the empirical research literature

Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.

Module 2 Chapter 3: What is Empirical Literature & Where can it be Found?

In Module 1, you read about the problem of pseudoscience. Here, we revisit the issue in addressing how to locate and assess scientific or empirical literature . In this chapter you will read about:

• distinguishing between what IS and IS NOT empirical literature
• how and where to locate empirical literature for understanding diverse populations, social work problems, and social phenomena.

Probably the most important take-home lesson from this chapter is that one source is not sufficient to being well-informed on a topic. It is important to locate multiple sources of information and to critically appraise the points of convergence and divergence in the information acquired from different sources. This is especially true in emerging and poorly understood topics, as well as in answering complex questions.

What Is Empirical Literature

Social workers often need to locate valid, reliable information concerning the dimensions of a population group or subgroup, a social work problem, or social phenomenon. They might also seek information about the way specific problems or resources are distributed among the populations encountered in professional practice. Or, social workers might be interested in finding out about the way that certain people experience an event or phenomenon. Empirical literature resources may provide answers to many of these types of social work questions. In addition, resources containing data regarding social indicators may also prove helpful. Social indicators are the “facts and figures” statistics that describe the social, economic, and psychological factors that have an impact on the well-being of a community or other population group.The United Nations (UN) and the World Health Organization (WHO) are examples of organizations that monitor social indicators at a global level: dimensions of population trends (size, composition, growth/loss), health status (physical, mental, behavioral, life expectancy, maternal and infant mortality, fertility/child-bearing, and diseases like HIV/AIDS), housing and quality of sanitation (water supply, waste disposal), education and literacy, and work/income/unemployment/economics, for example.

Three characteristics stand out in empirical literature compared to other types of information available on a topic of interest: systematic observation and methodology, objectivity, and transparency/replicability/reproducibility. Let’s look a little more closely at these three features.

Systematic Observation and Methodology. The hallmark of empiricism is “repeated or reinforced observation of the facts or phenomena” (Holosko, 2006, p. 6). In empirical literature, established research methodologies and procedures are systematically applied to answer the questions of interest.

Objectivity. Gathering “facts,” whatever they may be, drives the search for empirical evidence (Holosko, 2006). Authors of empirical literature are expected to report the facts as observed, whether or not these facts support the investigators’ original hypotheses. Research integrity demands that the information be provided in an objective manner, reducing sources of investigator bias to the greatest possible extent.

Transparency and Replicability/Reproducibility.   Empirical literature is reported in such a manner that other investigators understand precisely what was done and what was found in a particular research study—to the extent that they could replicate the study to determine whether the findings are reproduced when repeated. The outcomes of an original and replication study may differ, but a reader could easily interpret the methods and procedures leading to each study’s findings.

What is NOT Empirical Literature

By now, it is probably obvious to you that literature based on “evidence” that is not developed in a systematic, objective, transparent manner is not empirical literature. On one hand, non-empirical types of professional literature may have great significance to social workers. For example, social work scholars may produce articles that are clearly identified as describing a new intervention or program without evaluative evidence, critiquing a policy or practice, or offering a tentative, untested theory about a phenomenon. These resources are useful in educating ourselves about possible issues or concerns. But, even if they are informed by evidence, they are not empirical literature. Here is a list of several sources of information that do not meet the standard of being called empirical literature:

• your course instructor’s lectures
• political statements
• advertisements
• newspapers & magazines (journalism)
• television news reports & analyses (journalism)
• many websites, Facebook postings, Twitter tweets, and blog postings
• the introductory literature review in an empirical article

You may be surprised to see the last two included in this list. Like the other sources of information listed, these sources also might lead you to look for evidence. But, they are not themselves sources of evidence. They may summarize existing evidence, but in the process of summarizing (like your instructor’s lectures), information is transformed, modified, reduced, condensed, and otherwise manipulated in such a manner that you may not see the entire, objective story. These are called secondary sources, as opposed to the original, primary source of evidence. In relying solely on secondary sources, you sacrifice your own critical appraisal and thinking about the original work—you are “buying” someone else’s interpretation and opinion about the original work, rather than developing your own interpretation and opinion. What if they got it wrong? How would you know if you did not examine the primary source for yourself? Consider the following as an example of “getting it wrong” being perpetuated.

Example: Bullying and School Shootings . One result of the heavily publicized April 1999 school shooting incident at Columbine High School (Colorado), was a heavy emphasis placed on bullying as a causal factor in these incidents (Mears, Moon, & Thielo, 2017), “creating a powerful master narrative about school shootings” (Raitanen, Sandberg, & Oksanen, 2017, p. 3). Naturally, with an identified cause, a great deal of effort was devoted to anti-bullying campaigns and interventions for enhancing resilience among youth who experience bullying.  However important these strategies might be for promoting positive mental health, preventing poor mental health, and possibly preventing suicide among school-aged children and youth, it is a mistaken belief that this can prevent school shootings (Mears, Moon, & Thielo, 2017). Many times the accounts of the perpetrators having been bullied come from potentially inaccurate third-party accounts, rather than the perpetrators themselves; bullying was not involved in all instances of school shooting; a perpetrator’s perception of being bullied/persecuted are not necessarily accurate; many who experience severe bullying do not perpetrate these incidents; bullies are the least targeted shooting victims; perpetrators of the shooting incidents were often bullying others; and, bullying is only one of many important factors associated with perpetrating such an incident (Ioannou, Hammond, & Simpson, 2015; Mears, Moon, & Thielo, 2017; Newman &Fox, 2009; Raitanen, Sandberg, & Oksanen, 2017). While mass media reports deliver bullying as a means of explaining the inexplicable, the reality is not so simple: “The connection between bullying and school shootings is elusive” (Langman, 2014), and “the relationship between bullying and school shooting is, at best, tenuous” (Mears, Moon, & Thielo, 2017, p. 940). The point is, when a narrative becomes this publicly accepted, it is difficult to sort out truth and reality without going back to original sources of information and evidence.

What May or May Not Be Empirical Literature: Literature Reviews

Investigators typically engage in a review of existing literature as they develop their own research studies. The review informs them about where knowledge gaps exist, methods previously employed by other scholars, limitations of prior work, and previous scholars’ recommendations for directing future research. These reviews may appear as a published article, without new study data being reported (see Fields, Anderson, & Dabelko-Schoeny, 2014 for example). Or, the literature review may appear in the introduction to their own empirical study report. These literature reviews are not considered to be empirical evidence sources themselves, although they may be based on empirical evidence sources. One reason is that the authors of a literature review may or may not have engaged in a systematic search process, identifying a full, rich, multi-sided pool of evidence reports.

There is, however, a type of review that applies systematic methods and is, therefore, considered to be more strongly rooted in evidence: the systematic review .

Systematic review of literature. A systematic reviewis a type of literature report where established methods have been systematically applied, objectively, in locating and synthesizing a body of literature. The systematic review report is characterized by a great deal of transparency about the methods used and the decisions made in the review process, and are replicable. Thus, it meets the criteria for empirical literature: systematic observation and methodology, objectivity, and transparency/reproducibility. We will work a great deal more with systematic reviews in the second course, SWK 3402, since they are important tools for understanding interventions. They are somewhat less common, but not unheard of, in helping us understand diverse populations, social work problems, and social phenomena.

Locating Empirical Evidence

Social workers have available a wide array of tools and resources for locating empirical evidence in the literature. These can be organized into four general categories.

Journal Articles. A number of professional journals publish articles where investigators report on the results of their empirical studies. However, it is important to know how to distinguish between empirical and non-empirical manuscripts in these journals. A key indicator, though not the only one, involves a peer review process . Many professional journals require that manuscripts undergo a process of peer review before they are accepted for publication. This means that the authors’ work is shared with scholars who provide feedback to the journal editor as to the quality of the submitted manuscript. The editor then makes a decision based on the reviewers’ feedback:

• Accept as is
• Accept with minor revisions
• Request that a revision be resubmitted (no assurance of acceptance)

When a “revise and resubmit” decision is made, the piece will go back through the review process to determine if it is now acceptable for publication and that all of the reviewers’ concerns have been adequately addressed. Editors may also reject a manuscript because it is a poor fit for the journal, based on its mission and audience, rather than sending it for review consideration.

Indicators of journal relevance. Various journals are not equally relevant to every type of question being asked of the literature. Journals may overlap to a great extent in terms of the topics they might cover; in other words, a topic might appear in multiple different journals, depending on how the topic was being addressed. For example, articles that might help answer a question about the relationship between community poverty and violence exposure might appear in several different journals, some with a focus on poverty, others with a focus on violence, and still others on community development or public health. Journal titles are sometimes a good starting point but may not give a broad enough picture of what they cover in their contents.

In focusing a literature search, it also helps to review a journal’s mission and target audience. For example, at least four different journals focus specifically on poverty:

• Journal of Children & Poverty
• Journal of Poverty
• Journal of Poverty and Social Justice
• Poverty & Public Policy

Let’s look at an example using the Journal of Poverty and Social Justice . Information about this journal is located on the journal’s webpage: http://policy.bristoluniversitypress.co.uk/journals/journal-of-poverty-and-social-justice . In the section headed “About the Journal” you can see that it is an internationally focused research journal, and that it addresses social justice issues in addition to poverty alone. The research articles are peer-reviewed (there appear to be non-empirical discussions published, as well). These descriptions about a journal are almost always available, sometimes listed as “scope” or “mission.” These descriptions also indicate the sponsorship of the journal—sponsorship may be institutional (a particular university or agency, such as Smith College Studies in Social Work ), a professional organization, such as the Council on Social Work Education (CSWE) or the National Association of Social Work (NASW), or a publishing company (e.g., Taylor & Frances, Wiley, or Sage).

Indicators of journal caliber.  Despite engaging in a peer review process, not all journals are equally rigorous. Some journals have very high rejection rates, meaning that many submitted manuscripts are rejected; others have fairly high acceptance rates, meaning that relatively few manuscripts are rejected. This is not necessarily the best indicator of quality, however, since newer journals may not be sufficiently familiar to authors with high quality manuscripts and some journals are very specific in terms of what they publish. Another index that is sometimes used is the journal’s impact factor . Impact factor is a quantitative number indicative of how often articles published in the journal are cited in the reference list of other journal articles—the statistic is calculated as the number of times on average each article published in a particular year were cited divided by the number of articles published (the number that could be cited). For example, the impact factor for the Journal of Poverty and Social Justice in our list above was 0.70 in 2017, and for the Journal of Poverty was 0.30. These are relatively low figures compared to a journal like the New England Journal of Medicine with an impact factor of 59.56! This means that articles published in that journal were, on average, cited more than 59 times in the next year or two.

Impact factors are not necessarily the best indicator of caliber, however, since many strong journals are geared toward practitioners rather than scholars, so they are less likely to be cited by other scholars but may have a large impact on a large readership. This may be the case for a journal like the one titled Social Work, the official journal of the National Association of Social Workers. It is distributed free to all members: over 120,000 practitioners, educators, and students of social work world-wide. The journal has a recent impact factor of.790. The journals with social work relevant content have impact factors in the range of 1.0 to 3.0 according to Scimago Journal & Country Rank (SJR), particularly when they are interdisciplinary journals (for example, Child Development , Journal of Marriage and Family , Child Abuse and Neglect , Child Maltreatmen t, Social Service Review , and British Journal of Social Work ). Once upon a time, a reader could locate different indexes comparing the “quality” of social work-related journals. However, the concept of “quality” is difficult to systematically define. These indexes have mostly been replaced by impact ratings, which are not necessarily the best, most robust indicators on which to rely in assessing journal quality. For example, new journals addressing cutting edge topics have not been around long enough to have been evaluated using this particular tool, and it takes a few years for articles to begin to be cited in other, later publications.

Beware of pseudo-, illegitimate, misleading, deceptive, and suspicious journals . Another side effect of living in the Age of Information is that almost anyone can circulate almost anything and call it whatever they wish. This goes for “journal” publications, as well. With the advent of open-access publishing in recent years (electronic resources available without subscription), we have seen an explosion of what are called predatory or junk journals . These are publications calling themselves journals, often with titles very similar to legitimate publications and often with fake editorial boards. These “publications” lack the integrity of legitimate journals. This caution is reminiscent of the discussions earlier in the course about pseudoscience and “snake oil” sales. The predatory nature of many apparent information dissemination outlets has to do with how scientists and scholars may be fooled into submitting their work, often paying to have their work peer-reviewed and published. There exists a “thriving black-market economy of publishing scams,” and at least two “journal blacklists” exist to help identify and avoid these scam journals (Anderson, 2017).

This issue is important to information consumers, because it creates a challenge in terms of identifying legitimate sources and publications. The challenge is particularly important to address when information from on-line, open-access journals is being considered. Open-access is not necessarily a poor choice—legitimate scientists may pay sizeable fees to legitimate publishers to make their work freely available and accessible as open-access resources. On-line access is also not necessarily a poor choice—legitimate publishers often make articles available on-line to provide timely access to the content, especially when publishing the article in hard copy will be delayed by months or even a year or more. On the other hand, stating that a journal engages in a peer-review process is no guarantee of quality—this claim may or may not be truthful. Pseudo- and junk journals may engage in some quality control practices, but may lack attention to important quality control processes, such as managing conflict of interest, reviewing content for objectivity or quality of the research conducted, or otherwise failing to adhere to industry standards (Laine & Winker, 2017).

One resource designed to assist with the process of deciphering legitimacy is the Directory of Open Access Journals (DOAJ). The DOAJ is not a comprehensive listing of all possible legitimate open-access journals, and does not guarantee quality, but it does help identify legitimate sources of information that are openly accessible and meet basic legitimacy criteria. It also is about open-access journals, not the many journals published in hard copy.

An additional caution: Search for article corrections. Despite all of the careful manuscript review and editing, sometimes an error appears in a published article. Most journals have a practice of publishing corrections in future issues. When you locate an article, it is helpful to also search for updates. Here is an example where data presented in an article’s original tables were erroneous, and a correction appeared in a later issue.

• Marchant, A., Hawton, K., Stewart A., Montgomery, P., Singaravelu, V., Lloyd, K., Purdy, N., Daine, K., & John, A. (2017). A systematic review of the relationship between internet use, self-harm and suicidal behaviour in young people: The good, the bad and the unknown. PLoS One, 12(8): e0181722. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5558917/
• Marchant, A., Hawton, K., Stewart A., Montgomery, P., Singaravelu, V., Lloyd, K., Purdy, N., Daine, K., & John, A. (2018).Correction—A systematic review of the relationship between internet use, self-harm and suicidal behaviour in young people: The good, the bad and the unknown. PLoS One, 13(3): e0193937.  http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0193937

Search Tools. In this age of information, it is all too easy to find items—the problem lies in sifting, sorting, and managing the vast numbers of items that can be found. For example, a simple Google® search for the topic “community poverty and violence” resulted in about 15,600,000 results! As a means of simplifying the process of searching for journal articles on a specific topic, a variety of helpful tools have emerged. One type of search tool has previously applied a filtering process for you: abstracting and indexing databases . These resources provide the user with the results of a search to which records have already passed through one or more filters. For example, PsycINFO is managed by the American Psychological Association and is devoted to peer-reviewed literature in behavioral science. It contains almost 4.5 million records and is growing every month. However, it may not be available to users who are not affiliated with a university library. Conducting a basic search for our topic of “community poverty and violence” in PsychINFO returned 1,119 articles. Still a large number, but far more manageable. Additional filters can be applied, such as limiting the range in publication dates, selecting only peer reviewed items, limiting the language of the published piece (English only, for example), and specified types of documents (either chapters, dissertations, or journal articles only, for example). Adding the filters for English, peer-reviewed journal articles published between 2010 and 2017 resulted in 346 documents being identified.

Just as was the case with journals, not all abstracting and indexing databases are equivalent. There may be overlap between them, but none is guaranteed to identify all relevant pieces of literature. Here are some examples to consider, depending on the nature of the questions asked of the literature:

• Academic Search Complete—multidisciplinary index of 9,300 peer-reviewed journals
• AgeLine—multidisciplinary index of aging-related content for over 600 journals
• Campbell Collaboration—systematic reviews in education, crime and justice, social welfare, international development
• Google Scholar—broad search tool for scholarly literature across many disciplines
• MEDLINE/ PubMed—National Library of medicine, access to over 15 million citations
• Oxford Bibliographies—annotated bibliographies, each is discipline specific (e.g., psychology, childhood studies, criminology, social work, sociology)
• PsycINFO/PsycLIT—international literature on material relevant to psychology and related disciplines
• SocINDEX—publications in sociology
• Social Sciences Abstracts—multiple disciplines
• Social Work Abstracts—many areas of social work are covered
• Web of Science—a “meta” search tool that searches other search tools, multiple disciplines

Placing our search for information about “community violence and poverty” into the Social Work Abstracts tool with no additional filters resulted in a manageable 54-item list. Finally, abstracting and indexing databases are another way to determine journal legitimacy: if a journal is indexed in a one of these systems, it is likely a legitimate journal. However, the converse is not necessarily true: if a journal is not indexed does not mean it is an illegitimate or pseudo-journal.

Government Sources. A great deal of information is gathered, analyzed, and disseminated by various governmental branches at the international, national, state, regional, county, and city level. Searching websites that end in.gov is one way to identify this type of information, often presented in articles, news briefs, and statistical reports. These government sources gather information in two ways: they fund external investigations through grants and contracts and they conduct research internally, through their own investigators. Here are some examples to consider, depending on the nature of the topic for which information is sought:

• Agency for Healthcare Research and Quality (AHRQ) at https://www.ahrq.gov/
• Bureau of Justice Statistics (BJS) at https://www.bjs.gov/
• Census Bureau at https://www.census.gov
• Morbidity and Mortality Weekly Report of the CDC (MMWR-CDC) at https://www.cdc.gov/mmwr/index.html
• Child Welfare Information Gateway at https://www.childwelfare.gov
• Children’s Bureau/Administration for Children & Families at https://www.acf.hhs.gov
• Forum on Child and Family Statistics at https://www.childstats.gov
• National Institutes of Health (NIH) at https://www.nih.gov , including (not limited to):
• National Institute on Aging (NIA at https://www.nia.nih.gov
• National Institute on Alcohol Abuse and Alcoholism (NIAAA) at https://www.niaaa.nih.gov
• National Institute of Child Health and Human Development (NICHD) at https://www.nichd.nih.gov
• National Institute on Drug Abuse (NIDA) at https://www.nida.nih.gov
• National Institute of Environmental Health Sciences at https://www.niehs.nih.gov
• National Institute of Mental Health (NIMH) at https://www.nimh.nih.gov
• National Institute on Minority Health and Health Disparities at https://www.nimhd.nih.gov
• National Institute of Justice (NIJ) at https://www.nij.gov
• Substance Abuse and Mental Health Services Administration (SAMHSA) at https://www.samhsa.gov/
• United States Agency for International Development at https://usaid.gov

Each state and many counties or cities have similar data sources and analysis reports available, such as Ohio Department of Health at https://www.odh.ohio.gov/healthstats/dataandstats.aspx and Franklin County at https://statisticalatlas.com/county/Ohio/Franklin-County/Overview . Data are available from international/global resources (e.g., United Nations and World Health Organization), as well.

Other Sources. The Health and Medicine Division (HMD) of the National Academies—previously the Institute of Medicine (IOM)—is a nonprofit institution that aims to provide government and private sector policy and other decision makers with objective analysis and advice for making informed health decisions. For example, in 2018 they produced reports on topics in substance use and mental health concerning the intersection of opioid use disorder and infectious disease,  the legal implications of emerging neurotechnologies, and a global agenda concerning the identification and prevention of violence (see http://www.nationalacademies.org/hmd/Global/Topics/Substance-Abuse-Mental-Health.aspx ). The exciting aspect of this resource is that it addresses many topics that are current concerns because they are hoping to help inform emerging policy. The caution to consider with this resource is the evidence is often still emerging, as well.

Numerous “think tank” organizations exist, each with a specific mission. For example, the Rand Corporation is a nonprofit organization offering research and analysis to address global issues since 1948. The institution’s mission is to help improve policy and decision making “to help individuals, families, and communities throughout the world be safer and more secure, healthier and more prosperous,” addressing issues of energy, education, health care, justice, the environment, international affairs, and national security (https://www.rand.org/about/history.html). And, for example, the Robert Woods Johnson Foundation is a philanthropic organization supporting research and research dissemination concerning health issues facing the United States. The foundation works to build a culture of health across systems of care (not only medical care) and communities (https://www.rwjf.org).

While many of these have a great deal of helpful evidence to share, they also may have a strong political bias. Objectivity is often lacking in what information these organizations provide: they provide evidence to support certain points of view. That is their purpose—to provide ideas on specific problems, many of which have a political component. Think tanks “are constantly researching solutions to a variety of the world’s problems, and arguing, advocating, and lobbying for policy changes at local, state, and federal levels” (quoted from https://thebestschools.org/features/most-influential-think-tanks/ ). Helpful information about what this one source identified as the 50 most influential U.S. think tanks includes identifying each think tank’s political orientation. For example, The Heritage Foundation is identified as conservative, whereas Human Rights Watch is identified as liberal.

While not the same as think tanks, many mission-driven organizations also sponsor or report on research, as well. For example, the National Association for Children of Alcoholics (NACOA) in the United States is a registered nonprofit organization. Its mission, along with other partnering organizations, private-sector groups, and federal agencies, is to promote policy and program development in research, prevention and treatment to provide information to, for, and about children of alcoholics (of all ages). Based on this mission, the organization supports knowledge development and information gathering on the topic and disseminates information that serves the needs of this population. While this is a worthwhile mission, there is no guarantee that the information meets the criteria for evidence with which we have been working. Evidence reported by think tank and mission-driven sources must be utilized with a great deal of caution and critical analysis!

In many instances an empirical report has not appeared in the published literature, but in the form of a technical or final report to the agency or program providing the funding for the research that was conducted. One such example is presented by a team of investigators funded by the National Institute of Justice to evaluate a program for training professionals to collect strong forensic evidence in instances of sexual assault (Patterson, Resko, Pierce-Weeks, & Campbell, 2014): https://www.ncjrs.gov/pdffiles1/nij/grants/247081.pdf . Investigators may serve in the capacity of consultant to agencies, programs, or institutions, and provide empirical evidence to inform activities and planning. One such example is presented by Maguire-Jack (2014) as a report to a state’s child maltreatment prevention board: https://preventionboard.wi.gov/Documents/InvestmentInPreventionPrograming_Final.pdf .

When Direct Answers to Questions Cannot Be Found. Sometimes social workers are interested in finding answers to complex questions or questions related to an emerging, not-yet-understood topic. This does not mean giving up on empirical literature. Instead, it requires a bit of creativity in approaching the literature. A Venn diagram might help explain this process. Consider a scenario where a social worker wishes to locate literature to answer a question concerning issues of intersectionality. Intersectionality is a social justice term applied to situations where multiple categorizations or classifications come together to create overlapping, interconnected, or multiplied disadvantage. For example, women with a substance use disorder and who have been incarcerated face a triple threat in terms of successful treatment for a substance use disorder: intersectionality exists between being a woman, having a substance use disorder, and having been in jail or prison. After searching the literature, little or no empirical evidence might have been located on this specific triple-threat topic. Instead, the social worker will need to seek literature on each of the threats individually, and possibly will find literature on pairs of topics (see Figure 3-1). There exists some literature about women’s outcomes for treatment of a substance use disorder (a), some literature about women during and following incarceration (b), and some literature about substance use disorders and incarceration (c). Despite not having a direct line on the center of the intersecting spheres of literature (d), the social worker can develop at least a partial picture based on the overlapping literatures.

Figure 3-1. Venn diagram of intersecting literature sets.

Take a moment to complete the following activity. For each statement about empirical literature, decide if it is true or false.

Social Work 3401 Coursebook Copyright © by Dr. Audrey Begun is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License , except where otherwise noted.

Share This Book

• Ask a Librarian

Research: Overview & Approaches

• Getting Started with Undergraduate Research
• Planning & Getting Started
• Building Your Knowledge Base
• Locating Sources
• Reading Scholarly Articles
• Creating a Literature Review
• Productivity & Organizing Research
• Scholarly and Professional Relationships

Introduction to Empirical Research

Databases for finding empirical research, guided search, google scholar, examples of empirical research, sources and further reading.

• Interpretive Research
• Action-Based Research
• Creative & Experimental Approaches

Your Librarian

• Introductory Video This video covers what empirical research is, what kinds of questions and methods empirical researchers use, and some tips for finding empirical research articles in your discipline.

• Guided Search: Finding Empirical Research Articles This is a hands-on tutorial that will allow you to use your own search terms to find resources.

• Study on radiation transfer in human skin for cosmetics
• Long-Term Mobile Phone Use and the Risk of Vestibular Schwannoma: A Danish Nationwide Cohort Study
• Emissions Impacts and Benefits of Plug-In Hybrid Electric Vehicles and Vehicle-to-Grid Services
• Review of design considerations and technological challenges for successful development and deployment of plug-in hybrid electric vehicles
• Endocrine disrupters and human health: could oestrogenic chemicals in body care cosmetics adversely affect breast cancer incidence in women?

• << Previous: Scholarly and Professional Relationships
• Next: Interpretive Research >>
• Last Updated: May 23, 2024 11:51 AM
• URL: https://guides.lib.purdue.edu/research_approaches

Penn State University Libraries

Empirical research in the social sciences and education.

• What is Empirical Research and How to Read It
• Finding Empirical Research in Library Databases
• Designing Empirical Research
• Ethics, Cultural Responsiveness, and Anti-Racism in Research
• Citing, Writing, and Presenting Your Work

Contact the Librarian at your campus for more help!

Introduction: What is Empirical Research?

Empirical research is based on observed and measured phenomena and derives knowledge from actual experience rather than from theory or belief. 

How do you know if a study is empirical? Read the subheadings within the article, book, or report and look for a description of the research "methodology."  Ask yourself: Could I recreate this study and test these results?

Key characteristics to look for:

• Specific research questions to be answered
• Definition of the population, behavior, or   phenomena being studied
• Description of the process used to study this population or phenomena, including selection criteria, controls, and testing instruments (such as surveys)

Another hint: some scholarly journals use a specific layout, called the "IMRaD" format, to communicate empirical research findings. Such articles typically have 4 components:

• Introduction : sometimes called "literature review" -- what is currently known about the topic -- usually includes a theoretical framework and/or discussion of previous studies
• Methodology: sometimes called "research design" -- how to recreate the study -- usually describes the population, research process, and analytical tools used in the present study
• Results : sometimes called "findings" -- what was learned through the study -- usually appears as statistical data or as substantial quotations from research participants
• Discussion : sometimes called "conclusion" or "implications" -- why the study is important -- usually describes how the research results influence professional practices or future studies

Reading and Evaluating Scholarly Materials

Reading research can be a challenge. However, the tutorials and videos below can help. They explain what scholarly articles look like, how to read them, and how to evaluate them:

• CRAAP Checklist A frequently-used checklist that helps you examine the currency, relevance, authority, accuracy, and purpose of an information source.
• IF I APPLY A newer model of evaluating sources which encourages you to think about your own biases as a reader, as well as concerns about the item you are reading.
• Credo Video: How to Read Scholarly Materials (4 min.)
• Credo Tutorial: How to Read Scholarly Materials
• Credo Tutorial: Evaluating Information
• Credo Video: Evaluating Statistics (4 min.)
• Next: Finding Empirical Research in Library Databases >>
• Last Updated: Feb 18, 2024 8:33 PM
• URL: https://guides.libraries.psu.edu/emp

• Skip to main content
• Skip to primary sidebar
• Skip to footer
• QuestionPro

• Solutions Industries Gaming Automotive Sports and events Education Government Travel & Hospitality Financial Services Healthcare Cannabis Technology Use Case NPS+ Communities Audience Contactless surveys Mobile LivePolls Member Experience GDPR Positive People Science 360 Feedback Surveys
• Resources Blog eBooks Survey Templates Case Studies Training Help center

Home Market Research

Empirical Research: Definition, Methods, Types and Examples

Content Index

Empirical research: Definition

Empirical research: origin, quantitative research methods, qualitative research methods, steps for conducting empirical research, empirical research methodology cycle, advantages of empirical research, disadvantages of empirical research, why is there a need for empirical research.

Empirical research is defined as any research where conclusions of the study is strictly drawn from concretely empirical evidence, and therefore “verifiable” evidence.

This empirical evidence can be gathered using quantitative market research and  qualitative market research  methods.

For example: A research is being conducted to find out if listening to happy music in the workplace while working may promote creativity? An experiment is conducted by using a music website survey on a set of audience who are exposed to happy music and another set who are not listening to music at all, and the subjects are then observed. The results derived from such a research will give empirical evidence if it does promote creativity or not.

LEARN ABOUT: Behavioral Research

You must have heard the quote” I will not believe it unless I see it”. This came from the ancient empiricists, a fundamental understanding that powered the emergence of medieval science during the renaissance period and laid the foundation of modern science, as we know it today. The word itself has its roots in greek. It is derived from the greek word empeirikos which means “experienced”.

In today’s world, the word empirical refers to collection of data using evidence that is collected through observation or experience or by using calibrated scientific instruments. All of the above origins have one thing in common which is dependence of observation and experiments to collect data and test them to come up with conclusions.

LEARN ABOUT: Causal Research

Types and methodologies of empirical research

Empirical research can be conducted and analysed using qualitative or quantitative methods.

• Quantitative research : Quantitative research methods are used to gather information through numerical data. It is used to quantify opinions, behaviors or other defined variables . These are predetermined and are in a more structured format. Some of the commonly used methods are survey, longitudinal studies, polls, etc
• Qualitative research:   Qualitative research methods are used to gather non numerical data.  It is used to find meanings, opinions, or the underlying reasons from its subjects. These methods are unstructured or semi structured. The sample size for such a research is usually small and it is a conversational type of method to provide more insight or in-depth information about the problem Some of the most popular forms of methods are focus groups, experiments, interviews, etc.

Data collected from these will need to be analysed. Empirical evidence can also be analysed either quantitatively and qualitatively. Using this, the researcher can answer empirical questions which have to be clearly defined and answerable with the findings he has got. The type of research design used will vary depending on the field in which it is going to be used. Many of them might choose to do a collective research involving quantitative and qualitative method to better answer questions which cannot be studied in a laboratory setting.

LEARN ABOUT: Qualitative Research Questions and Questionnaires

Quantitative research methods aid in analyzing the empirical evidence gathered. By using these a researcher can find out if his hypothesis is supported or not.

• Survey research: Survey research generally involves a large audience to collect a large amount of data. This is a quantitative method having a predetermined set of closed questions which are pretty easy to answer. Because of the simplicity of such a method, high responses are achieved. It is one of the most commonly used methods for all kinds of research in today’s world.

Previously, surveys were taken face to face only with maybe a recorder. However, with advancement in technology and for ease, new mediums such as emails , or social media have emerged.

For example: Depletion of energy resources is a growing concern and hence there is a need for awareness about renewable energy. According to recent studies, fossil fuels still account for around 80% of energy consumption in the United States. Even though there is a rise in the use of green energy every year, there are certain parameters because of which the general population is still not opting for green energy. In order to understand why, a survey can be conducted to gather opinions of the general population about green energy and the factors that influence their choice of switching to renewable energy. Such a survey can help institutions or governing bodies to promote appropriate awareness and incentive schemes to push the use of greener energy.

Learn more: Renewable Energy Survey Template Descriptive Research vs Correlational Research

• Experimental research: In experimental research , an experiment is set up and a hypothesis is tested by creating a situation in which one of the variable is manipulated. This is also used to check cause and effect. It is tested to see what happens to the independent variable if the other one is removed or altered. The process for such a method is usually proposing a hypothesis, experimenting on it, analyzing the findings and reporting the findings to understand if it supports the theory or not.

For example: A particular product company is trying to find what is the reason for them to not be able to capture the market. So the organisation makes changes in each one of the processes like manufacturing, marketing, sales and operations. Through the experiment they understand that sales training directly impacts the market coverage for their product. If the person is trained well, then the product will have better coverage.

• Correlational research: Correlational research is used to find relation between two set of variables . Regression analysis is generally used to predict outcomes of such a method. It can be positive, negative or neutral correlation.

LEARN ABOUT: Level of Analysis

For example: Higher educated individuals will get higher paying jobs. This means higher education enables the individual to high paying job and less education will lead to lower paying jobs.

• Longitudinal study: Longitudinal study is used to understand the traits or behavior of a subject under observation after repeatedly testing the subject over a period of time. Data collected from such a method can be qualitative or quantitative in nature.

For example: A research to find out benefits of exercise. The target is asked to exercise everyday for a particular period of time and the results show higher endurance, stamina, and muscle growth. This supports the fact that exercise benefits an individual body.

• Cross sectional: Cross sectional study is an observational type of method, in which a set of audience is observed at a given point in time. In this type, the set of people are chosen in a fashion which depicts similarity in all the variables except the one which is being researched. This type does not enable the researcher to establish a cause and effect relationship as it is not observed for a continuous time period. It is majorly used by healthcare sector or the retail industry.

For example: A medical study to find the prevalence of under-nutrition disorders in kids of a given population. This will involve looking at a wide range of parameters like age, ethnicity, location, incomes  and social backgrounds. If a significant number of kids coming from poor families show under-nutrition disorders, the researcher can further investigate into it. Usually a cross sectional study is followed by a longitudinal study to find out the exact reason.

• Causal-Comparative research : This method is based on comparison. It is mainly used to find out cause-effect relationship between two variables or even multiple variables.

For example: A researcher measured the productivity of employees in a company which gave breaks to the employees during work and compared that to the employees of the company which did not give breaks at all.

LEARN ABOUT: Action Research

Some research questions need to be analysed qualitatively, as quantitative methods are not applicable there. In many cases, in-depth information is needed or a researcher may need to observe a target audience behavior, hence the results needed are in a descriptive analysis form. Qualitative research results will be descriptive rather than predictive. It enables the researcher to build or support theories for future potential quantitative research. In such a situation qualitative research methods are used to derive a conclusion to support the theory or hypothesis being studied.

LEARN ABOUT: Qualitative Interview

• Case study: Case study method is used to find more information through carefully analyzing existing cases. It is very often used for business research or to gather empirical evidence for investigation purpose. It is a method to investigate a problem within its real life context through existing cases. The researcher has to carefully analyse making sure the parameter and variables in the existing case are the same as to the case that is being investigated. Using the findings from the case study, conclusions can be drawn regarding the topic that is being studied.

For example: A report mentioning the solution provided by a company to its client. The challenges they faced during initiation and deployment, the findings of the case and solutions they offered for the problems. Such case studies are used by most companies as it forms an empirical evidence for the company to promote in order to get more business.

• Observational method:   Observational method is a process to observe and gather data from its target. Since it is a qualitative method it is time consuming and very personal. It can be said that observational research method is a part of ethnographic research which is also used to gather empirical evidence. This is usually a qualitative form of research, however in some cases it can be quantitative as well depending on what is being studied.

For example: setting up a research to observe a particular animal in the rain-forests of amazon. Such a research usually take a lot of time as observation has to be done for a set amount of time to study patterns or behavior of the subject. Another example used widely nowadays is to observe people shopping in a mall to figure out buying behavior of consumers.

• One-on-one interview: Such a method is purely qualitative and one of the most widely used. The reason being it enables a researcher get precise meaningful data if the right questions are asked. It is a conversational method where in-depth data can be gathered depending on where the conversation leads.

For example: A one-on-one interview with the finance minister to gather data on financial policies of the country and its implications on the public.

• Focus groups: Focus groups are used when a researcher wants to find answers to why, what and how questions. A small group is generally chosen for such a method and it is not necessary to interact with the group in person. A moderator is generally needed in case the group is being addressed in person. This is widely used by product companies to collect data about their brands and the product.

For example: A mobile phone manufacturer wanting to have a feedback on the dimensions of one of their models which is yet to be launched. Such studies help the company meet the demand of the customer and position their model appropriately in the market.

• Text analysis: Text analysis method is a little new compared to the other types. Such a method is used to analyse social life by going through images or words used by the individual. In today’s world, with social media playing a major part of everyone’s life, such a method enables the research to follow the pattern that relates to his study.

For example: A lot of companies ask for feedback from the customer in detail mentioning how satisfied are they with their customer support team. Such data enables the researcher to take appropriate decisions to make their support team better.

Sometimes a combination of the methods is also needed for some questions that cannot be answered using only one type of method especially when a researcher needs to gain a complete understanding of complex subject matter.

We recently published a blog that talks about examples of qualitative data in education ; why don’t you check it out for more ideas?

Since empirical research is based on observation and capturing experiences, it is important to plan the steps to conduct the experiment and how to analyse it. This will enable the researcher to resolve problems or obstacles which can occur during the experiment.

Step #1: Define the purpose of the research

This is the step where the researcher has to answer questions like what exactly do I want to find out? What is the problem statement? Are there any issues in terms of the availability of knowledge, data, time or resources. Will this research be more beneficial than what it will cost.

Before going ahead, a researcher has to clearly define his purpose for the research and set up a plan to carry out further tasks.

Step #2 : Supporting theories and relevant literature

The researcher needs to find out if there are theories which can be linked to his research problem . He has to figure out if any theory can help him support his findings. All kind of relevant literature will help the researcher to find if there are others who have researched this before, or what are the problems faced during this research. The researcher will also have to set up assumptions and also find out if there is any history regarding his research problem

Step #3: Creation of Hypothesis and measurement

Before beginning the actual research he needs to provide himself a working hypothesis or guess what will be the probable result. Researcher has to set up variables, decide the environment for the research and find out how can he relate between the variables.

Researcher will also need to define the units of measurements, tolerable degree for errors, and find out if the measurement chosen will be acceptable by others.

Step #4: Methodology, research design and data collection

In this step, the researcher has to define a strategy for conducting his research. He has to set up experiments to collect data which will enable him to propose the hypothesis. The researcher will decide whether he will need experimental or non experimental method for conducting the research. The type of research design will vary depending on the field in which the research is being conducted. Last but not the least, the researcher will have to find out parameters that will affect the validity of the research design. Data collection will need to be done by choosing appropriate samples depending on the research question. To carry out the research, he can use one of the many sampling techniques. Once data collection is complete, researcher will have empirical data which needs to be analysed.

LEARN ABOUT: Best Data Collection Tools

Step #5: Data Analysis and result

Data analysis can be done in two ways, qualitatively and quantitatively. Researcher will need to find out what qualitative method or quantitative method will be needed or will he need a combination of both. Depending on the unit of analysis of his data, he will know if his hypothesis is supported or rejected. Analyzing this data is the most important part to support his hypothesis.

Step #6: Conclusion

A report will need to be made with the findings of the research. The researcher can give the theories and literature that support his research. He can make suggestions or recommendations for further research on his topic.

A.D. de Groot, a famous dutch psychologist and a chess expert conducted some of the most notable experiments using chess in the 1940’s. During his study, he came up with a cycle which is consistent and now widely used to conduct empirical research. It consists of 5 phases with each phase being as important as the next one. The empirical cycle captures the process of coming up with hypothesis about how certain subjects work or behave and then testing these hypothesis against empirical data in a systematic and rigorous approach. It can be said that it characterizes the deductive approach to science. Following is the empirical cycle.

• Observation: At this phase an idea is sparked for proposing a hypothesis. During this phase empirical data is gathered using observation. For example: a particular species of flower bloom in a different color only during a specific season.
• Induction: Inductive reasoning is then carried out to form a general conclusion from the data gathered through observation. For example: As stated above it is observed that the species of flower blooms in a different color during a specific season. A researcher may ask a question “does the temperature in the season cause the color change in the flower?” He can assume that is the case, however it is a mere conjecture and hence an experiment needs to be set up to support this hypothesis. So he tags a few set of flowers kept at a different temperature and observes if they still change the color?
• Deduction: This phase helps the researcher to deduce a conclusion out of his experiment. This has to be based on logic and rationality to come up with specific unbiased results.For example: In the experiment, if the tagged flowers in a different temperature environment do not change the color then it can be concluded that temperature plays a role in changing the color of the bloom.
• Testing: This phase involves the researcher to return to empirical methods to put his hypothesis to the test. The researcher now needs to make sense of his data and hence needs to use statistical analysis plans to determine the temperature and bloom color relationship. If the researcher finds out that most flowers bloom a different color when exposed to the certain temperature and the others do not when the temperature is different, he has found support to his hypothesis. Please note this not proof but just a support to his hypothesis.
• Evaluation: This phase is generally forgotten by most but is an important one to keep gaining knowledge. During this phase the researcher puts forth the data he has collected, the support argument and his conclusion. The researcher also states the limitations for the experiment and his hypothesis and suggests tips for others to pick it up and continue a more in-depth research for others in the future. LEARN MORE: Population vs Sample

LEARN MORE: Population vs Sample

There is a reason why empirical research is one of the most widely used method. There are a few advantages associated with it. Following are a few of them.

• It is used to authenticate traditional research through various experiments and observations.
• This research methodology makes the research being conducted more competent and authentic.
• It enables a researcher understand the dynamic changes that can happen and change his strategy accordingly.
• The level of control in such a research is high so the researcher can control multiple variables.
• It plays a vital role in increasing internal validity .

Even though empirical research makes the research more competent and authentic, it does have a few disadvantages. Following are a few of them.

• Such a research needs patience as it can be very time consuming. The researcher has to collect data from multiple sources and the parameters involved are quite a few, which will lead to a time consuming research.
• Most of the time, a researcher will need to conduct research at different locations or in different environments, this can lead to an expensive affair.
• There are a few rules in which experiments can be performed and hence permissions are needed. Many a times, it is very difficult to get certain permissions to carry out different methods of this research.
• Collection of data can be a problem sometimes, as it has to be collected from a variety of sources through different methods.

LEARN ABOUT:  Social Communication Questionnaire

Empirical research is important in today’s world because most people believe in something only that they can see, hear or experience. It is used to validate multiple hypothesis and increase human knowledge and continue doing it to keep advancing in various fields.

For example: Pharmaceutical companies use empirical research to try out a specific drug on controlled groups or random groups to study the effect and cause. This way, they prove certain theories they had proposed for the specific drug. Such research is very important as sometimes it can lead to finding a cure for a disease that has existed for many years. It is useful in science and many other fields like history, social sciences, business, etc.

LEARN ABOUT: 12 Best Tools for Researchers

With the advancement in today’s world, empirical research has become critical and a norm in many fields to support their hypothesis and gain more knowledge. The methods mentioned above are very useful for carrying out such research. However, a number of new methods will keep coming up as the nature of new investigative questions keeps getting unique or changing.

Create a single source of real data with a built-for-insights platform. Store past data, add nuggets of insights, and import research data from various sources into a CRM for insights. Build on ever-growing research with a real-time dashboard in a unified research management platform to turn insights into knowledge.

LEARN MORE         FREE TRIAL

MORE LIKE THIS

What Are My Employees Really Thinking? The Power of Open-ended Survey Analysis

May 24, 2024

I Am Disconnected – Tuesday CX Thoughts

May 21, 2024

20 Best Customer Success Tools of 2024

May 20, 2024

AI-Based Services Buying Guide for Market Research (based on ESOMAR’s 20 Questions) 

Other categories.

• Academic Research
• Artificial Intelligence
• Assessments
• Brand Awareness
• Case Studies
• Communities
• Consumer Insights
• Customer effort score
• Customer Engagement
• Customer Experience
• Customer Loyalty
• Customer Research
• Customer Satisfaction
• Employee Benefits
• Employee Engagement
• Employee Retention
• Friday Five
• General Data Protection Regulation
• Insights Hub
• Life@QuestionPro
• Market Research
• Mobile diaries
• Mobile Surveys
• New Features
• Online Communities
• Question Types
• Questionnaire
• QuestionPro Products
• Release Notes
• Research Tools and Apps
• Revenue at Risk
• Survey Templates
• Training Tips
• Uncategorized
• Video Learning Series
• What’s Coming Up
• Workforce Intelligence

• quicklinks Academic admin council Academic calendar Academic stds cte Admission Advising African studies Alumni engagement American studies Anthropology/sociology Arabic Arboretum Archives Arcus center Art Assessment committee Athletics Athletic training Biology Biology&chem center Black faculty&staff assoc Bookstore BrandK Business office Campus event calendar Campus safety Catalog Career & prof dev Health science Ctr for civic engagement Ctr for international pgrms Chemistry Chinese Classics College communication Community & global health Community council Complex systems studies Computer science Copyright Counseling Council of student reps Crisis response Critical ethnic studies Critical theory Development Dining services Directories Disability services Donor relations East Asian studies Economics and business Educational policies cte Educational quality assmt Engineering Environmental stewardship Environmental studies English Experiential education cte Facilities management Facilities reservations Faculty development cte Faculty executive cte Faculty grants Faculty personnel cte Fellowships & grants Festival playhouse Film & media studies Financial aid First year experience Fitness & wellness ctr French Gardens & growing spaces German Global crossroads Health center Jewish studies History Hornet hive Hornet HQ Hornet sports Human resources Inclusive excellence Index (student newspaper) Information services Institutional research Institutional review board Intercultural student life International & area studies International programs Intramural sports Japanese LandSea Learning commons Learning support Lgbtqai+ student resources Library Mail and copy center Math Math/physics center Microsoft Stream Microsoft Teams Moodle Movies (ch 22 online) Music OneDrive Outdoor programs Parents' resources Payroll Phi Beta Kappa Philharmonia Philosophy Physics Physical education Political science Pre-law advising Provost Psychology Public pol & urban affairs Recycling Registrar Religion Religious & spiritual life Research Guides (libguides) Residential life Safety (security) Sexual safety Shared passages program SharePoint online Sophomore experience Spanish Strategic plan Student accounts Student development Student activities Student organizations Study abroad Support staff Sustainability Teaching and learning cte Teaching commons Theatre arts Title IX Webmail Women, gender & sexuality Writing center

PSYC 301: Intro to Research Methods

• Advanced Search Strategies
• Tracking the Research Process
• Annotations
• Article Cards
• Organizing Sources
• Writing an Outline
• Citing Sources

Finding Empirical Research

Empirical research is published in books and in scholarly, peer-reviewed journals. PsycInfo  offers straightforward ways to identify empirical research, unlike most other databases.

Finding Empirical Research in PsycInfo

• PsycInfo Choose "Advanced Search" Scroll down the page to "Methodology," and choose "Empirical Study" Type your keywords into the search boxes Choose other limits, such as publication date, if needed Click on the "Search" button

Slideshow showing how to find empirical research in APA PsycInfo

Video of finding empirical articles in psycinfo.

• Searching for Peer-Reviewed Empirical Articles (YouTube Video) Created by the APA

What is Empirical Research?

Empirical research  is based on observed and measured phenomena and derives knowledge from actual experience rather than from theory or belief. 

How do you know if a study is empirical? Read the subheadings within the article, book, or report and look for a description of the research "methodology." Ask yourself: Could I recreate this study and test these results?

Key characteristics to look for:

• Specific research questions  to be answered
• Definition of the  population, behavior, or   phenomena  being studied
• Description of the  process  used to study this population or phenomena, including selection criteria, controls, and testing instruments (such as surveys)

Another hint: some scholarly journals use a specific layout, called the "IMRaD" format, to communicate empirical research findings. Such articles typically have 4 components:

• Introduction : sometimes called "literature review" -- what is currently known about the topic -- usually includes a theoretical framework and/or discussion of previous studies
• Methodology:  sometimes called "research design" -- how to recreate the study -- usually describes the population, research process, and analytical tools
• Results : sometimes called "findings"  --  what was learned through the study -- usually appears as statistical data or as substantial quotations from research participants
• Discussion : sometimes called "conclusion" or "implications" -- why the study is important -- usually describes how the research results influence professional practices or future studies

Adapted from PennState University Libraries, Empirical Research in the Social Sciences and Education

Using PsycInfo

• Narrowing a Search (Canva Slideshow) Created by K Librarians
• Searching with the Thesaurus and Index Terms (YouTube Video) Created by the APA
• << Previous: Home
• Next: Advanced Search Strategies >>
• Last Updated: Apr 29, 2024 12:20 PM
• URL: https://libguides.kzoo.edu/psyc301

Canvas | University | Ask a Librarian

• Library Homepage
• Arrendale Library

Empirical Research: Quantitative & Qualitative

• Empirical Research

Introduction: What is Empirical Research?

Quantitative methods, qualitative methods.

• Quantitative vs. Qualitative
• Reference Works for Social Sciences Research
• Contact Us!

 Call us at 706-776-0111

  Chat with a Librarian

  Send Us Email

  Library Hours

Empirical research  is based on phenomena that can be observed and measured. Empirical research derives knowledge from actual experience rather than from theory or belief. 

Key characteristics of empirical research include:

• Specific research questions to be answered;
• Definitions of the population, behavior, or phenomena being studied;
• Description of the methodology or research design used to study this population or phenomena, including selection criteria, controls, and testing instruments (such as surveys);
• Two basic research processes or methods in empirical research: quantitative methods and qualitative methods (see the rest of the guide for more about these methods).

(based on the original from the Connelly LIbrary of LaSalle University)

Empirical Research: Qualitative vs. Quantitative

Learn about common types of journal articles that use APA Style, including empirical studies; meta-analyses; literature reviews; and replication, theoretical, and methodological articles.

Academic Writer

© 2024 American Psychological Association.

• More about Academic Writer ...

Quantitative Research

A quantitative research project is characterized by having a population about which the researcher wants to draw conclusions, but it is not possible to collect data on the entire population.

• For an observational study, it is necessary to select a proper, statistical random sample and to use methods of statistical inference to draw conclusions about the population. 
• For an experimental study, it is necessary to have a random assignment of subjects to experimental and control groups in order to use methods of statistical inference.

Statistical methods are used in all three stages of a quantitative research project.

For observational studies, the data are collected using statistical sampling theory. Then, the sample data are analyzed using descriptive statistical analysis. Finally, generalizations are made from the sample data to the entire population using statistical inference.

For experimental studies, the subjects are allocated to experimental and control group using randomizing methods. Then, the experimental data are analyzed using descriptive statistical analysis. Finally, just as for observational data, generalizations are made to a larger population.

Iversen, G. (2004). Quantitative research . In M. Lewis-Beck, A. Bryman, & T. Liao (Eds.), Encyclopedia of social science research methods . (pp. 897-898). Thousand Oaks, CA: SAGE Publications, Inc.

Qualitative Research

What makes a work deserving of the label qualitative research is the demonstrable effort to produce richly and relevantly detailed descriptions and particularized interpretations of people and the social, linguistic, material, and other practices and events that shape and are shaped by them.

Qualitative research typically includes, but is not limited to, discerning the perspectives of these people, or what is often referred to as the actor’s point of view. Although both philosophically and methodologically a highly diverse entity, qualitative research is marked by certain defining imperatives that include its case (as opposed to its variable) orientation, sensitivity to cultural and historical context, and reflexivity. 

In its many guises, qualitative research is a form of empirical inquiry that typically entails some form of purposive sampling for information-rich cases; in-depth interviews and open-ended interviews, lengthy participant/field observations, and/or document or artifact study; and techniques for analysis and interpretation of data that move beyond the data generated and their surface appearances. 

Sandelowski, M. (2004).  Qualitative research . In M. Lewis-Beck, A. Bryman, & T. Liao (Eds.),  Encyclopedia of social science research methods . (pp. 893-894). Thousand Oaks, CA: SAGE Publications, Inc.

• Next: Quantitative vs. Qualitative >>
• Last Updated: Mar 22, 2024 10:47 AM
• URL: https://library.piedmont.edu/empirical-research
• Ebooks & Online Video
• New Materials
• Renew Checkouts
• Faculty Resources
• Library Friends
• Library Services
• Our Mission
• Library History
• Ask a Librarian!
• Making Citations
• Working Online

Arrendale Library Piedmont University 706-776-0111

Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.

2.3 Reviewing the Research Literature

Learning objectives.

• Define the research literature in psychology and give examples of sources that are part of the research literature and sources that are not.
• Describe and use several methods for finding previous research on a particular research idea or question.

Reviewing the research literature means finding, reading, and summarizing the published research relevant to your question. An empirical research report written in American Psychological Association (APA) style always includes a written literature review, but it is important to review the literature early in the research process for several reasons.

• It can help you turn a research idea into an interesting research question.
• It can tell you if a research question has already been answered.
• It can help you evaluate the interestingness of a research question.
• It can give you ideas for how to conduct your own study.
• It can tell you how your study fits into the research literature.

What Is the Research Literature?

The research literature in any field is all the published research in that field. The research literature in psychology is enormous—including millions of scholarly articles and books dating to the beginning of the field—and it continues to grow. Although its boundaries are somewhat fuzzy, the research literature definitely does not include self-help and other pop psychology books, dictionary and encyclopedia entries, websites, and similar sources that are intended mainly for the general public. These are considered unreliable because they are not reviewed by other researchers and are often based on little more than common sense or personal experience. Wikipedia contains much valuable information, but the fact that its authors are anonymous and its content continually changes makes it unsuitable as a basis of sound scientific research. For our purposes, it helps to define the research literature as consisting almost entirely of two types of sources: articles in professional journals, and scholarly books in psychology and related fields.

Professional Journals

Professional journals are periodicals that publish original research articles. There are thousands of professional journals that publish research in psychology and related fields. They are usually published monthly or quarterly in individual issues, each of which contains several articles. The issues are organized into volumes, which usually consist of all the issues for a calendar year. Some journals are published in hard copy only, others in both hard copy and electronic form, and still others in electronic form only.

Most articles in professional journals are one of two basic types: empirical research reports and review articles. Empirical research reports describe one or more new empirical studies conducted by the authors. They introduce a research question, explain why it is interesting, review previous research, describe their method and results, and draw their conclusions. Review articles summarize previously published research on a topic and usually present new ways to organize or explain the results. When a review article is devoted primarily to presenting a new theory, it is often referred to as a theoretical article .

Figure 2.6 Small Sample of the Thousands of Professional Journals That Publish Research in Psychology and Related Fields

Most professional journals in psychology undergo a process of peer review . Researchers who want to publish their work in the journal submit a manuscript to the editor—who is generally an established researcher too—who in turn sends it to two or three experts on the topic. Each reviewer reads the manuscript, writes a critical review, and sends the review back to the editor along with his or her recommendations. The editor then decides whether to accept the article for publication, ask the authors to make changes and resubmit it for further consideration, or reject it outright. In any case, the editor forwards the reviewers’ written comments to the researchers so that they can revise their manuscript accordingly. Peer review is important because it ensures that the work meets basic standards of the field before it can enter the research literature.

Scholarly Books

Scholarly books are books written by researchers and practitioners mainly for use by other researchers and practitioners. A monograph is written by a single author or a small group of authors and usually gives a coherent presentation of a topic much like an extended review article. Edited volumes have an editor or a small group of editors who recruit many authors to write separate chapters on different aspects of the same topic. Although edited volumes can also give a coherent presentation of the topic, it is not unusual for each chapter to take a different perspective or even for the authors of different chapters to openly disagree with each other. In general, scholarly books undergo a peer review process similar to that used by professional journals.

Literature Search Strategies

Using psycinfo and other databases.

The primary method used to search the research literature involves using one or more electronic databases. These include Academic Search Premier, JSTOR, and ProQuest for all academic disciplines, ERIC for education, and PubMed for medicine and related fields. The most important for our purposes, however, is PsycINFO , which is produced by the APA. PsycINFO is so comprehensive—covering thousands of professional journals and scholarly books going back more than 100 years—that for most purposes its content is synonymous with the research literature in psychology. Like most such databases, PsycINFO is usually available through your college or university library.

PsycINFO consists of individual records for each article, book chapter, or book in the database. Each record includes basic publication information, an abstract or summary of the work, and a list of other works cited by that work. A computer interface allows entering one or more search terms and returns any records that contain those search terms. (These interfaces are provided by different vendors and therefore can look somewhat different depending on the library you use.) Each record also contains lists of keywords that describe the content of the work and also a list of index terms. The index terms are especially helpful because they are standardized. Research on differences between women and men, for example, is always indexed under “Human Sex Differences.” Research on touching is always indexed under the term “Physical Contact.” If you do not know the appropriate index terms, PsycINFO includes a thesaurus that can help you find them.

Given that there are nearly three million records in PsycINFO, you may have to try a variety of search terms in different combinations and at different levels of specificity before you find what you are looking for. Imagine, for example, that you are interested in the question of whether women and men differ in terms of their ability to recall experiences from when they were very young. If you were to enter “memory for early experiences” as your search term, PsycINFO would return only six records, most of which are not particularly relevant to your question. However, if you were to enter the search term “memory,” it would return 149,777 records—far too many to look through individually. This is where the thesaurus helps. Entering “memory” into the thesaurus provides several more specific index terms—one of which is “early memories.” While searching for “early memories” among the index terms returns 1,446 records—still too many too look through individually—combining it with “human sex differences” as a second search term returns 37 articles, many of which are highly relevant to the topic.

Depending on the vendor that provides the interface to PsycINFO, you may be able to save, print, or e-mail the relevant PsycINFO records. The records might even contain links to full-text copies of the works themselves. (PsycARTICLES is a database that provides full-text access to articles in all journals published by the APA.) If not, and you want a copy of the work, you will have to find out if your library carries the journal or has the book and the hard copy on the library shelves. Be sure to ask a librarian if you need help.

Using Other Search Techniques

In addition to entering search terms into PsycINFO and other databases, there are several other techniques you can use to search the research literature. First, if you have one good article or book chapter on your topic—a recent review article is best—you can look through the reference list of that article for other relevant articles, books, and book chapters. In fact, you should do this with any relevant article or book chapter you find. You can also start with a classic article or book chapter on your topic, find its record in PsycINFO (by entering the author’s name or article’s title as a search term), and link from there to a list of other works in PsycINFO that cite that classic article. This works because other researchers working on your topic are likely to be aware of the classic article and cite it in their own work. You can also do a general Internet search using search terms related to your topic or the name of a researcher who conducts research on your topic. This might lead you directly to works that are part of the research literature (e.g., articles in open-access journals or posted on researchers’ own websites). The search engine Google Scholar is especially useful for this purpose. A general Internet search might also lead you to websites that are not part of the research literature but might provide references to works that are. Finally, you can talk to people (e.g., your instructor or other faculty members in psychology) who know something about your topic and can suggest relevant articles and book chapters.

What to Search For

When you do a literature review, you need to be selective. Not every article, book chapter, and book that relates to your research idea or question will be worth obtaining, reading, and integrating into your review. Instead, you want to focus on sources that help you do four basic things: (a) refine your research question, (b) identify appropriate research methods, (c) place your research in the context of previous research, and (d) write an effective research report. Several basic principles can help you find the most useful sources.

First, it is best to focus on recent research, keeping in mind that what counts as recent depends on the topic. For newer topics that are actively being studied, “recent” might mean published in the past year or two. For older topics that are receiving less attention right now, “recent” might mean within the past 10 years. You will get a feel for what counts as recent for your topic when you start your literature search. A good general rule, however, is to start with sources published in the past five years. The main exception to this rule would be classic articles that turn up in the reference list of nearly every other source. If other researchers think that this work is important, even though it is old, then by all means you should include it in your review.

Second, you should look for review articles on your topic because they will provide a useful overview of it—often discussing important definitions, results, theories, trends, and controversies—giving you a good sense of where your own research fits into the literature. You should also look for empirical research reports addressing your question or similar questions, which can give you ideas about how to operationally define your variables and collect your data. As a general rule, it is good to use methods that others have already used successfully unless you have good reasons not to. Finally, you should look for sources that provide information that can help you argue for the interestingness of your research question. For a study on the effects of cell phone use on driving ability, for example, you might look for information about how widespread cell phone use is, how frequent and costly motor vehicle crashes are, and so on.

How many sources are enough for your literature review? This is a difficult question because it depends on how extensively your topic has been studied and also on your own goals. One study found that across a variety of professional journals in psychology, the average number of sources cited per article was about 50 (Adair & Vohra, 2003). This gives a rough idea of what professional researchers consider to be adequate. As a student, you might be assigned a much lower minimum number of references to use, but the principles for selecting the most useful ones remain the same.

Key Takeaways

• The research literature in psychology is all the published research in psychology, consisting primarily of articles in professional journals and scholarly books.
• Early in the research process, it is important to conduct a review of the research literature on your topic to refine your research question, identify appropriate research methods, place your question in the context of other research, and prepare to write an effective research report.
• There are several strategies for finding previous research on your topic. Among the best is using PsycINFO, a computer database that catalogs millions of articles, books, and book chapters in psychology and related fields.
• Practice: Use the techniques discussed in this section to find 10 journal articles and book chapters on one of the following research ideas: memory for smells, aggressive driving, the causes of narcissistic personality disorder, the functions of the intraparietal sulcus, or prejudice against the physically handicapped.

Adair, J. G., & Vohra, N. (2003). The explosion of knowledge, references, and citations: Psychology’s unique response to a crisis. American Psychologist, 58 , 15–23.

Research Methods in Psychology Copyright © 2016 by University of Minnesota is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License , except where otherwise noted.

• Meriam Library

SWRK 330 - Social Work Research Methods

• Literature Reviews and Empirical Research
• Databases and Search Tips
• Article Citations
• Scholarly Journal Evaulation
• Statistical Sources
• Books and eBooks

What is a Literature Review?

Empirical research.

• Annotated Bibliographies

A literature review  summarizes and discusses previous publications  on a topic.

It should also:

explore past research and its strengths and weaknesses.

be used to validate the target and methods you have chosen for your proposed research.

consist of books and scholarly journals that provide research examples of populations or settings similar to your own, as well as community resources to document the need for your proposed research.

The literature review does not present new  primary  scholarship. 

be completed in the correct citation format requested by your professor  (see the  C itations Tab)

Access Purdue  OWL's Social Work Literature Review Guidelines here .  

Empirical Research  is  research  that is based on experimentation or observation, i.e. Evidence. Such  research  is often conducted to answer a specific question or to test a hypothesis (educated guess).

How do you know if a study is empirical? Read the subheadings within the article, book, or report and look for a description of the research "methodology."  Ask yourself: Could I recreate this study and test these results?

These are some key features to look for when identifying empirical research.

NOTE:  Not all of these features will be in every empirical research article, some may be excluded, use this only as a guide.

• Statement of methodology
• Research questions are clear and measurable
• Individuals, group, subjects which are being studied are identified/defined
• Data is presented regarding the findings
• Controls or instruments such as surveys or tests were conducted
• There is a literature review
• There is discussion of the results included
• Citations/references are included

See also Empirical Research Guide

• << Previous: Citations
• Next: Annotated Bibliographies >>
• Last Updated: Feb 6, 2024 8:38 AM
• URL: https://libguides.csuchico.edu/SWRK330

Meriam Library | CSU, Chico

• University of Memphis Libraries
• Research Guides

Empirical Research: Defining, Identifying, & Finding

Searching for empirical research.

• Defining Empirical Research
• Introduction

Where Do I Find Empirical Research?

How do i find more empirical research in my search.

• Database Tools
• Search Terms
• Image Descriptions

Because empirical research refers to the method of investigation rather than a method of publication, it can be published in a number of places. In many disciplines empirical research is most commonly published in scholarly, peer-reviewed journals . Putting empirical research through the peer review process helps ensure that the research is high quality. 

Finding Peer-Reviewed Articles

You can find peer-reviewed articles in a general web search along with a lot of other types of sources. However, these specialized tools are more likely to find peer-reviewed articles:

• Library databases
• Academic search engines such as Google Scholar

Common Types of Articles That Are Not Empirical

However, just finding an article in a peer-reviewed journal is not enough to say it is empirical, since not all the articles in a peer-reviewed journal will be empirical research or even peer reviewed. Knowing how to quickly identify some types non-empirical research articles in peer-reviewed journals can help speed up your search. 

• Peer-reviewed articles that systematically discuss and propose abstract concepts and methods for a field without primary data collection.
• Example: Grosser, K. & Moon, J. (2019). CSR and feminist organization studies: Towards an integrated theorization for the analysis of gender issues .
• Peer-reviewed articles that systematically describe, summarize, and often categorize and evaluate previous research on a topic without collecting new data.
• Example: Heuer, S. & Willer, R. (2020). How is quality of life assessed in people with dementia? A systematic literature review and a primer for speech-language pathologists .
• Note: empirical research articles will have a literature review section as part of the Introduction , but in an empirical research article the literature review exists to give context to the empirical research, which is the primary focus of the article. In a literature review article, the literature review is the focus. 
• While these articles are not empirical, they are often a great source of information on previous empirical research on a topic with citations to find that research.
• Non-peer-reviewed articles where the authors discuss their thoughts on a particular topic without data collection and a systematic method. There are a few differences between these types of articles.
• Written by the editors or guest editors of the journal. 
• Example:  Naples, N. A., Mauldin, L., & Dillaway, H. (2018). From the guest editors: Gender, disability, and intersectionality .
• Written by guest authors. The journal may have a non-peer-reviewed process for authors to submit these articles, and the editors of the journal may invite authors to write opinion articles.
• Example: García, J. J.-L., & Sharif, M. Z. (2015). Black lives matter: A commentary on racism and public health . 
• Written by the readers of a journal, often in response to an article previously-published in the journal.
• Example: Nathan, M. (2013). Letters: Perceived discrimination and racial/ethnic disparities in youth problem behaviors . 
• Non-peer-reviewed articles that describe and evaluate books, products, services, and other things the audience of the journal would be interested in. 
• Example: Robinson, R. & Green, J. M. (2020). Book review: Microaggressions and traumatic stress: Theory, research, and clinical treatment .

Even once you know how to recognize empirical research and where it is published, it would be nice to improve your search results so that more empirical research shows up for your topic.

There are two major ways to find the empirical research in a database search:

• Use built-in database tools to limit results to empirical research.
• Include search terms that help identify empirical research.
• << Previous: Discussion
• Next: Database Tools >>
• Last Updated: Apr 2, 2024 11:25 AM
• URL: https://libguides.memphis.edu/empirical-research

Empirical Research

Introduction, what is empirical research, attribution.

• Finding Empirical Research in Library Databases
• Designing Empirical Research
• Case Sudies

Empirical research is based on observed and measured phenomena and derives knowledge from actual experience rather than from theory or belief. 

How do you know if a study is empirical? Read the subheadings within the article, book, or report and look for a description of the research "methodology."  Ask yourself: Could I recreate this study and test these results?

Key characteristics to look for:

• Specific research questions to be answered
• Definition of the population, behavior, or   phenomena being studied
• Description of the process used to study this population or phenomena, including selection criteria, controls, and testing instruments (such as surveys)

Another hint: some scholarly journals use a specific layout, called the "IMRaD" format, to communicate empirical research findings. Such articles typically have 4 components:

• Introduction : sometimes called "literature review" -- what is currently known about the topic -- usually includes a theoretical framework and/or discussion of previous studies
• Methodology: sometimes called "research design" -- how to recreate the study -- usually describes the population, research process, and analytical tools
• Results : sometimes called "findings" -- what was learned through the study -- usually appears as statistical data or as substantial quotations from research participants
• Discussion : sometimes called "conclusion" or "implications" -- why the study is important -- usually describes how the research results influence professional practices or future studies

Portions of this guide were built using suggestions from other libraries, including Penn State and Utah State University libraries.

• Next: Finding Empirical Research in Library Databases >>
• Last Updated: Jan 10, 2023 8:31 AM
• URL: https://enmu.libguides.com/EmpiricalResearch

A systematic literature review of empirical research on ChatGPT in education

• Open access
• Published: 26 May 2024
• Volume 3 , article number  60 , ( 2024 )

Cite this article

You have full access to this open access article

• Yazid Albadarin   ORCID: orcid.org/0009-0005-8068-8902 1 ,
• Mohammed Saqr 1 ,
• Nicolas Pope 1 &
• Markku Tukiainen 1  

Over the last four decades, studies have investigated the incorporation of Artificial Intelligence (AI) into education. A recent prominent AI-powered technology that has impacted the education sector is ChatGPT. This article provides a systematic review of 14 empirical studies incorporating ChatGPT into various educational settings, published in 2022 and before the 10th of April 2023—the date of conducting the search process. It carefully followed the essential steps outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines, as well as Okoli’s (Okoli in Commun Assoc Inf Syst, 2015) steps for conducting a rigorous and transparent systematic review. In this review, we aimed to explore how students and teachers have utilized ChatGPT in various educational settings, as well as the primary findings of those studies. By employing Creswell’s (Creswell in Educational research: planning, conducting, and evaluating quantitative and qualitative research [Ebook], Pearson Education, London, 2015) coding techniques for data extraction and interpretation, we sought to gain insight into their initial attempts at ChatGPT incorporation into education. This approach also enabled us to extract insights and considerations that can facilitate its effective and responsible use in future educational contexts. The results of this review show that learners have utilized ChatGPT as a virtual intelligent assistant, where it offered instant feedback, on-demand answers, and explanations of complex topics. Additionally, learners have used it to enhance their writing and language skills by generating ideas, composing essays, summarizing, translating, paraphrasing texts, or checking grammar. Moreover, learners turned to it as an aiding tool to facilitate their directed and personalized learning by assisting in understanding concepts and homework, providing structured learning plans, and clarifying assignments and tasks. However, the results of specific studies (n = 3, 21.4%) show that overuse of ChatGPT may negatively impact innovative capacities and collaborative learning competencies among learners. Educators, on the other hand, have utilized ChatGPT to create lesson plans, generate quizzes, and provide additional resources, which helped them enhance their productivity and efficiency and promote different teaching methodologies. Despite these benefits, the majority of the reviewed studies recommend the importance of conducting structured training, support, and clear guidelines for both learners and educators to mitigate the drawbacks. This includes developing critical evaluation skills to assess the accuracy and relevance of information provided by ChatGPT, as well as strategies for integrating human interaction and collaboration into learning activities that involve AI tools. Furthermore, they also recommend ongoing research and proactive dialogue with policymakers, stakeholders, and educational practitioners to refine and enhance the use of AI in learning environments. This review could serve as an insightful resource for practitioners who seek to integrate ChatGPT into education and stimulate further research in the field.

Avoid common mistakes on your manuscript.

1 Introduction

Educational technology, a rapidly evolving field, plays a crucial role in reshaping the landscape of teaching and learning [ 82 ]. One of the most transformative technological innovations of our era that has influenced the field of education is Artificial Intelligence (AI) [ 50 ]. Over the last four decades, AI in education (AIEd) has gained remarkable attention for its potential to make significant advancements in learning, instructional methods, and administrative tasks within educational settings [ 11 ]. In particular, a large language model (LLM), a type of AI algorithm that applies artificial neural networks (ANNs) and uses massively large data sets to understand, summarize, generate, and predict new content that is almost difficult to differentiate from human creations [ 79 ], has opened up novel possibilities for enhancing various aspects of education, from content creation to personalized instruction [ 35 ]. Chatbots that leverage the capabilities of LLMs to understand and generate human-like responses have also presented the capacity to enhance student learning and educational outcomes by engaging students, offering timely support, and fostering interactive learning experiences [ 46 ].

The ongoing and remarkable technological advancements in chatbots have made their use more convenient, increasingly natural and effortless, and have expanded their potential for deployment across various domains [ 70 ]. One prominent example of chatbot applications is the Chat Generative Pre-Trained Transformer, known as ChatGPT, which was introduced by OpenAI, a leading AI research lab, on November 30th, 2022. ChatGPT employs a variety of deep learning techniques to generate human-like text, with a particular focus on recurrent neural networks (RNNs). Long short-term memory (LSTM) allows it to grasp the context of the text being processed and retain information from previous inputs. Also, the transformer architecture, a neural network architecture based on the self-attention mechanism, allows it to analyze specific parts of the input, thereby enabling it to produce more natural-sounding and coherent output. Additionally, the unsupervised generative pre-training and the fine-tuning methods allow ChatGPT to generate more relevant and accurate text for specific tasks [ 31 , 62 ]. Furthermore, reinforcement learning from human feedback (RLHF), a machine learning approach that combines reinforcement learning techniques with human-provided feedback, has helped improve ChatGPT’s model by accelerating the learning process and making it significantly more efficient.

This cutting-edge natural language processing (NLP) tool is widely recognized as one of today's most advanced LLMs-based chatbots [ 70 ], allowing users to ask questions and receive detailed, coherent, systematic, personalized, convincing, and informative human-like responses [ 55 ], even within complex and ambiguous contexts [ 63 , 77 ]. ChatGPT is considered the fastest-growing technology in history: in just three months following its public launch, it amassed an estimated 120 million monthly active users [ 16 ] with an estimated 13 million daily queries [ 49 ], surpassing all other applications [ 64 ]. This remarkable growth can be attributed to the unique features and user-friendly interface that ChatGPT offers. Its intuitive design allows users to interact seamlessly with the technology, making it accessible to a diverse range of individuals, regardless of their technical expertise [ 78 ]. Additionally, its exceptional performance results from a combination of advanced algorithms, continuous enhancements, and extensive training on a diverse dataset that includes various text sources such as books, articles, websites, and online forums [ 63 ], have contributed to a more engaging and satisfying user experience [ 62 ]. These factors collectively explain its remarkable global growth and set it apart from predecessors like Bard, Bing Chat, ERNIE, and others.

In this context, several studies have explored the technological advancements of chatbots. One noteworthy recent research effort, conducted by Schöbel et al. [ 70 ], stands out for its comprehensive analysis of more than 5,000 studies on communication agents. This study offered a comprehensive overview of the historical progression and future prospects of communication agents, including ChatGPT. Moreover, other studies have focused on making comparisons, particularly between ChatGPT and alternative chatbots like Bard, Bing Chat, ERNIE, LaMDA, BlenderBot, and various others. For example, O’Leary [ 53 ] compared two chatbots, LaMDA and BlenderBot, with ChatGPT and revealed that ChatGPT outperformed both. This superiority arises from ChatGPT’s capacity to handle a wider range of questions and generate slightly varied perspectives within specific contexts. Similarly, ChatGPT exhibited an impressive ability to formulate interpretable responses that were easily understood when compared with Google's feature snippet [ 34 ]. Additionally, ChatGPT was compared to other LLMs-based chatbots, including Bard and BERT, as well as ERNIE. The findings indicated that ChatGPT exhibited strong performance in the given tasks, often outperforming the other models [ 59 ].

Furthermore, in the education context, a comprehensive study systematically compared a range of the most promising chatbots, including Bard, Bing Chat, ChatGPT, and Ernie across a multidisciplinary test that required higher-order thinking. The study revealed that ChatGPT achieved the highest score, surpassing Bing Chat and Bard [ 64 ]. Similarly, a comparative analysis was conducted to compare ChatGPT with Bard in answering a set of 30 mathematical questions and logic problems, grouped into two question sets. Set (A) is unavailable online, while Set (B) is available online. The results revealed ChatGPT's superiority in Set (A) over Bard. Nevertheless, Bard's advantage emerged in Set (B) due to its capacity to access the internet directly and retrieve answers, a capability that ChatGPT does not possess [ 57 ]. However, through these varied assessments, ChatGPT consistently highlights its exceptional prowess compared to various alternatives in the ever-evolving chatbot technology.

The widespread adoption of chatbots, especially ChatGPT, by millions of students and educators, has sparked extensive discussions regarding its incorporation into the education sector [ 64 ]. Accordingly, many scholars have contributed to the discourse, expressing both optimism and pessimism regarding the incorporation of ChatGPT into education. For example, ChatGPT has been highlighted for its capabilities in enriching the learning and teaching experience through its ability to support different learning approaches, including adaptive learning, personalized learning, and self-directed learning [ 58 , 60 , 91 ]), deliver summative and formative feedback to students and provide real-time responses to questions, increase the accessibility of information [ 22 , 40 , 43 ], foster students’ performance, engagement and motivation [ 14 , 44 , 58 ], and enhance teaching practices [ 17 , 18 , 64 , 74 ].

On the other hand, concerns have been also raised regarding its potential negative effects on learning and teaching. These include the dissemination of false information and references [ 12 , 23 , 61 , 85 ], biased reinforcement [ 47 , 50 ], compromised academic integrity [ 18 , 40 , 66 , 74 ], and the potential decline in students' skills [ 43 , 61 , 64 , 74 ]. As a result, ChatGPT has been banned in multiple countries, including Russia, China, Venezuela, Belarus, and Iran, as well as in various educational institutions in India, Italy, Western Australia, France, and the United States [ 52 , 90 ].

Clearly, the advent of chatbots, especially ChatGPT, has provoked significant controversy due to their potential impact on learning and teaching. This indicates the necessity for further exploration to gain a deeper understanding of this technology and carefully evaluate its potential benefits, limitations, challenges, and threats to education [ 79 ]. Therefore, conducting a systematic literature review will provide valuable insights into the potential prospects and obstacles linked to its incorporation into education. This systematic literature review will primarily focus on ChatGPT, driven by the aforementioned key factors outlined above.

However, the existing literature lacks a systematic literature review of empirical studies. Thus, this systematic literature review aims to address this gap by synthesizing the existing empirical studies conducted on chatbots, particularly ChatGPT, in the field of education, highlighting how ChatGPT has been utilized in educational settings, and identifying any existing gaps. This review may be particularly useful for researchers in the field and educators who are contemplating the integration of ChatGPT or any chatbot into education. The following research questions will guide this study:

What are students' and teachers' initial attempts at utilizing ChatGPT in education?

What are the main findings derived from empirical studies that have incorporated ChatGPT into learning and teaching?

2 Methodology

To conduct this study, the authors followed the essential steps of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) and Okoli’s [ 54 ] steps for conducting a systematic review. These included identifying the study’s purpose, drafting a protocol, applying a practical screening process, searching the literature, extracting relevant data, evaluating the quality of the included studies, synthesizing the studies, and ultimately writing the review. The subsequent section provides an extensive explanation of how these steps were carried out in this study.

2.1 Identify the purpose

Given the widespread adoption of ChatGPT by students and teachers for various educational purposes, often without a thorough understanding of responsible and effective use or a clear recognition of its potential impact on learning and teaching, the authors recognized the need for further exploration of ChatGPT's impact on education in this early stage. Therefore, they have chosen to conduct a systematic literature review of existing empirical studies that incorporate ChatGPT into educational settings. Despite the limited number of empirical studies due to the novelty of the topic, their goal is to gain a deeper understanding of this technology and proactively evaluate its potential benefits, limitations, challenges, and threats to education. This effort could help to understand initial reactions and attempts at incorporating ChatGPT into education and bring out insights and considerations that can inform the future development of education.

2.2 Draft the protocol

The next step is formulating the protocol. This protocol serves to outline the study process in a rigorous and transparent manner, mitigating researcher bias in study selection and data extraction [ 88 ]. The protocol will include the following steps: generating the research question, predefining a literature search strategy, identifying search locations, establishing selection criteria, assessing the studies, developing a data extraction strategy, and creating a timeline.

2.3 Apply practical screen

The screening step aims to accurately filter the articles resulting from the searching step and select the empirical studies that have incorporated ChatGPT into educational contexts, which will guide us in answering the research questions and achieving the objectives of this study. To ensure the rigorous execution of this step, our inclusion and exclusion criteria were determined based on the authors' experience and informed by previous successful systematic reviews [ 21 ]. Table 1 summarizes the inclusion and exclusion criteria for study selection.

2.4 Literature search

We conducted a thorough literature search to identify articles that explored, examined, and addressed the use of ChatGPT in Educational contexts. We utilized two research databases: Dimensions.ai, which provides access to a large number of research publications, and lens.org, which offers access to over 300 million articles, patents, and other research outputs from diverse sources. Additionally, we included three databases, Scopus, Web of Knowledge, and ERIC, which contain relevant research on the topic that addresses our research questions. To browse and identify relevant articles, we used the following search formula: ("ChatGPT" AND "Education"), which included the Boolean operator "AND" to get more specific results. The subject area in the Scopus and ERIC databases were narrowed to "ChatGPT" and "Education" keywords, and in the WoS database was limited to the "Education" category. The search was conducted between the 3rd and 10th of April 2023, which resulted in 276 articles from all selected databases (111 articles from Dimensions.ai, 65 from Scopus, 28 from Web of Science, 14 from ERIC, and 58 from Lens.org). These articles were imported into the Rayyan web-based system for analysis. The duplicates were identified automatically by the system. Subsequently, the first author manually reviewed the duplicated articles ensured that they had the same content, and then removed them, leaving us with 135 unique articles. Afterward, the titles, abstracts, and keywords of the first 40 manuscripts were scanned and reviewed by the first author and were discussed with the second and third authors to resolve any disagreements. Subsequently, the first author proceeded with the filtering process for all articles and carefully applied the inclusion and exclusion criteria as presented in Table  1 . Articles that met any one of the exclusion criteria were eliminated, resulting in 26 articles. Afterward, the authors met to carefully scan and discuss them. The authors agreed to eliminate any empirical studies solely focused on checking ChatGPT capabilities, as these studies do not guide us in addressing the research questions and achieving the study's objectives. This resulted in 14 articles eligible for analysis.

2.5 Quality appraisal

The examination and evaluation of the quality of the extracted articles is a vital step [ 9 ]. Therefore, the extracted articles were carefully evaluated for quality using Fink’s [ 24 ] standards, which emphasize the necessity for detailed descriptions of methodology, results, conclusions, strengths, and limitations. The process began with a thorough assessment of each study's design, data collection, and analysis methods to ensure their appropriateness and comprehensive execution. The clarity, consistency, and logical progression from data to results and conclusions were also critically examined. Potential biases and recognized limitations within the studies were also scrutinized. Ultimately, two articles were excluded for failing to meet Fink’s criteria, particularly in providing sufficient detail on methodology, results, conclusions, strengths, or limitations. The review process is illustrated in Fig.  1 .

The study selection process

2.6 Data extraction

The next step is data extraction, the process of capturing the key information and categories from the included studies. To improve efficiency, reduce variation among authors, and minimize errors in data analysis, the coding categories were constructed using Creswell's [ 15 ] coding techniques for data extraction and interpretation. The coding process involves three sequential steps. The initial stage encompasses open coding , where the researcher examines the data, generates codes to describe and categorize it, and gains a deeper understanding without preconceived ideas. Following open coding is axial coding , where the interrelationships between codes from open coding are analyzed to establish more comprehensive categories or themes. The process concludes with selective coding , refining and integrating categories or themes to identify core concepts emerging from the data. The first coder performed the coding process, then engaged in discussions with the second and third authors to finalize the coding categories for the first five articles. The first coder then proceeded to code all studies and engaged again in discussions with the other authors to ensure the finalization of the coding process. After a comprehensive analysis and capturing of the key information from the included studies, the data extraction and interpretation process yielded several themes. These themes have been categorized and are presented in Table  2 . It is important to note that open coding results were removed from Table  2 for aesthetic reasons, as it included many generic aspects, such as words, short phrases, or sentences mentioned in the studies.

2.7 Synthesize studies

In this stage, we will gather, discuss, and analyze the key findings that emerged from the selected studies. The synthesis stage is considered a transition from an author-centric to a concept-centric focus, enabling us to map all the provided information to achieve the most effective evaluation of the data [ 87 ]. Initially, the authors extracted data that included general information about the selected studies, including the author(s)' names, study titles, years of publication, educational levels, research methodologies, sample sizes, participants, main aims or objectives, raw data sources, and analysis methods. Following that, all key information and significant results from the selected studies were compiled using Creswell’s [ 15 ] coding techniques for data extraction and interpretation to identify core concepts and themes emerging from the data, focusing on those that directly contributed to our research questions and objectives, such as the initial utilization of ChatGPT in learning and teaching, learners' and educators' familiarity with ChatGPT, and the main findings of each study. Finally, the data related to each selected study were extracted into an Excel spreadsheet for data processing. The Excel spreadsheet was reviewed by the authors, including a series of discussions to ensure the finalization of this process and prepare it for further analysis. Afterward, the final result being analyzed and presented in various types of charts and graphs. Table 4 presents the extracted data from the selected studies, with each study labeled with a capital 'S' followed by a number.

This section consists of two main parts. The first part provides a descriptive analysis of the data compiled from the reviewed studies. The second part presents the answers to the research questions and the main findings of these studies.

3.1 Part 1: descriptive analysis

This section will provide a descriptive analysis of the reviewed studies, including educational levels and fields, participants distribution, country contribution, research methodologies, study sample size, study population, publication year, list of journals, familiarity with ChatGPT, source of data, and the main aims and objectives of the studies. Table 4 presents a comprehensive overview of the extracted data from the selected studies.

3.1.1 The number of the reviewed studies and publication years

The total number of the reviewed studies was 14. All studies were empirical studies and published in different journals focusing on Education and Technology. One study was published in 2022 [S1], while the remaining were published in 2023 [S2]-[S14]. Table 3 illustrates the year of publication, the names of the journals, and the number of reviewed studies published in each journal for the studies reviewed.

3.1.2 Educational levels and fields

The majority of the reviewed studies, 11 studies, were conducted in higher education institutions [S1]-[S10] and [S13]. Two studies did not specify the educational level of the population [S12] and [S14], while one study focused on elementary education [S11]. However, the reviewed studies covered various fields of education. Three studies focused on Arts and Humanities Education [S8], [S11], and [S14], specifically English Education. Two studies focused on Engineering Education, with one in Computer Engineering [S2] and the other in Construction Education [S3]. Two studies focused on Mathematics Education [S5] and [S12]. One study focused on Social Science Education [S13]. One study focused on Early Education [S4]. One study focused on Journalism Education [S9]. Finally, three studies did not specify the field of education [S1], [S6], and [S7]. Figure  2 represents the educational levels in the reviewed studies, while Fig.  3 represents the context of the reviewed studies.

Educational levels in the reviewed studies

Context of the reviewed studies

3.1.3 Participants distribution and countries contribution

The reviewed studies have been conducted across different geographic regions, providing a diverse representation of the studies. The majority of the studies, 10 in total, [S1]-[S3], [S5]-[S9], [S11], and [S14], primarily focused on participants from single countries such as Pakistan, the United Arab Emirates, China, Indonesia, Poland, Saudi Arabia, South Korea, Spain, Tajikistan, and the United States. In contrast, four studies, [S4], [S10], [S12], and [S13], involved participants from multiple countries, including China and the United States [S4], China, the United Kingdom, and the United States [S10], the United Arab Emirates, Oman, Saudi Arabia, and Jordan [S12], Turkey, Sweden, Canada, and Australia [ 13 ]. Figures  4 and 5 illustrate the distribution of participants, whether from single or multiple countries, and the contribution of each country in the reviewed studies, respectively.

The reviewed studies conducted in single or multiple countries

The Contribution of each country in the studies

3.1.4 Study population and sample size

Four study populations were included: university students, university teachers, university teachers and students, and elementary school teachers. Six studies involved university students [S2], [S3], [S5] and [S6]-[S8]. Three studies focused on university teachers [S1], [S4], and [S6], while one study specifically targeted elementary school teachers [S11]. Additionally, four studies included both university teachers and students [S10] and [ 12 , 13 , 14 ], and among them, study [S13] specifically included postgraduate students. In terms of the sample size of the reviewed studies, nine studies included a small sample size of less than 50 participants [S1], [S3], [S6], [S8], and [S10]-[S13]. Three studies had 50–100 participants [S2], [S9], and [S14]. Only one study had more than 100 participants [S7]. It is worth mentioning that study [S4] adopted a mixed methods approach, including 10 participants for qualitative analysis and 110 participants for quantitative analysis.

3.1.5 Participants’ familiarity with using ChatGPT

The reviewed studies recruited a diverse range of participants with varying levels of familiarity with ChatGPT. Five studies [S2], [S4], [S6], [S8], and [S12] involved participants already familiar with ChatGPT, while eight studies [S1], [S3], [S5], [S7], [S9], [S10], [S13] and [S14] included individuals with differing levels of familiarity. Notably, one study [S11] had participants who were entirely unfamiliar with ChatGPT. It is important to note that four studies [S3], [S5], [S9], and [S11] provided training or guidance to their participants before conducting their studies, while ten studies [S1], [S2], [S4], [S6]-[S8], [S10], and [S12]-[S14] did not provide training due to the participants' existing familiarity with ChatGPT.

3.1.6 Research methodology approaches and source(S) of data

The reviewed studies adopted various research methodology approaches. Seven studies adopted qualitative research methodology [S1], [S4], [S6], [S8], [S10], [S11], and [S12], while three studies adopted quantitative research methodology [S3], [S7], and [S14], and four studies employed mixed-methods, which involved a combination of both the strengths of qualitative and quantitative methods [S2], [S5], [S9], and [S13].

In terms of the source(s) of data, the reviewed studies obtained their data from various sources, such as interviews, questionnaires, and pre-and post-tests. Six studies relied on interviews as their primary source of data collection [S1], [S4], [S6], [S10], [S11], and [S12], four studies relied on questionnaires [S2], [S7], [S13], and [S14], two studies combined the use of pre-and post-tests and questionnaires for data collection [S3] and [S9], while two studies combined the use of questionnaires and interviews to obtain the data [S5] and [S8]. It is important to note that six of the reviewed studies were quasi-experimental [S3], [S5], [S8], [S9], [S12], and [S14], while the remaining ones were experimental studies [S1], [S2], [S4], [S6], [S7], [S10], [S11], and [S13]. Figures  6 and 7 illustrate the research methodologies and the source (s) of data used in the reviewed studies, respectively.

Research methodologies in the reviewed studies

Source of data in the reviewed studies

3.1.7 The aim and objectives of the studies

The reviewed studies encompassed a diverse set of aims, with several of them incorporating multiple primary objectives. Six studies [S3], [S6], [S7], [S8], [S11], and [S12] examined the integration of ChatGPT in educational contexts, and four studies [S4], [S5], [S13], and [S14] investigated the various implications of its use in education, while three studies [S2], [S9], and [S10] aimed to explore both its integration and implications in education. Additionally, seven studies explicitly explored attitudes and perceptions of students [S2] and [S3], educators [S1] and [S6], or both [S10], [S12], and [S13] regarding the utilization of ChatGPT in educational settings.

3.2 Part 2: research questions and main findings of the reviewed studies

This part will present the answers to the research questions and the main findings of the reviewed studies, classified into two main categories (learning and teaching) according to AI Education classification by [ 36 ]. Figure  8 summarizes the main findings of the reviewed studies in a visually informative diagram. Table 4 provides a detailed list of the key information extracted from the selected studies that led to generating these themes.

The main findings in the reviewed studies

4 Students' initial attempts at utilizing ChatGPT in learning and main findings from students' perspective

4.1 virtual intelligent assistant.

Nine studies demonstrated that ChatGPT has been utilized by students as an intelligent assistant to enhance and support their learning. Students employed it for various purposes, such as answering on-demand questions [S2]-[S5], [S8], [S10], and [S12], providing valuable information and learning resources [S2]-[S5], [S6], and [S8], as well as receiving immediate feedback [S2], [S4], [S9], [S10], and [S12]. In this regard, students generally were confident in the accuracy of ChatGPT's responses, considering them relevant, reliable, and detailed [S3], [S4], [S5], and [S8]. However, some students indicated the need for improvement, as they found that answers are not always accurate [S2], and that misleading information may have been provided or that it may not always align with their expectations [S6] and [S10]. It was also observed by the students that the accuracy of ChatGPT is dependent on several factors, including the quality and specificity of the user's input, the complexity of the question or topic, and the scope and relevance of its training data [S12]. Many students felt that ChatGPT's answers were not always accurate and most of them believed that it requires good background knowledge to work with.

4.2 Writing and language proficiency assistant

Six of the reviewed studies highlighted that ChatGPT has been utilized by students as a valuable assistant tool to improve their academic writing skills and language proficiency. Among these studies, three mainly focused on English education, demonstrating that students showed sufficient mastery in using ChatGPT for generating ideas, summarizing, paraphrasing texts, and completing writing essays [S8], [S11], and [S14]. Furthermore, ChatGPT helped them in writing by making students active investigators rather than passive knowledge recipients and facilitated the development of their writing skills [S11] and [S14]. Similarly, ChatGPT allowed students to generate unique ideas and perspectives, leading to deeper analysis and reflection on their journalism writing [S9]. In terms of language proficiency, ChatGPT allowed participants to translate content into their home languages, making it more accessible and relevant to their context [S4]. It also enabled them to request changes in linguistic tones or flavors [S8]. Moreover, participants used it to check grammar or as a dictionary [S11].

4.3 Valuable resource for learning approaches

Five studies demonstrated that students used ChatGPT as a valuable complementary resource for self-directed learning. It provided learning resources and guidance on diverse educational topics and created a supportive home learning environment [S2] and [S4]. Moreover, it offered step-by-step guidance to grasp concepts at their own pace and enhance their understanding [S5], streamlined task and project completion carried out independently [S7], provided comprehensive and easy-to-understand explanations on various subjects [S10], and assisted in studying geometry operations, thereby empowering them to explore geometry operations at their own pace [S12]. Three studies showed that students used ChatGPT as a valuable learning resource for personalized learning. It delivered age-appropriate conversations and tailored teaching based on a child's interests [S4], acted as a personalized learning assistant, adapted to their needs and pace, which assisted them in understanding mathematical concepts [S12], and enabled personalized learning experiences in social sciences by adapting to students' needs and learning styles [S13]. On the other hand, it is important to note that, according to one study [S5], students suggested that using ChatGPT may negatively affect collaborative learning competencies between students.

4.4 Enhancing students' competencies

Six of the reviewed studies have shown that ChatGPT is a valuable tool for improving a wide range of skills among students. Two studies have provided evidence that ChatGPT led to improvements in students' critical thinking, reasoning skills, and hazard recognition competencies through engaging them in interactive conversations or activities and providing responses related to their disciplines in journalism [S5] and construction education [S9]. Furthermore, two studies focused on mathematical education have shown the positive impact of ChatGPT on students' problem-solving abilities in unraveling problem-solving questions [S12] and enhancing the students' understanding of the problem-solving process [S5]. Lastly, one study indicated that ChatGPT effectively contributed to the enhancement of conversational social skills [S4].

4.5 Supporting students' academic success

Seven of the reviewed studies highlighted that students found ChatGPT to be beneficial for learning as it enhanced learning efficiency and improved the learning experience. It has been observed to improve students' efficiency in computer engineering studies by providing well-structured responses and good explanations [S2]. Additionally, students found it extremely useful for hazard reporting [S3], and it also enhanced their efficiency in solving mathematics problems and capabilities [S5] and [S12]. Furthermore, by finding information, generating ideas, translating texts, and providing alternative questions, ChatGPT aided students in deepening their understanding of various subjects [S6]. It contributed to an increase in students' overall productivity [S7] and improved efficiency in composing written tasks [S8]. Regarding learning experiences, ChatGPT was instrumental in assisting students in identifying hazards that they might have otherwise overlooked [S3]. It also improved students' learning experiences in solving mathematics problems and developing abilities [S5] and [S12]. Moreover, it increased students' successful completion of important tasks in their studies [S7], particularly those involving average difficulty writing tasks [S8]. Additionally, ChatGPT increased the chances of educational success by providing students with baseline knowledge on various topics [S10].

5 Teachers' initial attempts at utilizing ChatGPT in teaching and main findings from teachers' perspective

5.1 valuable resource for teaching.

The reviewed studies showed that teachers have employed ChatGPT to recommend, modify, and generate diverse, creative, organized, and engaging educational contents, teaching materials, and testing resources more rapidly [S4], [S6], [S10] and [S11]. Additionally, teachers experienced increased productivity as ChatGPT facilitated quick and accurate responses to questions, fact-checking, and information searches [S1]. It also proved valuable in constructing new knowledge [S6] and providing timely answers to students' questions in classrooms [S11]. Moreover, ChatGPT enhanced teachers' efficiency by generating new ideas for activities and preplanning activities for their students [S4] and [S6], including interactive language game partners [S11].

5.2 Improving productivity and efficiency

The reviewed studies showed that participants' productivity and work efficiency have been significantly enhanced by using ChatGPT as it enabled them to allocate more time to other tasks and reduce their overall workloads [S6], [S10], [S11], [S13], and [S14]. However, three studies [S1], [S4], and [S11], indicated a negative perception and attitude among teachers toward using ChatGPT. This negativity stemmed from a lack of necessary skills to use it effectively [S1], a limited familiarity with it [S4], and occasional inaccuracies in the content provided by it [S10].

5.3 Catalyzing new teaching methodologies

Five of the reviewed studies highlighted that educators found the necessity of redefining their teaching profession with the assistance of ChatGPT [S11], developing new effective learning strategies [S4], and adapting teaching strategies and methodologies to ensure the development of essential skills for future engineers [S5]. They also emphasized the importance of adopting new educational philosophies and approaches that can evolve with the introduction of ChatGPT into the classroom [S12]. Furthermore, updating curricula to focus on improving human-specific features, such as emotional intelligence, creativity, and philosophical perspectives [S13], was found to be essential.

5.4 Effective utilization of CHATGPT in teaching

According to the reviewed studies, effective utilization of ChatGPT in education requires providing teachers with well-structured training, support, and adequate background on how to use ChatGPT responsibly [S1], [S3], [S11], and [S12]. Establishing clear rules and regulations regarding its usage is essential to ensure it positively impacts the teaching and learning processes, including students' skills [S1], [S4], [S5], [S8], [S9], and [S11]-[S14]. Moreover, conducting further research and engaging in discussions with policymakers and stakeholders is indeed crucial for the successful integration of ChatGPT in education and to maximize the benefits for both educators and students [S1], [S6]-[S10], and [S12]-[S14].

6 Discussion

The purpose of this review is to conduct a systematic review of empirical studies that have explored the utilization of ChatGPT, one of today’s most advanced LLM-based chatbots, in education. The findings of the reviewed studies showed several ways of ChatGPT utilization in different learning and teaching practices as well as it provided insights and considerations that can facilitate its effective and responsible use in future educational contexts. The results of the reviewed studies came from diverse fields of education, which helped us avoid a biased review that is limited to a specific field. Similarly, the reviewed studies have been conducted across different geographic regions. This kind of variety in geographic representation enriched the findings of this review.

In response to RQ1 , "What are students' and teachers' initial attempts at utilizing ChatGPT in education?", the findings from this review provide comprehensive insights. Chatbots, including ChatGPT, play a crucial role in supporting student learning, enhancing their learning experiences, and facilitating diverse learning approaches [ 42 , 43 ]. This review found that this tool, ChatGPT, has been instrumental in enhancing students' learning experiences by serving as a virtual intelligent assistant, providing immediate feedback, on-demand answers, and engaging in educational conversations. Additionally, students have benefited from ChatGPT’s ability to generate ideas, compose essays, and perform tasks like summarizing, translating, paraphrasing texts, or checking grammar, thereby enhancing their writing and language competencies. Furthermore, students have turned to ChatGPT for assistance in understanding concepts and homework, providing structured learning plans, and clarifying assignments and tasks, which fosters a supportive home learning environment, allowing them to take responsibility for their own learning and cultivate the skills and approaches essential for supportive home learning environment [ 26 , 27 , 28 ]. This finding aligns with the study of Saqr et al. [ 68 , 69 ] who highlighted that, when students actively engage in their own learning process, it yields additional advantages, such as heightened motivation, enhanced achievement, and the cultivation of enthusiasm, turning them into advocates for their own learning.

Moreover, students have utilized ChatGPT for tailored teaching and step-by-step guidance on diverse educational topics, streamlining task and project completion, and generating and recommending educational content. This personalization enhances the learning environment, leading to increased academic success. This finding aligns with other recent studies [ 26 , 27 , 28 , 60 , 66 ] which revealed that ChatGPT has the potential to offer personalized learning experiences and support an effective learning process by providing students with customized feedback and explanations tailored to their needs and abilities. Ultimately, fostering students' performance, engagement, and motivation, leading to increase students' academic success [ 14 , 44 , 58 ]. This ultimate outcome is in line with the findings of Saqr et al. [ 68 , 69 ], which emphasized that learning strategies are important catalysts of students' learning, as students who utilize effective learning strategies are more likely to have better academic achievement.

Teachers, too, have capitalized on ChatGPT's capabilities to enhance productivity and efficiency, using it for creating lesson plans, generating quizzes, providing additional resources, generating and preplanning new ideas for activities, and aiding in answering students’ questions. This adoption of technology introduces new opportunities to support teaching and learning practices, enhancing teacher productivity. This finding aligns with those of Day [ 17 ], De Castro [ 18 ], and Su and Yang [ 74 ] as well as with those of Valtonen et al. [ 82 ], who revealed that emerging technological advancements have opened up novel opportunities and means to support teaching and learning practices, and enhance teachers’ productivity.

In response to RQ2 , "What are the main findings derived from empirical studies that have incorporated ChatGPT into learning and teaching?", the findings from this review provide profound insights and raise significant concerns. Starting with the insights, chatbots, including ChatGPT, have demonstrated the potential to reshape and revolutionize education, creating new, novel opportunities for enhancing the learning process and outcomes [ 83 ], facilitating different learning approaches, and offering a range of pedagogical benefits [ 19 , 43 , 72 ]. In this context, this review found that ChatGPT could open avenues for educators to adopt or develop new effective learning and teaching strategies that can evolve with the introduction of ChatGPT into the classroom. Nonetheless, there is an evident lack of research understanding regarding the potential impact of generative machine learning models within diverse educational settings [ 83 ]. This necessitates teachers to attain a high level of proficiency in incorporating chatbots, such as ChatGPT, into their classrooms to create inventive, well-structured, and captivating learning strategies. In the same vein, the review also found that teachers without the requisite skills to utilize ChatGPT realized that it did not contribute positively to their work and could potentially have adverse effects [ 37 ]. This concern could lead to inequity of access to the benefits of chatbots, including ChatGPT, as individuals who lack the necessary expertise may not be able to harness their full potential, resulting in disparities in educational outcomes and opportunities. Therefore, immediate action is needed to address these potential issues. A potential solution is offering training, support, and competency development for teachers to ensure that all of them can leverage chatbots, including ChatGPT, effectively and equitably in their educational practices [ 5 , 28 , 80 ], which could enhance accessibility and inclusivity, and potentially result in innovative outcomes [ 82 , 83 ].

Additionally, chatbots, including ChatGPT, have the potential to significantly impact students' thinking abilities, including retention, reasoning, analysis skills [ 19 , 45 ], and foster innovation and creativity capabilities [ 83 ]. This review found that ChatGPT could contribute to improving a wide range of skills among students. However, it found that frequent use of ChatGPT may result in a decrease in innovative capacities, collaborative skills and cognitive capacities, and students' motivation to attend classes, as well as could lead to reduced higher-order thinking skills among students [ 22 , 29 ]. Therefore, immediate action is needed to carefully examine the long-term impact of chatbots such as ChatGPT, on learning outcomes as well as to explore its incorporation into educational settings as a supportive tool without compromising students' cognitive development and critical thinking abilities. In the same vein, the review also found that it is challenging to draw a consistent conclusion regarding the potential of ChatGPT to aid self-directed learning approach. This finding aligns with the recent study of Baskara [ 8 ]. Therefore, further research is needed to explore the potential of ChatGPT for self-directed learning. One potential solution involves utilizing learning analytics as a novel approach to examine various aspects of students' learning and support them in their individual endeavors [ 32 ]. This approach can bridge this gap by facilitating an in-depth analysis of how learners engage with ChatGPT, identifying trends in self-directed learning behavior, and assessing its influence on their outcomes.

Turning to the significant concerns, on the other hand, a fundamental challenge with LLM-based chatbots, including ChatGPT, is the accuracy and quality of the provided information and responses, as they provide false information as truth—a phenomenon often referred to as "hallucination" [ 3 , 49 ]. In this context, this review found that the provided information was not entirely satisfactory. Consequently, the utilization of chatbots presents potential concerns, such as generating and providing inaccurate or misleading information, especially for students who utilize it to support their learning. This finding aligns with other findings [ 6 , 30 , 35 , 40 ] which revealed that incorporating chatbots such as ChatGPT, into education presents challenges related to its accuracy and reliability due to its training on a large corpus of data, which may contain inaccuracies and the way users formulate or ask ChatGPT. Therefore, immediate action is needed to address these potential issues. One possible solution is to equip students with the necessary skills and competencies, which include a background understanding of how to use it effectively and the ability to assess and evaluate the information it generates, as the accuracy and the quality of the provided information depend on the input, its complexity, the topic, and the relevance of its training data [ 28 , 49 , 86 ]. However, it's also essential to examine how learners can be educated about how these models operate, the data used in their training, and how to recognize their limitations, challenges, and issues [ 79 ].

Furthermore, chatbots present a substantial challenge concerning maintaining academic integrity [ 20 , 56 ] and copyright violations [ 83 ], which are significant concerns in education. The review found that the potential misuse of ChatGPT might foster cheating, facilitate plagiarism, and threaten academic integrity. This issue is also affirmed by the research conducted by Basic et al. [ 7 ], who presented evidence that students who utilized ChatGPT in their writing assignments had more plagiarism cases than those who did not. These findings align with the conclusions drawn by Cotton et al. [ 13 ], Hisan and Amri [ 33 ] and Sullivan et al. [ 75 ], who revealed that the integration of chatbots such as ChatGPT into education poses a significant challenge to the preservation of academic integrity. Moreover, chatbots, including ChatGPT, have increased the difficulty in identifying plagiarism [ 47 , 67 , 76 ]. The findings from previous studies [ 1 , 84 ] indicate that AI-generated text often went undetected by plagiarism software, such as Turnitin. However, Turnitin and other similar plagiarism detection tools, such as ZeroGPT, GPTZero, and Copyleaks, have since evolved, incorporating enhanced techniques to detect AI-generated text, despite the possibility of false positives, as noted in different studies that have found these tools still not yet fully ready to accurately and reliably identify AI-generated text [ 10 , 51 ], and new novel detection methods may need to be created and implemented for AI-generated text detection [ 4 ]. This potential issue could lead to another concern, which is the difficulty of accurately evaluating student performance when they utilize chatbots such as ChatGPT assistance in their assignments. Consequently, the most LLM-driven chatbots present a substantial challenge to traditional assessments [ 64 ]. The findings from previous studies indicate the importance of rethinking, improving, and redesigning innovative assessment methods in the era of chatbots [ 14 , 20 , 64 , 75 ]. These methods should prioritize the process of evaluating students' ability to apply knowledge to complex cases and demonstrate comprehension, rather than solely focusing on the final product for assessment. Therefore, immediate action is needed to address these potential issues. One possible solution would be the development of clear guidelines, regulatory policies, and pedagogical guidance. These measures would help regulate the proper and ethical utilization of chatbots, such as ChatGPT, and must be established before their introduction to students [ 35 , 38 , 39 , 41 , 89 ].

In summary, our review has delved into the utilization of ChatGPT, a prominent example of chatbots, in education, addressing the question of how ChatGPT has been utilized in education. However, there remain significant gaps, which necessitate further research to shed light on this area.

7 Conclusions

This systematic review has shed light on the varied initial attempts at incorporating ChatGPT into education by both learners and educators, while also offering insights and considerations that can facilitate its effective and responsible use in future educational contexts. From the analysis of 14 selected studies, the review revealed the dual-edged impact of ChatGPT in educational settings. On the positive side, ChatGPT significantly aided the learning process in various ways. Learners have used it as a virtual intelligent assistant, benefiting from its ability to provide immediate feedback, on-demand answers, and easy access to educational resources. Additionally, it was clear that learners have used it to enhance their writing and language skills, engaging in practices such as generating ideas, composing essays, and performing tasks like summarizing, translating, paraphrasing texts, or checking grammar. Importantly, other learners have utilized it in supporting and facilitating their directed and personalized learning on a broad range of educational topics, assisting in understanding concepts and homework, providing structured learning plans, and clarifying assignments and tasks. Educators, on the other hand, found ChatGPT beneficial for enhancing productivity and efficiency. They used it for creating lesson plans, generating quizzes, providing additional resources, and answers learners' questions, which saved time and allowed for more dynamic and engaging teaching strategies and methodologies.

However, the review also pointed out negative impacts. The results revealed that overuse of ChatGPT could decrease innovative capacities and collaborative learning among learners. Specifically, relying too much on ChatGPT for quick answers can inhibit learners' critical thinking and problem-solving skills. Learners might not engage deeply with the material or consider multiple solutions to a problem. This tendency was particularly evident in group projects, where learners preferred consulting ChatGPT individually for solutions over brainstorming and collaborating with peers, which negatively affected their teamwork abilities. On a broader level, integrating ChatGPT into education has also raised several concerns, including the potential for providing inaccurate or misleading information, issues of inequity in access, challenges related to academic integrity, and the possibility of misusing the technology.

Accordingly, this review emphasizes the urgency of developing clear rules, policies, and regulations to ensure ChatGPT's effective and responsible use in educational settings, alongside other chatbots, by both learners and educators. This requires providing well-structured training to educate them on responsible usage and understanding its limitations, along with offering sufficient background information. Moreover, it highlights the importance of rethinking, improving, and redesigning innovative teaching and assessment methods in the era of ChatGPT. Furthermore, conducting further research and engaging in discussions with policymakers and stakeholders are essential steps to maximize the benefits for both educators and learners and ensure academic integrity.

It is important to acknowledge that this review has certain limitations. Firstly, the limited inclusion of reviewed studies can be attributed to several reasons, including the novelty of the technology, as new technologies often face initial skepticism and cautious adoption; the lack of clear guidelines or best practices for leveraging this technology for educational purposes; and institutional or governmental policies affecting the utilization of this technology in educational contexts. These factors, in turn, have affected the number of studies available for review. Secondly, the utilization of the original version of ChatGPT, based on GPT-3 or GPT-3.5, implies that new studies utilizing the updated version, GPT-4 may lead to different findings. Therefore, conducting follow-up systematic reviews is essential once more empirical studies on ChatGPT are published. Additionally, long-term studies are necessary to thoroughly examine and assess the impact of ChatGPT on various educational practices.

Despite these limitations, this systematic review has highlighted the transformative potential of ChatGPT in education, revealing its diverse utilization by learners and educators alike and summarized the benefits of incorporating it into education, as well as the forefront critical concerns and challenges that must be addressed to facilitate its effective and responsible use in future educational contexts. This review could serve as an insightful resource for practitioners who seek to integrate ChatGPT into education and stimulate further research in the field.

Data availability

The data supporting our findings are available upon request.

Abbreviations

• Artificial intelligence

AI in education

Large language model

Artificial neural networks

Chat Generative Pre-Trained Transformer

Recurrent neural networks

Long short-term memory

Reinforcement learning from human feedback

Natural language processing

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

AlAfnan MA, Dishari S, Jovic M, Lomidze K. ChatGPT as an educational tool: opportunities, challenges, and recommendations for communication, business writing, and composition courses. J Artif Intell Technol. 2023. https://doi.org/10.37965/jait.2023.0184 .

Article   Google Scholar  

Ali JKM, Shamsan MAA, Hezam TA, Mohammed AAQ. Impact of ChatGPT on learning motivation. J Engl Stud Arabia Felix. 2023;2(1):41–9. https://doi.org/10.56540/jesaf.v2i1.51 .

Alkaissi H, McFarlane SI. Artificial hallucinations in ChatGPT: implications in scientific writing. Cureus. 2023. https://doi.org/10.7759/cureus.35179 .

Anderson N, Belavý DL, Perle SM, Hendricks S, Hespanhol L, Verhagen E, Memon AR. AI did not write this manuscript, or did it? Can we trick the AI text detector into generated texts? The potential future of ChatGPT and AI in sports & exercise medicine manuscript generation. BMJ Open Sport Exerc Med. 2023;9(1): e001568. https://doi.org/10.1136/bmjsem-2023-001568 .

Ausat AMA, Massang B, Efendi M, Nofirman N, Riady Y. Can chat GPT replace the role of the teacher in the classroom: a fundamental analysis. J Educ. 2023;5(4):16100–6.

Google Scholar  

Baidoo-Anu D, Ansah L. Education in the Era of generative artificial intelligence (AI): understanding the potential benefits of ChatGPT in promoting teaching and learning. Soc Sci Res Netw. 2023. https://doi.org/10.2139/ssrn.4337484 .

Basic Z, Banovac A, Kruzic I, Jerkovic I. Better by you, better than me, chatgpt3 as writing assistance in students essays. 2023. arXiv preprint arXiv:2302.04536 .‏

Baskara FR. The promises and pitfalls of using chat GPT for self-determined learning in higher education: an argumentative review. Prosiding Seminar Nasional Fakultas Tarbiyah dan Ilmu Keguruan IAIM Sinjai. 2023;2:95–101. https://doi.org/10.47435/sentikjar.v2i0.1825 .

Behera RK, Bala PK, Dhir A. The emerging role of cognitive computing in healthcare: a systematic literature review. Int J Med Inform. 2019;129:154–66. https://doi.org/10.1016/j.ijmedinf.2019.04.024 .

Chaka C. Detecting AI content in responses generated by ChatGPT, YouChat, and Chatsonic: the case of five AI content detection tools. J Appl Learn Teach. 2023. https://doi.org/10.37074/jalt.2023.6.2.12 .

Chiu TKF, Xia Q, Zhou X, Chai CS, Cheng M. Systematic literature review on opportunities, challenges, and future research recommendations of artificial intelligence in education. Comput Educ Artif Intell. 2023;4:100118. https://doi.org/10.1016/j.caeai.2022.100118 .

Choi EPH, Lee JJ, Ho M, Kwok JYY, Lok KYW. Chatting or cheating? The impacts of ChatGPT and other artificial intelligence language models on nurse education. Nurse Educ Today. 2023;125:105796. https://doi.org/10.1016/j.nedt.2023.105796 .

Cotton D, Cotton PA, Shipway JR. Chatting and cheating: ensuring academic integrity in the era of ChatGPT. Innov Educ Teach Int. 2023. https://doi.org/10.1080/14703297.2023.2190148 .

Crawford J, Cowling M, Allen K. Leadership is needed for ethical ChatGPT: Character, assessment, and learning using artificial intelligence (AI). J Univ Teach Learn Pract. 2023. https://doi.org/10.53761/1.20.3.02 .

Creswell JW. Educational research: planning, conducting, and evaluating quantitative and qualitative research [Ebook]. 4th ed. London: Pearson Education; 2015.

Curry D. ChatGPT Revenue and Usage Statistics (2023)—Business of Apps. 2023. https://www.businessofapps.com/data/chatgpt-statistics/

Day T. A preliminary investigation of fake peer-reviewed citations and references generated by ChatGPT. Prof Geogr. 2023. https://doi.org/10.1080/00330124.2023.2190373 .

De Castro CA. A Discussion about the Impact of ChatGPT in education: benefits and concerns. J Bus Theor Pract. 2023;11(2):p28. https://doi.org/10.22158/jbtp.v11n2p28 .

Deng X, Yu Z. A meta-analysis and systematic review of the effect of Chatbot technology use in sustainable education. Sustainability. 2023;15(4):2940. https://doi.org/10.3390/su15042940 .

Eke DO. ChatGPT and the rise of generative AI: threat to academic integrity? J Responsib Technol. 2023;13:100060. https://doi.org/10.1016/j.jrt.2023.100060 .

Elmoazen R, Saqr M, Tedre M, Hirsto L. A systematic literature review of empirical research on epistemic network analysis in education. IEEE Access. 2022;10:17330–48. https://doi.org/10.1109/access.2022.3149812 .

Farrokhnia M, Banihashem SK, Noroozi O, Wals AEJ. A SWOT analysis of ChatGPT: implications for educational practice and research. Innov Educ Teach Int. 2023. https://doi.org/10.1080/14703297.2023.2195846 .

Fergus S, Botha M, Ostovar M. Evaluating academic answers generated using ChatGPT. J Chem Educ. 2023;100(4):1672–5. https://doi.org/10.1021/acs.jchemed.3c00087 .

Fink A. Conducting research literature reviews: from the Internet to Paper. Incorporated: SAGE Publications; 2010.

Firaina R, Sulisworo D. Exploring the usage of ChatGPT in higher education: frequency and impact on productivity. Buletin Edukasi Indonesia (BEI). 2023;2(01):39–46. https://doi.org/10.56741/bei.v2i01.310 .

Firat, M. (2023). How chat GPT can transform autodidactic experiences and open education.  Department of Distance Education, Open Education Faculty, Anadolu Unive .‏ https://orcid.org/0000-0001-8707-5918

Firat M. What ChatGPT means for universities: perceptions of scholars and students. J Appl Learn Teach. 2023. https://doi.org/10.37074/jalt.2023.6.1.22 .

Fuchs K. Exploring the opportunities and challenges of NLP models in higher education: is Chat GPT a blessing or a curse? Front Educ. 2023. https://doi.org/10.3389/feduc.2023.1166682 .

García-Peñalvo FJ. La percepción de la inteligencia artificial en contextos educativos tras el lanzamiento de ChatGPT: disrupción o pánico. Educ Knowl Soc. 2023;24: e31279. https://doi.org/10.14201/eks.31279 .

Gilson A, Safranek CW, Huang T, Socrates V, Chi L, Taylor A, Chartash D. How does ChatGPT perform on the United States medical Licensing examination? The implications of large language models for medical education and knowledge assessment. JMIR Med Educ. 2023;9: e45312. https://doi.org/10.2196/45312 .

Hashana AJ, Brundha P, Ayoobkhan MUA, Fazila S. Deep Learning in ChatGPT—A Survey. In   2023 7th international conference on trends in electronics and informatics (ICOEI) . 2023. (pp. 1001–1005). IEEE. https://doi.org/10.1109/icoei56765.2023.10125852

Hirsto L, Saqr M, López-Pernas S, Valtonen T. (2022). A systematic narrative review of learning analytics research in K-12 and schools.  Proceedings . https://ceur-ws.org/Vol-3383/FLAIEC22_paper_9536.pdf

Hisan UK, Amri MM. ChatGPT and medical education: a double-edged sword. J Pedag Educ Sci. 2023;2(01):71–89. https://doi.org/10.13140/RG.2.2.31280.23043/1 .

Hopkins AM, Logan JM, Kichenadasse G, Sorich MJ. Artificial intelligence chatbots will revolutionize how cancer patients access information: ChatGPT represents a paradigm-shift. JNCI Cancer Spectr. 2023. https://doi.org/10.1093/jncics/pkad010 .

Househ M, AlSaad R, Alhuwail D, Ahmed A, Healy MG, Latifi S, Sheikh J. Large Language models in medical education: opportunities, challenges, and future directions. JMIR Med Educ. 2023;9: e48291. https://doi.org/10.2196/48291 .

Ilkka T. The impact of artificial intelligence on learning, teaching, and education. Minist de Educ. 2018. https://doi.org/10.2760/12297 .

Iqbal N, Ahmed H, Azhar KA. Exploring teachers’ attitudes towards using CHATGPT. Globa J Manag Adm Sci. 2022;3(4):97–111. https://doi.org/10.46568/gjmas.v3i4.163 .

Irfan M, Murray L, Ali S. Integration of Artificial intelligence in academia: a case study of critical teaching and learning in Higher education. Globa Soc Sci Rev. 2023;8(1):352–64. https://doi.org/10.31703/gssr.2023(viii-i).32 .

Jeon JH, Lee S. Large language models in education: a focus on the complementary relationship between human teachers and ChatGPT. Educ Inf Technol. 2023. https://doi.org/10.1007/s10639-023-11834-1 .

Khan RA, Jawaid M, Khan AR, Sajjad M. ChatGPT—Reshaping medical education and clinical management. Pak J Med Sci. 2023. https://doi.org/10.12669/pjms.39.2.7653 .

King MR. A conversation on artificial intelligence, Chatbots, and plagiarism in higher education. Cell Mol Bioeng. 2023;16(1):1–2. https://doi.org/10.1007/s12195-022-00754-8 .

Kooli C. Chatbots in education and research: a critical examination of ethical implications and solutions. Sustainability. 2023;15(7):5614. https://doi.org/10.3390/su15075614 .

Kuhail MA, Alturki N, Alramlawi S, Alhejori K. Interacting with educational chatbots: a systematic review. Educ Inf Technol. 2022;28(1):973–1018. https://doi.org/10.1007/s10639-022-11177-3 .

Lee H. The rise of ChatGPT: exploring its potential in medical education. Anat Sci Educ. 2023. https://doi.org/10.1002/ase.2270 .

Li L, Subbareddy R, Raghavendra CG. AI intelligence Chatbot to improve students learning in the higher education platform. J Interconnect Netw. 2022. https://doi.org/10.1142/s0219265921430325 .

Limna P. A Review of Artificial Intelligence (AI) in Education during the Digital Era. 2022. https://ssrn.com/abstract=4160798

Lo CK. What is the impact of ChatGPT on education? A rapid review of the literature. Educ Sci. 2023;13(4):410. https://doi.org/10.3390/educsci13040410 .

Luo W, He H, Liu J, Berson IR, Berson MJ, Zhou Y, Li H. Aladdin’s genie or pandora’s box For early childhood education? Experts chat on the roles, challenges, and developments of ChatGPT. Early Educ Dev. 2023. https://doi.org/10.1080/10409289.2023.2214181 .

Meyer JG, Urbanowicz RJ, Martin P, O’Connor K, Li R, Peng P, Moore JH. ChatGPT and large language models in academia: opportunities and challenges. Biodata Min. 2023. https://doi.org/10.1186/s13040-023-00339-9 .

Mhlanga D. Open AI in education, the responsible and ethical use of ChatGPT towards lifelong learning. Soc Sci Res Netw. 2023. https://doi.org/10.2139/ssrn.4354422 .

Neumann, M., Rauschenberger, M., & Schön, E. M. (2023). “We Need To Talk About ChatGPT”: The Future of AI and Higher Education.‏ https://doi.org/10.1109/seeng59157.2023.00010

Nolan B. Here are the schools and colleges that have banned the use of ChatGPT over plagiarism and misinformation fears. Business Insider . 2023. https://www.businessinsider.com

O’Leary DE. An analysis of three chatbots: BlenderBot, ChatGPT and LaMDA. Int J Intell Syst Account, Financ Manag. 2023;30(1):41–54. https://doi.org/10.1002/isaf.1531 .

Okoli C. A guide to conducting a standalone systematic literature review. Commun Assoc Inf Syst. 2015. https://doi.org/10.17705/1cais.03743 .

OpenAI. (2023). https://openai.com/blog/chatgpt

Perkins M. Academic integrity considerations of AI large language models in the post-pandemic era: ChatGPT and beyond. J Univ Teach Learn Pract. 2023. https://doi.org/10.53761/1.20.02.07 .

Plevris V, Papazafeiropoulos G, Rios AJ. Chatbots put to the test in math and logic problems: A preliminary comparison and assessment of ChatGPT-3.5, ChatGPT-4, and Google Bard. arXiv (Cornell University) . 2023. https://doi.org/10.48550/arxiv.2305.18618

Rahman MM, Watanobe Y (2023) ChatGPT for education and research: opportunities, threats, and strategies. Appl Sci 13(9):5783. https://doi.org/10.3390/app13095783

Ram B, Verma P. Artificial intelligence AI-based Chatbot study of ChatGPT, google AI bard and baidu AI. World J Adv Eng Technol Sci. 2023;8(1):258–61. https://doi.org/10.30574/wjaets.2023.8.1.0045 .

Rasul T, Nair S, Kalendra D, Robin M, de Oliveira Santini F, Ladeira WJ, Heathcote L. The role of ChatGPT in higher education: benefits, challenges, and future research directions. J Appl Learn Teach. 2023. https://doi.org/10.37074/jalt.2023.6.1.29 .

Ratnam M, Sharm B, Tomer A. ChatGPT: educational artificial intelligence. Int J Adv Trends Comput Sci Eng. 2023;12(2):84–91. https://doi.org/10.30534/ijatcse/2023/091222023 .

Ray PP. ChatGPT: a comprehensive review on background, applications, key challenges, bias, ethics, limitations and future scope. Internet Things Cyber-Phys Syst. 2023;3:121–54. https://doi.org/10.1016/j.iotcps.2023.04.003 .

Roumeliotis KI, Tselikas ND. ChatGPT and Open-AI models: a preliminary review. Future Internet. 2023;15(6):192. https://doi.org/10.3390/fi15060192 .

Rudolph J, Tan S, Tan S. War of the chatbots: Bard, Bing Chat, ChatGPT, Ernie and beyond. The new AI gold rush and its impact on higher education. J Appl Learn Teach. 2023. https://doi.org/10.37074/jalt.2023.6.1.23 .

Ruiz LMS, Moll-López S, Nuñez-Pérez A, Moraño J, Vega-Fleitas E. ChatGPT challenges blended learning methodologies in engineering education: a case study in mathematics. Appl Sci. 2023;13(10):6039. https://doi.org/10.3390/app13106039 .

Sallam M, Salim NA, Barakat M, Al-Tammemi AB. ChatGPT applications in medical, dental, pharmacy, and public health education: a descriptive study highlighting the advantages and limitations. Narra J. 2023;3(1): e103. https://doi.org/10.52225/narra.v3i1.103 .

Salvagno M, Taccone FS, Gerli AG. Can artificial intelligence help for scientific writing? Crit Care. 2023. https://doi.org/10.1186/s13054-023-04380-2 .

Saqr M, López-Pernas S, Helske S, Hrastinski S. The longitudinal association between engagement and achievement varies by time, students’ profiles, and achievement state: a full program study. Comput Educ. 2023;199:104787. https://doi.org/10.1016/j.compedu.2023.104787 .

Saqr M, Matcha W, Uzir N, Jovanović J, Gašević D, López-Pernas S. Transferring effective learning strategies across learning contexts matters: a study in problem-based learning. Australas J Educ Technol. 2023;39(3):9.

Schöbel S, Schmitt A, Benner D, Saqr M, Janson A, Leimeister JM. Charting the evolution and future of conversational agents: a research agenda along five waves and new frontiers. Inf Syst Front. 2023. https://doi.org/10.1007/s10796-023-10375-9 .

Shoufan A. Exploring students’ perceptions of CHATGPT: thematic analysis and follow-up survey. IEEE Access. 2023. https://doi.org/10.1109/access.2023.3268224 .

Sonderegger S, Seufert S. Chatbot-mediated learning: conceptual framework for the design of Chatbot use cases in education. Gallen: Institute for Educational Management and Technologies, University of St; 2022. https://doi.org/10.5220/0010999200003182 .

Book   Google Scholar  

Strzelecki A. To use or not to use ChatGPT in higher education? A study of students’ acceptance and use of technology. Interact Learn Environ. 2023. https://doi.org/10.1080/10494820.2023.2209881 .

Su J, Yang W. Unlocking the power of ChatGPT: a framework for applying generative AI in education. ECNU Rev Educ. 2023. https://doi.org/10.1177/20965311231168423 .

Sullivan M, Kelly A, McLaughlan P. ChatGPT in higher education: Considerations for academic integrity and student learning. J ApplLearn Teach. 2023;6(1):1–10. https://doi.org/10.37074/jalt.2023.6.1.17 .

Szabo A. ChatGPT is a breakthrough in science and education but fails a test in sports and exercise psychology. Balt J Sport Health Sci. 2023;1(128):25–40. https://doi.org/10.33607/bjshs.v127i4.1233 .

Taecharungroj V. “What can ChatGPT do?” analyzing early reactions to the innovative AI chatbot on Twitter. Big Data Cognit Comput. 2023;7(1):35. https://doi.org/10.3390/bdcc7010035 .

Tam S, Said RB. User preferences for ChatGPT-powered conversational interfaces versus traditional methods. Biomed Eng Soc. 2023. https://doi.org/10.58496/mjcsc/2023/004 .

Tedre M, Kahila J, Vartiainen H. (2023). Exploration on how co-designing with AI facilitates critical evaluation of ethics of AI in craft education. In: Langran E, Christensen P, Sanson J (Eds).  Proceedings of Society for Information Technology and Teacher Education International Conference . 2023. pp. 2289–2296.

Tlili A, Shehata B, Adarkwah MA, Bozkurt A, Hickey DT, Huang R, Agyemang B. What if the devil is my guardian angel: ChatGPT as a case study of using chatbots in education. Smart Learn Environ. 2023. https://doi.org/10.1186/s40561-023-00237-x .

Uddin SMJ, Albert A, Ovid A, Alsharef A. Leveraging CHATGPT to aid construction hazard recognition and support safety education and training. Sustainability. 2023;15(9):7121. https://doi.org/10.3390/su15097121 .

Valtonen T, López-Pernas S, Saqr M, Vartiainen H, Sointu E, Tedre M. The nature and building blocks of educational technology research. Comput Hum Behav. 2022;128:107123. https://doi.org/10.1016/j.chb.2021.107123 .

Vartiainen H, Tedre M. Using artificial intelligence in craft education: crafting with text-to-image generative models. Digit Creat. 2023;34(1):1–21. https://doi.org/10.1080/14626268.2023.2174557 .

Ventayen RJM. OpenAI ChatGPT generated results: similarity index of artificial intelligence-based contents. Soc Sci Res Netw. 2023. https://doi.org/10.2139/ssrn.4332664 .

Wagner MW, Ertl-Wagner BB. Accuracy of information and references using ChatGPT-3 for retrieval of clinical radiological information. Can Assoc Radiol J. 2023. https://doi.org/10.1177/08465371231171125 .

Wardat Y, Tashtoush MA, AlAli R, Jarrah AM. ChatGPT: a revolutionary tool for teaching and learning mathematics. Eurasia J Math, Sci Technol Educ. 2023;19(7):em2286. https://doi.org/10.29333/ejmste/13272 .

Webster J, Watson RT. Analyzing the past to prepare for the future: writing a literature review. Manag Inf Syst Quart. 2002;26(2):3.

Xiao Y, Watson ME. Guidance on conducting a systematic literature review. J Plan Educ Res. 2017;39(1):93–112. https://doi.org/10.1177/0739456x17723971 .

Yan D. Impact of ChatGPT on learners in a L2 writing practicum: an exploratory investigation. Educ Inf Technol. 2023. https://doi.org/10.1007/s10639-023-11742-4 .

Yu H. Reflection on whether Chat GPT should be banned by academia from the perspective of education and teaching. Front Psychol. 2023;14:1181712. https://doi.org/10.3389/fpsyg.2023.1181712 .

Zhu C, Sun M, Luo J, Li T, Wang M. How to harness the potential of ChatGPT in education? Knowl Manag ELearn. 2023;15(2):133–52. https://doi.org/10.34105/j.kmel.2023.15.008 .

Download references

The paper is co-funded by the Academy of Finland (Suomen Akatemia) Research Council for Natural Sciences and Engineering for the project Towards precision education: Idiographic learning analytics (TOPEILA), Decision Number 350560.

Author information

Authors and affiliations.

School of Computing, University of Eastern Finland, 80100, Joensuu, Finland

Yazid Albadarin, Mohammed Saqr, Nicolas Pope & Markku Tukiainen

You can also search for this author in PubMed   Google Scholar

Contributions

YA contributed to the literature search, data analysis, discussion, and conclusion. Additionally, YA contributed to the manuscript’s writing, editing, and finalization. MS contributed to the study’s design, conceptualization, acquisition of funding, project administration, allocation of resources, supervision, validation, literature search, and analysis of results. Furthermore, MS contributed to the manuscript's writing, revising, and approving it in its finalized state. NP contributed to the results, and discussions, and provided supervision. NP also contributed to the writing process, revisions, and the final approval of the manuscript in its finalized state. MT contributed to the study's conceptualization, resource management, supervision, writing, revising the manuscript, and approving it.

Corresponding author

Correspondence to Yazid Albadarin .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

See Table  4

The process of synthesizing the data presented in Table  4 involved identifying the relevant studies through a search process of databases (ERIC, Scopus, Web of Knowledge, Dimensions.ai, and lens.org) using specific keywords "ChatGPT" and "education". Following this, inclusion/exclusion criteria were applied, and data extraction was performed using Creswell's [ 15 ] coding techniques to capture key information and identify common themes across the included studies.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Albadarin, Y., Saqr, M., Pope, N. et al. A systematic literature review of empirical research on ChatGPT in education. Discov Educ 3 , 60 (2024). https://doi.org/10.1007/s44217-024-00138-2

Download citation

Received : 22 October 2023

Accepted : 10 May 2024

Published : 26 May 2024

DOI : https://doi.org/10.1007/s44217-024-00138-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Large language models
• Educational technology
• Systematic review

Advertisement

• Find a journal
• Publish with us
• Track your research

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Experiencing more meaningful coincidences is associated with more real-life creativity? Insights from three empirical studies

Roles Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Software, Validation, Visualization, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliation Department of Psychology, University of Graz, Graz, Austria

Roles Resources, Writing – review & editing

Roles Conceptualization, Supervision, Writing – review & editing

• Christian Rominger, 
• Andreas Fink, 
• Corinna M. Perchtold-Stefan

• Published: May 24, 2024
• https://doi.org/10.1371/journal.pone.0300121
• Reader Comments

Literature suggests a link between creativity and the perception of meaningful patterns in random arrangements, which is coined apophenia, patternicity, synchronicity, or the experience of meaningful coincidences. However, empirical research did not establish a clear link between real-life creativity and the experience of meaningful coincidences. In this three-study approach, we consistently found a connection between the experience of meaningful coincidences and creative activities as well as creative achievements. However, we did not obtain a consistent link with openness to experience or with peoples’ creative potential. By applying an internet daily diary approach, we found that the experience of meaningful coincidences fluctuates from day to day and that the number of perceived coincidences is associated with positive and negative affect. A third preregistered study showed that positive and negative affect might not serve as a strong mechanism that mediates the link between meaningful coincidences and real-life creative activities. We need further research to explore the reason for this robust link between meaningful coincidences and real-life creativity.

Citation: Rominger C, Fink A, Perchtold-Stefan CM (2024) Experiencing more meaningful coincidences is associated with more real-life creativity? Insights from three empirical studies. PLoS ONE 19(5): e0300121. https://doi.org/10.1371/journal.pone.0300121

Editor: Xinwen Bai, IPCAS: Institute of Psychology Chinese Academy of Sciences, CHINA

Received: September 20, 2023; Accepted: February 22, 2024; Published: May 24, 2024

Copyright: © 2024 Rominger et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Data affiliated with this manuscript are available via the Open Science Framework at the following link: https://osf.io/vgj6a/?view_only=44421ae04b6b4ce9873c2b9e7caa0cf5 ".

Funding: The author(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The experience of meaningful coincidences is defined as the sudden perception of a tight and significant connection of events which are, however, objectively unrelated, such as thinking about a friend and receiving a phone call from exactly this friend [ 1 ]. In these moments, we believe to experience that an inner event causes an outer event, and that thinking about the friend might have led to the call of the friend. This elicits a strong feeling that somehow, both events are related and take place in synchrony. People, show interindividual differences in the number of these and similar experiences [ 2 ]. A higher frequency of meaningful coincidences is associated with paranormal beliefs and positive schizotypy [ 3 , 4 ], but at the same time presents its own, distinct characteristics [ 2 , 5 ]. While linked to positive schizotypy and at the extreme, to psychiatric illness of schizophrenia, the experience of illusory connections and meaningful patterns between unrelated events, semantic concepts, and random arrangements (i.e., apophenia, [ 6 ]; but see [ 7 ] for a critical discussion of pareidolias as a surrogate for visual hallucinations) might also represent a benevolent and useful human trait, which is relevant for creativity [ 2 , 8 – 10 ].

Creativity can be understood as a potential (i.e., creative potential), a personality trait (i.e., openness), an activity (i.e., creative activities), as well as an achievement (i.e., creative achievements; see e.g., [ 11 , 12 ]). The production of unique and original ideas, which are useful is at the core of all facets of creativity ([ 13 ]; but see e.g., [ 14 , 15 ] for an ongoing debate). Following this definition, seeing connections and meaningful patterns which are hidden from the view of others (cf . [ 16 ]) might help to overcome mental blocks, enhance thinking out of the box, and increase the chance to come up with more creative ideas. Therefore, the propensity to experience meaningful coincidences might represent a personality trait associated with a higher creative potential. In line with this, a meta-analysis indicated a positive relationship between creativity and positive aspects of schizotypy [ 17 ]. More specifically, several authors suggested a link between meaningful coincidences, and related phenomena such as synchronicity, patternicity, and apophenia with creativity [ 2 , 10 , 18 , 19 ]. However, there is only little direct empirical evidence available indicating a connection between creativity and the propensity to experience meaningful coincidences.

What has been done so far? Rominger et al. [ 5 ] reported a positive association between the self-rated propensity to experience meaningful coincidences and self-rated creative ideation behavior (see also [ 20 ]). In accordance with this, Diana et al. [ 21 ] found an association between the production of original ideas and the detection of meanings in pictures of natural landscapes. Rominger et al. [ 22 ] found that participants with a higher creative potential perceived a higher number of patterns in randomly arranged stimuli in a figural association task. Furthermore, Russo-Netzer and Icekson [ 23 ] reported a link between openness and the experience of meaningful coincidences (see also [ 20 ]). To sum up, this pattern of findings adds to the assumption that the experience of meaningful coincidences shares variance with several relevant measures of creativity such as creative ideation behavior, creative potential, as well as openness as the personality trait linked with creativity [ 24 , 25 ].

However, there is a clear gap in literature and no study to date investigated if people who experience more meaningful coincidences are more often engaged in real-life creativity such as creative activities and creative achievements [ 11 ]. This lack of knowledge makes it necessary to conduct this three-study research—assessing a broad range of creativity indices (i.e., creative potential, openness) along with real-life creativity (i.e., creative activities and achievements). This multi-study approach served two main aims. First, we investigated if the experience of meaningful coincidences is robustly associated with real-life creativity (as well as creative potential and openness). Second, we evaluated if PA (and NA) could serve a potential reason for the link between real-life creativity [ 26 ] and meaningful coincidences [ 23 ]. To achieve the second goal and to study potential mechanisms, we went step-by-step (i.e., from study to study) from a trait perspective (i.e., between-person) to a state perspective (i.e., within-person). The state-of-the-art assessment of more ecologically valid and dynamic data via internet daily diary allows to investigate if affective states as well as affective traits (i.e., PA and NA) might mediate the link between the perception of meaningful coincidences and creative activities (as well as achievements).

Therefore, and as the first step, study 1 investigated if the experience of meaningful coincidences is associated with real-life creativity such as creative activities and creative achievements in different domains like literature, sports, and music [ 11 ]. Additionally, we assessed participants’ creative potential, openness, and creative ideation behavior in an online survey to explore creativity relevant associations with the propensity to experience meaningful coincidences.

Participants.

In total 69 participants (45 women) took part in the first study. The mean age was 23.78 years ( SD = 5.06, min = 18, max = 54). According to self-report, all participants were free of cardiovascular, neurological, or mental disorders as well as psychotropic or cardiovascular medication. Participants were recruited via email and social media. The study was approved by the ethics review board (GZ. 39/100/63 ex 2020/21) and did not take place in a specific context. All participants gave written informed consent to participate in the study by clicking an agree to participate button, which is part of a larger project (see e.g., [ 27 ]). The recruitment period of this study was from 01 August 2021 to 31 December 2021. We used Limesurvey for this (Limesurvey GmbH. / LimeSurvey: An Open Source survey tool /LimeSurvey GmbH, Hamburg, Germany. http://www.limesurvey.org ).

The propensity to experience meaningful coincidences.

On the German version of the Coincidence Questionnaire, participants rated how frequently they have experienced several categories of "meaningful" coincidences [ 2 , 3 , 5 , 28 ]. The 7 items (e.g., perception of something distant in time such as having a dream that comes true) are rated from "never" to "very often" (5-point Likert scale). The propensity to perceive meaningful coincidences is the sum of all items ( M = 17.91, SD = 4.62, min = 8, max = 28, α = .76).

Creative potential.

Participants completed the abstract picture fragments of the Test for Creative Thinking–Drawing Production online (TCT-DP; [ 29 ]). The time limit was 15 min and was monitored during the online meeting. The generated drawings were then sent to the experimenter. Two independent and trained raters (one man and one woman) scored the TCT-DP in accordance with the test manual (e.g., unconventionality, inclusion of new elements, graphic combinations, etc.). The mean score of both raters was used as index of creative potential ( M = 1.56, SD = 0.54) with a high interrater reliability ( r = .97).

Creative personality (RIBS).

Creative ideation behavior was assessed by a German version of Runco’s Ideational Behavior Scale (RIBS; [ 30 ]; see e.g., [ 11 ]), which includes 17 statements such as “I come up with an idea or solution other people have never thought of”. Participants responded to the items on a scale ranging from 1 (never) to 5 (very often; M = 64.59, SD = 14.80; α = .92).

Creative activities and creative achievements.

We assessed creative activities (CAct) and creative achievements (CAch) with the Inventory of Creative Activities and Achievements (ICAA; 11). The questionnaire asks for 8 different domains of creative activities and achievements (i.e., literature, music, arts and crafts, cooking, sports, visual arts, performing arts, and science and engineering). The creative activities and creative achievements sum scores showed good internal consistency of α = .81 and α = .70 respectively.

We assessed participants’ openness via the NEO-FFI ([ 31 ], German translation; [ 32 ]). Openness was consistently associated with creative ideation performance and the potential for open problem solving [ 25 , 33 ], as well as real-life creative activities and creative achievements [ 11 ]. The internal consistency of openness in the present study was α = .75 ( M = 34.48, SD = 6.51).

Statistical analysis.

We calculated Pearson correlations to analyze the association between the experience of meaningful coincidences and measures of creativity. To follow-up significant associations between meaningful coincidences and creative activities as well as creative achievements, we calculated Pearson correlations with each sub-score of the ICAA. We calculated all statistical analyses via SPSS (vers. 29; IBM SPSS Statistics, Armonk, NY, USA). The level of significance was p < .05 (two-tailed).

As illustrated in Table 1 , the experience of meaningful coincidences was significantly associated with creative activities ( r = .35, p = .003) but failed to reach significance for creative achievements ( r = .17, p = .174). The experience of meaningful coincidences was not associated with openness, creative ideation behavior, and participants’ creative potential (see Table 1 ). As illustrated in Table 1 , creative ideation behavior was significantly associated with creative activities and achievements, which just slightly failed to reach significance for the TCT-DP. Openness was associated with creative activities, creative achievements, and self-rated creative ideation. Creative activities and creative achievements were intercorrelated. This pattern of findings indicates convergent validity of assessment. Age was negatively associated with creative activities and showed a trend for a negative correlation with the experience of meaningful coincidences. When controlling for age, the association between meaningful coincidences and creative activities remained virtually unchanged ( rs = .310, p = .010). Men showed a higher creative potential than women.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0300121.t001

Follow-up analysis indicated that the experience of meaningful coincidences was significantly associated with all sub-scores of creative activities, except music and science and engineering (for details see S1 Table ). For creative achievements, the correlation coefficients ranged between -.096 and .210 and did not reach significance for any subscale (for details see S1 Table ).

Discussion study 1

In this study, we investigated if the experience of meaningful coincidences is associated with more real-life creativity, which goes beyond studies investigating the association with openness and self-rated creative ideation behavior [ 5 ]. In line with the main aim of the study, we found an association between the propensity to experience meaningful coincidences and creative activities such as creating an original talk, writing a blog entry, or making up a rhyme. Creative achievements (as the sum score of 8 different categories) were positively but not significantly associated with the experience of meaningful coincidences (cooking and performing arts showed the highest correlation-coefficients). Nevertheless, for the very first time, this pattern of findings indicates that people who experience more meaningful coincidences are more likely to engage in real-life creative activities. This finding was not driven by gender and age and fits the association between positive schizotypy and the engagement in creative behaviors [ 34 ].

This novel finding is important since previous work strongly focused on peoples’ creative potential [ 5 , 21 ]. The present study adds a more real-life creativity finding to literature. In line with previous work, the association with self-rated creative behavior and openness showed the expected effect sizes (although not significant). However, we found virtually no association between people’s creative potential assessed via the TCT-DP and the propensity to experience meaningful coincidences. This might indicate that the experience of meaningful coincidences is more strongly related to self-rated creative activities but may not be that strongly associated with performance measures of creativity and people’s creative potential. This seems to be in some contrast to the study of Diana et al. [ 21 ]. However, it is important to hold in mind that Diana et al. reported an association between two performance measures (i.e., shared methods variance). In their approach, the number of detected meanings in pictures of natural landscapes might capture a potential [ 21 ] but not the propensity to perceive meaning (in randomness). In the present study, however, we targeted to assess the propensity of meaningful coincidences and found associations with creative activities (and creative achievements).

Of note, creativity indices showed the expected pattern of findings and therefore indicated convergent validity of the assessment. People’s creative potential was related with creative achievements at a trend (see e.g., [ 35 ]). Furthermore, creative ideation behavior was associated with other relevant parameters of creativity such as openness, creative activities, as well as creative achievements. In line with literature creative activities were associated with age [ 11 ]. Since, the association between meaningful coincidences and real-life creativity constitutes a novel finding, we were interested if this finding could (conceptually) replicate in an independent study 2.

In study 1, we retrospectively assessed the frequency of meaningful coincidences via an established questionnaire and came up with an interesting finding, indicating an association between meaningful coincidences and real-life creativity. However, in study 1, participants rated their frequency of experiencing meaningful coincidences on a scale from “never” to “very often” throughout their entire life. Such a procedure is not a direct assessment of the phenomenon of interest, as people might strongly rely on subjective representations of the frequency of experiences as well as on their beliefs [ 36 , 37 ]. Therefore, this retrospective method might dilute study results and the shared methods variance between questionnaires (e.g., coincidence questionnaire and ICAA) might further lead to an overestimation of effect size of the observed association between the variables of interest.

In order to assess the frequency with which people experience meaningful coincidences when they actually take place and to further increase our trust in the results of study 1, we used an internet daily dairy method in study 2 [ 37 ]. The first aim of study 2 was to replicate the association between the propensity to experience meaningful coincidences and real-life creativity. We monitored people’s experience of meaningful coincidences throughout seven days of a week (at most). This procedure allows to assess the dynamic of experiencing meaningful coincidences in more real-time (on a day-to-day basis) and to differentiate between intraindividual and interindividual (i.e., within-person and between-person) differences [ 37 , 38 ]. By applying multilevel models to these micro-longitudinal data, we can differentiate between-person and within-person effects and study cross-sectional and dynamic effects of the experience of meaningful coincidences at the very same time. This is novel in literature, which to date only focused on the interindividual perspective of meaningful coincidences and synchronicity [ 2 , 3 , 23 , 39 – 41 ]. Therefore, the second aim of study 2 is to investigate the fluctuations in the experience of meaning in everyday life for the very first time. In addition to meaningful coincidences, we further assessed participants’ positive and negative affective states throughout the assessment period of a week. This allows to investigate if peoples’ daily affect covaries with the frequency of meaningful coincidences.

First, we can assume that positive affect (PA) constitutes a state of a more flexible and broadened thinking style [ 42 , 43 ], which may be ideal for experiencing meaning in random arrangements. Additionally, the sudden perception of meaningful coincidences might enhance peoples’ PA [ 23 ]. Both directions of the effect should go along with a positive association between PA and meaningful coincidences. Second, the perception of meaningful coincidences might serve as a coping strategy. This means that people, who perceive more negative affect (NA) might also experience more meaningful coincidences in order to cope with their negative feelings. In other words, NA affect might increase the “need” to experience meaning in meaningless noise to cope with (random) situations [ 23 ]. This means that detecting coincidences might help to find coherence and purpose in life, which might arise from a desire for control ambiguity [ 28 ]. This is in line with studies showing associations between the search for meaning and unusual perceptions [ 44 ]. Additionally, PA and NA are linked to creativity [ 26 ], however, via different pathways [ 43 ]. Therefore, we can assume that PA and NA might potentially connect meaningful coincidences with real-life creativity and vice versa. With other words, the affect (PA and NA) might mediate the association between meaningful coincidences and creativity.

To broaden the assessment of creativity, in addition to real-life creative achievements and activities, we again assessed peoples’ openness and their creative potential [ 45 ]. Based on the findings of study 1, we assumed a positive association between meaningful coincidences and real-life creativity, although we assessed meaningful coincidences from day to day–which allows the assessment of the trait level of coincidences as well as the dynamic of these experiences.

In total, 98 participants took part in the internet daily diary study. Only participants who delivered valid answers for at least three days were included in the final study sample. This resulted in a final sample of 93 participants (74 women) with a mean age of 29.17 years ( SD = 15.07). All participants gave written informed consent before participating in the online survey by clicking an agree to participate button. The local ethics committee proved this study (GZ. 39/82/63 ex 2014/15), did not take place in a specific context, and the recruitment period of this study was from 1 to 31 May 2022.

Daily diary method.

Participants received a daily email with a link to the online survey that included 8 questions assessing the experience of meaningful coincidences and the 20 items of the PANAS to assess their daily affect. We used Limesurvey for this (Limesurvey GmbH. / LimeSurvey: An Open Source survey tool /LimeSurvey GmbH, Hamburg, Germany. http://www.limesurvey.org ).

Experience of meaningful coincidences.

Participants were asked how often they had perceived meaningful coincidences during the day by means of 8 items. They had to report the number of meaningful experiences they have experienced throughout the day in 7 different categories (taken from the coincidence questionnaire). Furthermore, we assessed other types of coincidences with one additional item (i.e., How many coincidences of other domains did you experience today?). For all calculations, we transformed the number of coincidences per day into a 5-point ordinal scale from 0 to 4 (i.e., 4 and more experiences per day). The mean number of coincidences per day was M = 0.25 ( SD = 0.32) ranging from 0 to 2. The mean number of coincidences at the between-person level was M = 0.25 ( SD = 0.26) ranging from 0 to 1.41. By applying generalizability theory analysis (GTA; [ 46 ]; for an application see e.g., [ 47 ]), we indicated excellent reliability for the assessment of the number of coincidences at the between-person level (R kR = .92) and acceptable reliability at the within-person level with R C = .39 [ 48 ].

Furthermore, each day we asked participants to answer all 20 items of the German version of the Positive and Negative affect Schedule (PANAS; [ 49 ]). The mean PA per day was M = 2.79 ( SD = 0.80; NA: M = 1.54, SD = 0.56) ranging from 1 to 5 (NA: 1 to 4.1). The mean PA at the between-person level was M = 2.78 ( SD = 0.55; NA: M = 1.55, SD = 0.40) ranging from 1.35 to 4.50 (NA: 1 to 2.62). The reliability of the PA was good for between-person (R kR = .84) and for within-person (R C = .90). Similarly, NA was assessed reliable with R kR = .85 for between-person and R C = .82 for within-person variation.

Online assessment of trait variables

Creative activities (CAct) and creative achievements (CAch) were assessed by means of the Inventory of Creative Activities and Achievements (ICAA; [ 11 ]). This questionnaire asks for eight different domains of creative activities and achievements (i.e., literature, music, arts & crafts, cooking, sports, visual arts, performing arts, and science and engineering). The sum score of creative activities showed an internal consistency of α = .79 and creative achievements score showed a α of .73.

Coincidence questionnaire.

We used the same coincidence questionnaire as in study 1. The Cronbach alpha was α = .78. This questionnaire served as validity measure for the daily diary assessment of meaningful coincidences with M = 17.45 ( SD = 4.91).

We used the divergent association task (DAT; [ 45 ]) to assess peoples’ creative potential. Participants were instructed to produce 10 unrelated words within 4 minutes. The unrelatedness served as an indicator of people’s creative potential. To calculate the distance between the words, we translated the German words into English. For the translated words we calculated the distance measures by means of the provided online application [ 45 ]. The mean distance score was 80.37 ( SD = 5.10).

Openness was assessed via the German version of the Big Five Inventory 2 (BFI-2; [ 50 ]), which assesses three sub-facets of openness (i.e., intellectual curiosity, aesthetic sensitivity, creative imagination). Cronbach alpha of the total score was α = .84.

Before the daily diary assessment all participants filled in the online questionnaires and worked on an online version of the DAT (LimeSurvey GmbH, Hamburg, Germany). Then participants registered for a daily dairy assessment, where they received an email per day with a link to the daily survey. Participants should answer the daily diary each day between 6 pm and 12 pm.

Statistical analyses.

First, to evaluate the proportion of variance associated with between-person and within-person variance for PA, NA, and meaningful coincidences respectively, we applied generalizability theory analyses (GTA; [ 46 ]) by the use of the software psych (Version 2.3.3; [ 51 ]) running in R (Version 3.4.2; [ 52 ]). Following Netzlek [ 48 ], we differentiated between Level 1 (items), Level 2 (days), and Level 3 (person). GTA is especially suited to assess reliability of daily-diary data, allowing the partitioning of between-person, within-person, and error variance by decomposing the observed variance associated with person, item, day, and their respective interactions. Second, we calculated Pearson correlations to investigate if the number of experienced meaningful coincidences assessed in everyday life was significantly associated with real-life creativity (and the other measures of creativity, i.e., creative potential and openness). Third, to investigate if the association between meaningful coincidences and real-life creativity was mediated by PA (and NA), we calculated mediation analyses with the lavaan package (vers. 0.6–14). Third, to evaluate the association between meaningful coincidences and affect from a trait as well as a state perspective, we calculated a multilevel model with PA and NA as fixed effects predicting the number of meaningful coincidences per day. In this robust multilevel model, we used participants as random factor and level 2 (grand mean centered) and level 1 (group mean centered) affect (i.e. PA and NA) as fixed effects. We applied robust linear mixed effects modeling (package robustlmm; [ 53 ]) using R [ 52 ] to account for potentially biasing outliers. The level of significance was fixed at p < .05 (two-tailed).

The experience of meaningful coincidences at the aggregated between-person level was significantly associated with the score on the coincidence questionnaire ( r = .38, p < .001). This indicates validity of the frequency of meaningful coincidences assessed via the daily diary method. 93% of participants experienced at least one meaningful coincidence throughout the assessment. As illustrated in Table 2 , 14% of variance of meaningful coincidences was between-person and 5% of variance was due to day-to-day variance (i.e., interaction between person and time; see Table 2 ). The proportion of between-person variance was comparable for PA and NA.

https://doi.org/10.1371/journal.pone.0300121.t002

Meaningful coincidences and creativity.

Creative achievements ( r = .31, p = .003) and creative activities ( r = .35, p < .001) were significantly associated with the number of coincidences experienced during the seven days of the daily diary assessment (see Fig 1 ).

https://doi.org/10.1371/journal.pone.0300121.g001

Follow-up analyses indicated that for creative activities this association was driven by literature, music, arts and crafts, as well as sports (see S1 Table ). For creative achievements, the follow-up analyses indicated that the association was driven by music, cooking, sports, and performing arts (see S1 Table ).

As illustrated in Table 3 , Openness was not associated with the experience of coincidences ( r = .12, p = .233). Age was negatively associated with NA and creative achievements. Therefore, the association between creative achievements/activities and coincidences are virtually unchanged when controlling for age.

https://doi.org/10.1371/journal.pone.0300121.t003

Furthermore, PA and NA were positively related with the number of meaningful coincidences. Only NA was significantly related to creative activities. The creative potential was associated with creative achievements indicating some validity.

Mediating effects of affect on the association between creativity and meaningful coincidences.

The mediation analysis indicated no significant indirect path of coincidences via PA to creative activities (Beta = -1.58, p = .540). The total effect (direct plus indirect) was significant (Beta = 34.00, p < .001). We observed a similar pattern for NA, with no significant indirect path (Beta = 4.66, p = .127; total effect: Beta = 34.00, p < .001). This pattern of findings argues for independent associations of coincidences and PA with creative activities.

A dynamic perspective on meaningful coincidences.

The robust multilevel model indicated that more meaningful coincidences were perceived on days on which participants showed higher PA (Level 1; Beta = 0.08, p < .001). People with a higher trait PA (at Level 2) also perceived more coincidences in general (Beta = 0.11, p = .004). Similarly, NA was significantly associated with meaningful coincidences at Level 2 (Beta = 0.15, p = .004) and at Level 1 (Beta = 0.04, p = .029).

Discussion study 2

Study 2 replicated the association between the experience of meaningful coincidences and real-life creativity such as creative activities and creative achievements. This finding fits the results of Baas et al. [ 34 ], who reported an association between positive schizotypy and creative achievements. Furthermore, the experience of meaningful coincidences was again, unrelated to the measure of creative potential. For the very first time this study applied an internet daily diary method to assess the frequency of meaningful coincidences and found that people perceive a considerable number of meaningful coincidences within a week on a day-to-day basis. Furthermore, the highest proportion of the number of experienced coincidences variance was between-person (14%), which underlines the assumption that meaningful coincidences represent a trait, which we can reliably assess in everyday life by means of daily diary [ 2 , 3 ]. The within-person variance was assessed at the lower boundary of reliability [ 48 ]. However, this proportion of variance (i.e., 5%) was meaningfully associated with fluctuations of people’s affect, arguing for validity of assessment. Exactly during days on which participants experienced more PA and NA, they also experienced more meaning in random arrangements. This is partly in line with a study of Conner and Silva [ 54 ], who conducted a two-week internet daily diary study and showed that PA states were associated with creative behavior and creative activity [ 55 – 57 ].

The association between PA states and the number of experienced meaningful coincidences underlines the validity of the assessment and offers a novel state perspective on the link between PA and the perception of meaning in everyday life [ 23 ]. Not only peoples’ affect fluctuates from day-to-day, but their experience of meaningful coincidences seems to do so as well, and both seem to be related. Furthermore, the traits of PA and NA were associated with the number of meaningful coincidences througout the week. This is in accordance with cross-sectional research indicating a link between positive feelings, well-being, and life-satisfaction with the experience of more meaningful coincidences [ 23 ]. The correlation with NA argues for the assumption that the experience of meaningful coincidences may serve as a coping mechanism [ 28 ]. Similar to this finding, Russo-Netzer and Icekson [ 23 ] reported a positive link between coincidences and depressive symptoms (in a non-clinical sample). They suggested that the experience of meaningful coincidences (and synchronicity) may serve as one pathway to life satisfaction. The relationship with PA fits this assumption well.

The observed pattern of findings is convincing since PA and NA were entered simultaneously into our model. This approach rules out a simpler interpretation that the positive association between meaningful coincidences and PA and NA was because of a conformation bias or the tendency of perceivers to say “yes” due to a lower threshold for giving positive answers [ 18 , 58 ]. This would have affected all daily diary ratings to a similar extend, which would have resulted in non-significant findings.

Importantly, neither PA nor NA seems to serve as a mechanism explaining why people who perceive more meaningful coincidences also showed more real-life creativity. Study 2 investigated this association on the between-person level. Consequently, we conducted a study 3 to provide a more detailed look into the within-person dynamics of meaningful coincidences, affect, and creative activities.

In study 3, our first goal was to replicate the association between creative activities (and achievements) with the experience of meaningful coincidences. Based on the study findings outlined above, we pre-registered study 3 [ 59 ]. The second goal was to extend this finding and provide a conceptual replication by additionally using a different measure of the daily experience of synchronicity [ 23 ] and by assessing the creative activities for each day–again increasing ecological validity and reducing recall biases in the data.

As the third aim, we studied if creative activities and the experience of meaningful coincidences would show a dynamic relationship from day to day. Based on this relationship we targeted to investigate affect (PA and NA) as a potential mechanism why people show more creative activities when experiencing more meaningful coincidences (and vice versa; on a day-to-day basis). PA might serve a valuable mechanism because it goes along with more creative activities (e.g., [ 57 ]) and meaningful coincidences (see study 2). Furthermore, also NA shows associations with meaningful coincidences (see study 2) and is (theoretically) linked to creativity as well [ 43 ].

To sum up, study 3 assessed real-life creativity, creative potential, and openness cross-sectionally by means of questionnaires, and additionally measured the dynamic of meaningful coincidences, affect, and creative activities by means of an internet daily diary. We preregistered the cross-sectional association between meaningful coincidences and real-life creativity [ 59 ] and targeted a conceptual replication as well. Furthermore, we investigated if affect (PA, NA) might mediate the association between meaningful coincidences and creative activities at the within and the between-person level.

All data and scripts of Study 3 are available online [ 59 ]. We planned to collect a sample of at least 80 participants for up to seven days, which should be sufficient to find a significant Pearson correlation of medium size ( r = .30) with a power of .80 and an alpha of .05. We only analyzed participants who answered at least 3 daily diaries. The final sample comprised 80 participants with a mean age of 29.66 years ( SD = 12.82). In total, 35 men, 44 women, and one diverse person participated in the study, which did not take place in a specific context. The local ethics committee approved the study (GZ. 39/82/63 ex 2014/15), the recruitment period of this study was from 01 May to 30 June 2023, and all participants gave written informed consent by clicking an agree to participate button.

Procedure of the daily diary assessment.

Participants received a daily email with a link to the online survey including 8 questions on the experience of meaningful coincidences, 20 items of the PANAS, 8 questions asking for creative activities, and further 9 items assessing synchronicity (a concept similar to meaningful coincidences). The study started on a Monday, where we also applied the cross-sectional questionnaires, and ended on a Sunday. Participants had to answer the daily diary each day between 6 pm and 12 pm. We used LimeSurvey to apply the questionnaires (LimeSurvey GmbH, Hamburg, Germany).

Coincidences questionnaire.

We used the 8 items from study 2 [ 3 ]. Participants answered how often they perceived specific meaningful coincidences during the day. In accordance with study 2, we transformed the number of coincidences per day into a 5-point ordinal scale from 0 to 4. The mean number of perceived meaningful coincidences per day on the between-person level was 0.22 ( SD = 0.28) ranging from 0 to 1.5. The mean number of coincidences per day was 0.20 ( SD = 0.32) ranging from 0 to 1.88. Again between-person reliability was excellent (R kR = .94), and within-person level was acceptable R C = .42.

Synchronicity awareness.

To assess synchronicity (i.e., meaningful coincidences), we used the German translation of the Synchronicity Awareness scale [ 23 ]. Similar to the coincidence questionnaire [ 3 ], the SA scale referred to awareness of the occurrence of synchronicity events and involves 9 items. We used this newly developed assessment approach, see if results can be replicated between the two coincidence questionnaires. We used a 5-point Likert Scale from 0 to 4 (i.e., no synchronicity experience, 4 and more experiences per day). The mean number of synchronicity events per day on the between-person level was 0.34 ( SD = 0.28) ranging from 0 to 1.37. The mean score per day was 0.32 ( SD = 0.36) ranging from 0 to 3.11 The between-person (R kR = .86) and within-person reliability (R C = .62) were good.

Creative activities.

We asked participants to rate if they had performed creative activities on 8 different domains taken from the ICAA [ 11 ]. We calculated the sum score of creative activities per day. At the day level the mean score was 1.16 ( SD = 1.21) ranging from 0 to 7. At the between-person level the mean score was 1.24 ( SD = 0.97) ranging from 0 to 4.25. People reported at least one creative activity on 61.20% of the days. Creative activities were assessed reliable with R kR = .90 for between-person. The within-person reliability was low with R C = .13.

Positive and negative affect.

Again, we asked participants to answer all 20 items of the German version of the Positive and Negative affect Schedule (PANAS; [ 49 ]). The mean PA per day was M = 2.91 ( SD = 0.79; NA: M = 1.40, SD = 0.47) ranging from 1 to 5 (NA: 1 to 4.2). The mean PA at the between-person level was M = 2.92 ( SD = 0.57; NA: M = 1.42, SD = 0.35) ranging from 1.69 to 4.55 (NA: 1 to 2.55). The reliability of the positive affect was good for between-person (R kR = .84) and for within-person (R C = .89). Similarly, negative affect was assessed reliable with R kR = .85 for between-person and R C = .78 for within-person variation.

Creative activities (CAct) and creative achievements (CAch) were assessed by means of ICAA [ 11 ]. The sum score of creative activities showed an internal consistency of α = .75 (creative achievements: α = .71).

We used the alternate uses task (AUT; [ 60 ]) to assess peoples creative potential. Participants were instructed to produce as many creative uses as possible for the objects brick and paperclip within a 3-minute time limit for each. The Spearman-Brown corrected reliability of both items was .86. The mean fluency score was 6.24 ( SD = 3.03). To calculate the originality, we translated the ideas from German into English via google translate and consequently checked them. For the translated ideas, we calculated distance scores by means of the online application SemDis. We used multiplicative models for semantic distance computations [ 61 ] and calculated the maximum score of the semantic distance factor per item as a measure of originality. The mean score showed a Spearman-Brown corrected reliability of .56. The originality score was 0.08 ( SD = 0.04).

Openness to experiences.

Openness to experiences was assessed via the German version of the BFAS-G with 10 items [ 62 ]. The items were answered on a seven-point Likert scale. Cronbach alpha was α = .79.

First, and in accordance with study 2, we evaluated the proportion of variance associated with between-person and within-person variance for PA, NA, meaningful coincidences, synchronicity, and creative activities by applying the GTA [ 46 ] by the use of the software psych (Version 2.3.3; [ 51 ]) running in R (Version 3.4.2; [ 52 ]). Second, to replicate the findings of study 1 and 2 we calculated single-order Pearson correlations between meaningful coincidences and real-life creativity (and the other aspects of creativity). Third, to investigate if PA and NA (from a state and trait perspective) were associated with meaningful coincidences, we calculated multilevel models predicting coincidences. Fourth, to investigate if meaningful coincidences are associated with the fluctuation of creative activities from day to day, we calculated multilevel models predicting creative activities by means of coincidences. Fifth, we calculated multilevel models predicting creative activities by means of affect. We used robust linear mixed effects models (package robustlmm; [ 53 ]) using R [ 52 ] for all multi-level models. Finally, to investigate potential mediating effects of PA and NA for the association between meaningful coincidences and creative activities, we calculated mediation analyses taking the between-person (2-2-2) and within-person (1-1-1) level into account. We used the lavaan package (vers. 0.6–14). As a conceptual replication attempt, all analyses were also calculated for synchronicity awareness. The level of significance was fixed at p < .05 (two-sided).

The experience of meaningful coincidences at the aggregated between-person level was significantly associated with the synchronicity awareness score ( r = .76, p < .001). Only 6 participants perceived no single meaningful coincidence. 20% of variance of meaningful coincidences was between-person and 5% were due to day-to-day fluctuations. For synchronicity, the between-person was 10% and the within-person variances was 9%. Here, only 2 participants had no synchronicity experience at all (see Table 4 ). The proportions of variance for PA and NA were comparable to study 2. Creative activities showed 7% of variance between-person and only 1% within-person.

https://doi.org/10.1371/journal.pone.0300121.t004

Meaningful coincidences and real-life creativity.

Creative achievements ( r = .23, p = .044) and creative activities ( r = .31, p < .001) were significantly associated with the number of coincidences experienced during the seven days of the daily diary assessment. We found a similar pattern for synchronicity awareness (see Fig 2 ). Follow-up analyses indicated that the significant finding for creative activities was driven by literature, music, sports, visual arts, and performing arts (see S1 Table ). For creative achievements, the finding was mainly due to cooking, visual arts, and performing arts (see S1 Table ).

https://doi.org/10.1371/journal.pone.0300121.g002

As illustrated in Table 5 , age was significantly associated with NA and creative achievements, however, not with meaningful coincidences and synchronicity. Therefore, age does not substantially affect the correlation between coincidences/synchronicity and creative activities/achievements. Gender showed no significant association.

https://doi.org/10.1371/journal.pone.0300121.t005

Furthermore, openness was not associated with coincidences. PA and NA were positively associated with the number of meaningful coincidences as well as the synchronicity score. NA affect was significantly related to creative activities. PA affect and openness were associated with creative activities and achievements. The creative potential measured via fluency was associated with creative activities, achievements as well as openness. The semantic distance score was associated with the fluency score and at a trend with PA. This pattern of findings indicates some validity of measures.

The association between meaningful coincidences and affect.

The robust multilevel model indicated that people with a higher PA as well as higher NA as a trait (at Level 2) also perceived more coincidences in general. PA each day (Level 1) and as a trait (Level 2) also predicted the experience of synchronicity (see Table 6 ).

https://doi.org/10.1371/journal.pone.0300121.t006

The association between meaningful coincidences and creative activities.

As illustrated in Table 7 , the robust multilevel model indicated that meaningful coincidences as a trait (Level 2) predicted creative activities in everyday life. A similar pattern was found for synchronicity awareness, which additionally predicted creative activities from day to day.

https://doi.org/10.1371/journal.pone.0300121.t007

The association between creative activities and affect.

The robust multilevel model indicated that state PA (Level 1, Beta = 0.20, p = .002) and trait PA (Level 2; Beta = 0.48, p = .019) predicted creative activities in everyday life, but not NA (Level 1: Beta = -0.06, p = .565; Level 2: Beta = 0.27, p = .427).

The mediating effect of affect on the association between meaningful coincidences/synchronicity and creative activities in everyday life.

The mediation analysis indicated no significant indirect path of meaningful coincidences via PA to creative activities, neither for the between-person Level 2 (Beta = 0.17, p = .423), nor for the within-person Level 1 (Beta = 0.05, p = .150). The total effect (direct plus indirect) was significant for between-person (Beta = 1.76, p < .001) but not for within-person (Beta = 0.35, p = .113; see Fig 3 ). This pattern of findings argues for an independent association of coincidences and PA with creative activities at both levels of analysis.

https://doi.org/10.1371/journal.pone.0300121.g003

For synchronicity, we found a very similar pattern. However, the indirect path of synchronicity via PA to creative activities was significant for within-person (Beta = 0.08, p < .027) and the direct effect to creativity was not (Beta = 0.28, p = .068). This indicates a within-person mediation effect of the association between synchronicity and creative activities (see Fig 3 ).

We found no significant mediation effect of NA neither for meaningful coincidences nor for synchronicity.

Discussion of study 3

As preregistered, study 3 replicated the findings of study 1 and study 2. The propensity to experience meaningful coincidences was again related to real-life creativity such as creative activities and creative achievements. Furthermore, we were also able to conceptually replicate this pattern in a twofold manner. First, by means of an alternative questionnaire developed by Russo-Netzer and Icekson [ 23 ] applied in a daily-diary context and second, by assessing creative activities from day to day [ 57 ]. Both modifications showed the very same pattern of findings–that there is a link between synchronicity and creative activities at the between-person level.

Well in line with literature, PA at day level predicted the number of creative activities participants were engaged in [ 54 , 55 , 57 ]. However, creative activities on a day-to-day basis were not that strongly associated with meaningful coincidences. People who perceive more meaning in randomly arranged events are likely more often engaged in creative activities and report more creative achievements at the trait level. This strengthens the interpretation of the experience of meaningful coincidences as a propensity [ 2 , 3 ]. By applying an internet daily diary assessment, we were able to decrease potential influences of memory and belief processes on our findings [ 37 ].

However, study 3 did not replicate the finding of study 2, which had suggested that the dynamic of PA and NA from day to day is associated with the experience of meaningful coincidences. However, of note, we found an association of PA state and synchronicity awareness. The mediation effect on the association between synchronicity and creative activities was only found for PA at the within-person level. Therefore, it seems promising to conduct further investigations on why and when (positive) affective states might impact the experience of meaningful coincidences.

General discussion

In this three-study approach, we repeatedly found that the experience of meaningful coincidences is associated with real-life creativity such as creative activities (three out of three studies) and creative achievements (two out of three studies). This association was firstly found in a cross-sectional online study, then replicated in a consecutive internet daily dairy study and finally these associations were replicated in an additional preregistered daily diary study. In total, we consecutively investigated more than 200 participants with different methods and questionnaires targeting to measure a broad spectrum of creativity and creative potential indices. Our approach was motivated by the idea that although one single study might not tell us much about the link between creativity and meaningful coincidences, three studies might. Unfortunately, at the same time, a higher number of studies also leaves us with a higher number of open questions.

Although the experience of meaningful coincidences was linked to creative activities and creative achievements, follow-up analyses indicate a less coherent pattern of findings. For creative activities, literature, sports, and performing arts (at a trend) might be responsible for the significant relationship with meaningful coincidences throughout the three studies. For creative achievements cooking seems to show the most consistent pattern of associations. However, it is important to consider that these associations strongly depend on sample characteristics and if participants show creative achievements and creative activities in these very sub-categories. This means that while the sum score can reliably estimate participant’s creative achievements and activities, the sub-scores may not necessarily.

Thus, we should consider this report as a preliminary, first attempt to study the benevolent trajectory of meaningful coincidences indicated by the assocition with real-world creativity. Although the novel findings presented here are in accordance with the assumption of a connection between the experience of meaning and creativity (e.g., [ 2 , 10 ]), we still do not know much about the mechanisms providing this robust link.

What can we learn about meaningful coincidences?

Following from the study results, we can conclude that assessing the experience of meaningful coincidences is possible by means of an internet daily diary approach. Applying daily diaries reduces systematic recall biases, brings more ecological validity to our study results and allows to measure the dynamic of meaningful coincidences and synchronicity as well [ 37 ]. The assessment of changes from moment to moment is an important prerequisite for future research uncovering potential mechanisms responsible for the experience of meaningful coincidences. This is even more important since the experience of meaningful coincidences is a naturally occurring phenomenon, which experimenters cannot induce artificially. Until to date, research considered meaningful coincidences a (personality) trait only [ 2 ]. Since in our studies, most people perceived at least one meaningful coincidence and synchronicity during the span of a week, a more dynamic assessment by means of a daily diary or experience sampling methods might help to shed more light on this phenomenon [ 63 , 64 ]. Furthermore, the within-person variance of meaningful coincidences was assessed reliably (at the lower border) and was meaningfully associated with PA as well as NA (study 2). This finding was only partly replicated in study 3, yet it encourages future attempts to study the facilitating mechanisms of meaningful coincidences and synchronicity by means of these methods.

What can we learn about the association between meaningful coincidences and creativity?

Is the experience of meaningful coincidences related to creativity? The answer to this exciting question is complex. For one, creativity is not a single, homogeneous entity and can be understood as a potential, a personality trait, an activity, as well as an achievement. Furthermore, creativity and the experience of meaningful coincidences can be studied at two levels (between-person and within-person). The present studies did not find strong evidence suggesting that seeing connections and meaningful patterns which are hidden from the view of others (cf . [ 16 ]) help to overcome mental blocks, enhance thinking out of the box, or increase the chance to reach more creative ideas. Although, the association with real-life creativity seems robust, relationships with other aspects of creativity such as openness to experience or people’s creative potential are much weaker and show lower effect sizes ([ 18 , 34 ]). The association seems to be more about displaying creative activities such as cooking and writing blogs (as well as creative achievements) and less about the basic potential to come up with many original ideas to solve creative problems.

The reason for the obtained robust link between creative activities and meaningful coincidences is unclear. We did not find evidence for a convincing mechanism; however, PA might still represent a target for future studies. At least in study 3, we found that PA states seem to mediate the association between synchronicity and the number of creative activities in everyday life. Nevertheless, due to the study findings, we can rule out at least two reasons for the association between meaningful coincidences and creativity: First, the association between the propensity to experience meaningful coincidences and real-life creativity is not (strongly) based on a higher creative potential of these people. People who experienced more meaning in meaningless noise did not necessarily show a higher creative potential on the implemented creativity task, while at the same time, reporting higher creative achievement and more creative activities. The null finding for creative potential is in line with previous research indicating that people with a higher creative potential also have higher inhibitory and executive skills (e.g., [ 22 , 65 , 66 ]). These skills seem to somewhat oppose the experience of meaningful coincidences, who were reported to show reduced working memory capacities [ 67 ] and reduced inhibitory control [ 2 , 20 ] as well as reduced grey matter in relevant cortical areas [ 28 ]. In our three-study approach, we used three different measures of creative potential, a figural drawing task (TCT-DP; [ 68 ]), naming unrelated words (DAT; [ 45 ]), and finding original uses of everyday objects (AUT; [ 60 ]). Not a single index of all these measures was associated with meaningful coincidences (or synchronicity) at the between-person level. This is even more compelling since, the creative potential measures seemed to be valid and showed the expected associations with other creativity indices such as creative achievements (e.g., [ 35 ]). When taking the effect sizes into account, we can conclude that if there is any association at all, it is likely small. Therefore, future studies with larger samples have to investigate this association more deeply and should also place emphasis on possible within-person associations of meaningful coincidences and creative abilities (see e.g., [ 27 , 69 ]).

Second, affect might not explain the association between meaningful coincidences and creative activities. Negative affect was associated with more creative activities and meaningful coincidences, which might indicate that people with a higher NA have a higher need to give their lives a meaning [ 23 , 70 ]. However, the variance of this association was not shared between both variables, thus discouraging a mediating mechanism. Similarly, PA was associated with meaningful coincidences at the between-person and within-person level, which however only partly explains the link between real-life creative activities and the experience of meaningful coincidences.

A further target for future research is curiosity, described as the tendency to explore uncertain environments [ 71 ]. Following the proposed framework by Ivancovsky [ 71 ], curiosity is associated with increased creativity via novelty-seeking. This might explain why people who perceive more meaning in randomness might also engage in more creative activities and show more creative achievements (see e.g., [ 34 ] for positive schizotypy). However, this assumption needs further empirical support.

Conclusions

Taken together, our three consecutive studies allow three main conclusions. First, our studies showed that people who experienced more meaningful coincidences are also more often engaged in real-life creativity such as creative activities and creative achievements [ 11 ]. This association at the between-person level was also found for the within-person level, which indicates the relevance of meaningful coincidences for real-life creativity and vice versa. Second, PA and NA are related to more perceived meaningful coincidences, predominantly at the level of the person. There are good reasons to assume that this link exists at the within-person level too. Nevertheless, PA and NA seem not to serve as strong mechanisms explaining the link between meaningful coincidences and creative activities. Third, creative potential seems less strongly linked with the experience of meaningful coincidences as literature might have implicitly suggested. The robustly observed association between creative activities and achievements with the perception of meaning fits the ideas of Kaufman [ 70 ], who suggested that creativity can help people find meaning in their lives. This statement may also be true for the experience of meaningful coincidences, which can be understood as a creative act in which people connect unrelated but significant events and give these events and consequently, also their lives, some meaning.

Supporting information

S1 table. pearson correlations between coincidences and the subscales of the icaa for study 1, study 2, and study 3..

p values in parentheses. Creative achievements = CAch, Creative activities = CAct.

https://doi.org/10.1371/journal.pone.0300121.s001

Acknowledgments

The authors acknowledge the financial support for publication by the University of Graz.

Registration

This study was pre-registered in the Open Science Framework (OSF, [ 59 ]).

• View Article
• Google Scholar
• 4. van Elk M, Friston K, Bekkering H. The experience of coincidence: An integrated psychological and neurocognitive perspective. In: van Wolde E, Landsman K, editors. The challenge of chance: A multidisciplinary approach from science and the humanities. Springer; 2016. pp. 171–85.
• PubMed/NCBI
• 6. Conrad K. Die beginnende Schizophrenie. Versuche einer Gestaltanalyse des Wahns. Stuttgart: Georg Thieme Verlag; 1958.
• 9. DeYoung CG. Openness/intellect: A dimension of personality reflecting cognitive exploration. In: Mikulincer M, Shaver PR, editors. Attitudes and Social Cognition. 1st ed. Washington, DC: American Psychological Association; 2015. pp. 369–99.
• 19. Brugger P. From haunted brain to haunted science. A cognitive neuroscience view of paranormal and pseudoscientific thought. In: Houran J, Lange R, editors. Hauntings and poltergeists. Multidisciplinary perspectives. Jefferson: McFarland & Company Inc. Publishers; 2001. pp. 195–213.
• 37. Gunthert KC, Wenze SJ. Daily diary methods. In: Mehl MR, Conner TS, Csikszentmihalyi M, editors. Handbook of research methods for studying daily life. New York: Guilford Press; 2012.
• 38. Hamaker EL. Why researchers should think “within-person”: A paradigmatic rationale. In: Mehl MR, Conner TS, Csikszentmihalyi M, editors. Handbook of research methods for studying daily life. New York: Guilford Press; 2012. pp. 43–61.
• 46. Shrout PE, Lane SP. Psychometrics. In: Mehl MR, Conner TS, Csikszentmihalyi M, editors. Handbook of research methods for studying daily life. New York: Guilford Press; 2012. pp. 302–20.
• 52. R Core Team. R. A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2023.
• 60. Guilford JP. The nature of human intelligence. New York: McGraw-Hill; 1967.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Environ Sci Ecotechnol
• v.19; 2024 May
• PMC10656232

Smarter eco-cities and their leading-edge artificial intelligence of things solutions for environmental sustainability: A comprehensive systematic review

Simon elias bibri.

a School of Architecture, Civil and Environmental Engineering (ENAC), Civil Engineering Institute (IIC), Visual Intelligence for Transportation (VITA), Swiss Federal Institute of Technology Lausanne (EPFL), Lausanne, Switzerland

John Krogstie

b Department of Computer Science, Norwegian University of Science and Technology (NTNU), Trondheim, Norway

Amin Kaboli

c School of Engineering, Institute of Mechanical Engineering, Swiss Federal Institute of Technology Lausanne (EPFL), Lausanne, Switzerland

Alexandre Alahi

The recent advancements made in the realms of Artificial Intelligence (AI) and Artificial Intelligence of Things (AIoT) have unveiled transformative prospects and opportunities to enhance and optimize the environmental performance and efficiency of smart cities. These strides have, in turn, impacted smart eco-cities, catalyzing ongoing improvements and driving solutions to address complex environmental challenges. This aligns with the visionary concept of smarter eco-cities, an emerging paradigm of urbanism characterized by the seamless integration of advanced technologies and environmental strategies. However, there remains a significant gap in thoroughly understanding this new paradigm and the intricate spectrum of its multifaceted underlying dimensions. To bridge this gap, this study provides a comprehensive systematic review of the burgeoning landscape of smarter eco-cities and their leading-edge AI and AIoT solutions for environmental sustainability. To ensure thoroughness, the study employs a unified evidence synthesis framework integrating aggregative, configurative, and narrative synthesis approaches. At the core of this study lie these subsequent research inquiries: What are the foundational underpinnings of emerging smarter eco-cities, and how do they intricately interrelate, particularly urbanism paradigms, environmental solutions, and data-driven technologies? What are the key drivers and enablers propelling the materialization of smarter eco-cities? What are the primary AI and AIoT solutions that can be harnessed in the development of smarter eco-cities? In what ways do AI and AIoT technologies contribute to fostering environmental sustainability practices, and what potential benefits and opportunities do they offer for smarter eco-cities? What challenges and barriers arise in the implementation of AI and AIoT solutions for the development of smarter eco-cities? The findings significantly deepen and broaden our understanding of both the significant potential of AI and AIoT technologies to enhance sustainable urban development practices, as well as the formidable nature of the challenges they pose. Beyond theoretical enrichment, these findings offer invaluable insights and new perspectives poised to empower policymakers, practitioners, and researchers to advance the integration of eco-urbanism and AI- and AIoT-driven urbanism. Through an insightful exploration of the contemporary urban landscape and the identification of successfully applied AI and AIoT solutions, stakeholders gain the necessary groundwork for making well-informed decisions, implementing effective strategies, and designing policies that prioritize environmental well-being.

Graphical abstract

• • AI and AIoT hold significant potential to address complex environmental challenges.
• • AI and AIoT are being applied in smart cities to enhance their performance and efficiency.
• • The concept of smarter eco-cities is based on the groundbreaking convergence of AI and IoT.
• • Smarter eco-cities are leveraging AI and AIoT to advance their environmental sustainability goals.
• • AI and AIoT pose environmental, technical, ethical, social, and regulatory challenges.

1. Introduction

The rapid advancement and groundbreaking convergence of Artificial Intelligence (AI) and the Internet of Things (IoT) has triggered profound transformations across various domains, including environmental sustainability, climate change, and urban development. This surge has led to the emergence of smarter eco-cities, where Artificial Intelligence of Things (AIoT) takes center stage and garners significant attention. In this context, AIoT is poised to offer innovative solutions to the mounting environmental challenges confronting smart cities and, by extension, smart eco-cities. AIoT holds the potential to unlock avenues for improving resource efficiency, reducing energy consumption, streamlining waste management, enhancing transportation management, conserving biodiversity, and mitigating environmental impacts. These advancements hold the power to reshape the urban development landscape in response to the surging wave of urbanization and the increasing complexity of ecological degradation, transforming cities into hubs of intelligence, sustainability, and environmental consciousness.

AI is rapidly reshaping both the technological and urban landscapes, serving as a pivotal driving force behind the advancement of smart cities and smart eco-cities (e.g., Ref. [ [1] , [2] , [3] , [4] ]). Its potential for disruption and innovation (e.g., Ref. [ 5 , 6 ]) has positioned it at the forefront of these developments. This influence profoundly alters the functioning of urban systems and the intricate interactions, behaviors, and responses of their subsystems to the surrounding environment. As a result, urban processes and practices are undergoing significant transformation, marked by their increasing alignment with data-driven scientific urbanism. The impact of AI on urban systems and activities has been consistently expanding [ 7 , 8 ], with its computing capabilities experiencing exponential growth to accommodate the mounting influx of data from diverse sources facilitated by IoT. The potential of IoT lies in its capacity to facilitate data analysis through AI models and algorithms, a synergy poised to catalyze environmentally sustainable urban development.

The centralized infrastructure of IoT is grappling with significant challenges, primarily driven by the extensive strain imposed by the immense volume of data being generated and processed. Harnessing actionable insights from these data necessitates the integration of AI models and algorithms to effectively manage the data flow, storage, and processing inherent in the IoT infrastructure. The emergence of AIoT is underpinned by various factors that set it apart from traditional IoT. To begin with, AIoT capitalizes on the synergies between AI and IoT technologies, facilitating more intelligent and efficient data processing, analysis, and decision-making (e.g., Ref. [ [9] , [10] , [11] ]). Through the integration of AI and Machine Learning (ML)/Deep Learning (DL) capabilities into IoT devices and systems, AIoT empowers real-time data insights, predictive analytics, and adaptive responses, thereby optimizing the overall system performance and efficiency. A key distinction lies in AIoT's ability to overcome the limitations of IoT in managing the copious and diverse data generated by the vast network of interconnected devices. Additionally, AIoT addresses the challenges associated with transmitting the rapid torrent of data from distributed sensor network infrastructure [ 12 , 13 ]. By harnessing the power of AI, AIoT effectively processes and contextualizes intricate and multifaceted data streams, unlocking the potential for advanced applications. Most notably in certain domains, AIoT paves the way for autonomous and intelligent decision-making by IoT devices, enabling them to learn, adapt, and optimize operations in response to shifting environmental conditions and user requirements. In essence, the emergence of AIoT introduces a realm of possibilities for innovation, optimization, and automation across diverse domains, including environmental sustainability, climate change, and smart cities (e.g., Ref. [ 1 , [13] , [14] , [15] , [16] ]).

Similar to AI, AIoT has become integral to the functioning of smart cities and, hence, smarter eco-cities. Notably, it has demonstrated innovative potential in addressing complex environmental challenges. Recent research has concentrated on the practical applications of AI and AIoT across various domains of environmental sustainability and climate change (e.g., Ref. [ [17] , [18] , [19] , [20] , [21] ]). Toward the end of 2020 onward, this focus has expanded to encompass smart cities in terms of their management and planning (e.g., Ref. [ [22] , [23] , [24] , [25] , [26] ]). Fundamentally, however, there is a strong interconnection between smart cities and eco-cities in that they have significantly influenced one another over the last decade, particularly in the domains of environmental sustainability and climate change. Eco-cities have long been associated with these two domains, serving as a well-established paradigm of sustainable urbanism (e.g., Ref. [ [27] , [28] , [29] , [30] , [31] , [32] , [33] ]). However, these two domains have frequently been addressed separately or, more recently, in connection with smart cities, particularly within the context of AI and AIoT, instead of being collectively considered within the framework of smart eco-cities. This suggests that there has been a strong tendency to prioritize one approach over another, specifically the second approach has been given a little attention. Especially, eco-cities and, by extension, smart eco-cities serve as experimental grounds for innovative solutions and environmental transitions (e.g., Ref. [ 27 , [34] , [35] , [36] ]). The experimentation within eco-cities extends beyond climate change to encompass energy transition, resource conservation, transport efficiency, biodiversity conservation, and experimental simulation and modeling [ [37] , [38] , [39] ]. These strategies and principles, in turn, form the fundamental driving forces behind the evolution of smart eco-cities.

Amid a world increasingly beset by uncertainties, compounded by the urgency of the climate crisis and the rapid digital transformation of smart cities and eco-cities, catalyzed by the emergence of AI and AIoT technologies, a compelling trajectory is emerging. This trajectory envisages the integration of these technologies' applied solutions into smart eco-cities to deal with the complexity of environmental degradation and climate disruption — under what can be termed as “smarter eco-cities.” This visionary concept entails the implementation of innovative, forward-looking strategies to reshape the urban landscape of the future. However, it is worth acknowledging that while AI and AIoT technologies hold immense potential yet to unlock, they pose environmental risks and amplify a spectrum of societal, ethical, legal, and regulatory challenges.

The body of literature exploring AI and AIoT solutions within the realm of environmental sustainability, climate change, and smart cities is rapidly expanding. Nonetheless, to the best of our knowledge, no review study has systematically analyzed and synthesized the existing corpus of knowledge regarding the interconnections and synergies between these three domains, coupled with their intersection with smart eco-cities with respect to AI and AIoT technologies and solutions. Furthermore, while a few recent review studies have taken a broader perspective on smart eco-cities, none have ventured into exploring their emerging technological and environmental solutions from an integrative perspective. Bridging these gaps, the present study embarks on a comprehensive systematic review of the burgeoning landscape of smarter eco-cities and their leading-edge AI and AIoT solutions for environmental sustainability. To ensure thoroughness, the study employs a unified evidence synthesis framework that seamlessly integrates configurative, aggregative, and narrative synthesis approaches. To achieve the overarching aim, the study focuses on the following specific objectives:

• • Describe, illustrate, and make meaningful connections between the fundamental concepts underpinning emerging smarter eco-cities.
• • Present the existing work in the field and explain how the present study differs from it and why this is a “step forward” and a new contribution to knowledge.
• • Analyze, synthesize, interpret, and critically evaluate the existing knowledge in the areas of environmental sustainability, climate change, and smart cities to derive comprehensive insights into the role of AI and AIoT solutions in the advancement of emerging smarter eco-cities.
• • Categorize AI and AIoT solutions based on these three areas and further evaluate their impact on advancing environmental sustainability goals.
• • Capture the dynamic landscape of AI and AIoT solutions for these three areas by identifying emerging trends, innovations, and novel approaches within the framework of smarter eco-cities.
• • Identify and discuss the key challenges and barriers that arise when implementing AI and AIoT solutions in the development of emerging smarter eco-cities.
• • Identify existing gaps and explore relevant avenues for future research and areas requiring further investigation.

By pursuing these specific objectives, the study endeavors to provide a holistic and in-depth understanding of the integration of AI and AIoT solutions within emerging smarter eco-cities, ultimately contributing to the advancement of environmental sustainability practices in urban contexts. Toward this end, it pursues the following research questions:

• RQ1: What are the foundational underpinnings of emerging smarter eco-cities, and how do they intricately interrelate, particularly urbanism paradigms, environmental solutions, and data-driven technologies?
• RQ2: What are the key drivers and enablers propelling the materialization of smarter eco-cities?
• RQ3: What are the primary AI and AIoT solutions that can be harnessed in the development of smarter eco-cities?
• RQ4: In what ways do AI and AIoT technologies contribute to fostering environmental sustainability practices, and what potential benefits and opportunities do they offer in the realm of smarter eco-cities?
• RQ5: What challenges and barriers arise when implementing AI and AIoT solutions for the development of smarter eco-cities?

By synthesizing the existing evidence and analyzing the state-of-the-art research to answer these research questions, the systematic review contributes to consolidating, enhancing, and transforming the existing knowledge on smart eco-urbanism by:

• • Uncovering the dynamjc interplay between urbanism paradigms, environmental solutions, and data-driven technologies in emerging smarter eco-cities.
• • Identifying the driving forces behind the materialization of smarter eco-cities, namely technological advancements, environmental concerns, and policy instruments.
• • Examining the multifaceted roles of AI and AIoT in environmental sustainability, climate change, and smart cities and how these technologies can be harnessed in the development of smarter eco-cities.
• • Exploring the specific ways in which AI and AIoT technologies propel sustainable development pracrices and advance environmental goals.
• • Highlighting best practices where AI and AIoT solutions can significantly contribute to improving the environmental sustainability of smarter eco-cities.
• • Identifying and evaluating a spectrum of challenges linked to the implementation of AI and AIoT in smarter eco-cities, aiming to illuminate potential obstacles and devise strategies to mitigate or overcome them.
• • Unveiling uncharted territories and encouraging the exploration of AI and AIoT solutions to drive large-scale implementations of smarter eco-cities, propelling the discourse forward and fostering innovation in sustainable urban development.
• • Providing a visionary outlook by elucidating how AI and AIoT technologies contribute to sustainable urban futures and exploring the transformative potential of these technologies in redefining urban infrastructures and systems.

In essence, the study not only presents a comprehensive overview of the current landscape of smarter eco-cities but also highlights the potential of AI and AIoT technologies in shaping the future of sustainable urban development, in addition to providing a roadmap for advancing the discourse on smarter eco-cities and facilitating interdisciplinary collaborations. Moreover, the applied unified evidence synthesis approach offers a more holistic and nuanced understanding of the research topic addressed by enhancing the thoroughness, depth, and breadth of the systematic review. The insights derived from the systematic review will not only inform researchers and practitioners in the field but also guide policymakers and practitioners in making informed decisions regarding the adoption and implementation of AI and AIoT technologies in sustainable urban management and planning. Overall, by highlighting the solutions, opportunities, benefits, and challenges in the field of smarter eco-cities, the systematic review will further facilitate the advancement of research, policy, and practice in pursuing more sustainable and technologically advanced urban environments.

This study is structured as follows: Section 2 introduces, describes, and illustrates the key conceptual strands of the study. Section 3 addresses the research review related to the study. Section 4 describes and illustrates the methodology applied in the study. Section 5 presents the results of the literature analysis and synthesis. Section 6 provides a detailed discussion, covering key challenges, open issues, and limitations. Section 7 identifies relevant gaps and presents recommendations for potential research directions and areas that require more exploration. This study concludes, in Section 8, with a summary of key findings and implications.

2. Conceptual background

Key relevant concepts need to be clarified together with their integrative and synergistic facets. The value of linking these concepts ( Fig. 1 ) lies in facilitating a better understanding of the foundational underpinnings of emerging smarter eco-cities in terms of urbanism paradigms, environmental solutions, and data-driven technologies.

Smarter eco-cities and their underlying urbanism paradigms, environmental solutions, and data-driven technologies.

2.1. Smarter eco-cities and their underlying urbanism paradigms

2.1.1. smart cities.

Smart cities have gained significant attention as a potential solution to address sustainability, resource management, and urbanization challenges. Numerous attempts have been made to define the concept of smart cities. They suggest many definitions and a plethora of directions to smart city development (e.g., Ref. [ [40] , [41] , [42] ]). The concept has undergone many changes over the past two decades. In this regard, it promotes from a technology-oriented approach, i.e., infrastructures, architectures, platforms, systems, applications, and models, to a people-oriented approach, i.e., stakeholders, citizens, knowledge, services, and related data. Accordingly, it encompasses various dimensions, and there is no universally accepted definition up till now. However, the working definition for this study is justified by its alignment with the research objectives and scope. Accordingly, a smart city denotes an urban environment that leverages advanced technologies and data-driven approaches to conserve resources, minimize its environmental impact, and enhance overall ecological well-being. It prioritizes energy efficiency, sustainable transportation, waste reduction, water conservation, environmental monitoring, and green infrastructure to create a more eco-friendly and livable environment while fostering economic growth for all residents. Smart cities are increasingly emphasizing the role of technological advancements and scalable data-driven solutions to foster sustainable development practices (e.g., Ref. [ [43] , [44] , [45] , [46] ]). By integrating technology with environmental stewardship, smart cities strive to create greener and healthier living environments.

However, smart cities face several challenges that need to be addressed to ensure their successful implementation as well as their integration with other emerging paradigms of urbanism. As mentioned earlier, one of the primary problems in this regard is the lack of a standardized definition for smart cities. This lack of clarity has led to confusion and inconsistency in the planning and implementation of smart city initiatives. Moreover, the existing smart city infrastructures are not designed to support the integration of advanced technologies and data-driven systems. They need to support the connectivity, data collection, and efficient management of resources, which involves scalability and interoperability. As smart cities grow and more devices and sensors are connected, their infrastructure must handle the increasing volume of data and the growing number of users. Integration and seamless communication between different systems and devices are crucial for the smooth functioning of their infrastructure. In addition, with the extensive use of data and connected devices in smart cities, ensuring the security and privacy of their infrastructure becomes paramount. These infrastructures must have robust security measures in place to protect against cyber threats and safeguard the privacy of citizens' data. Several studies (e.g., Ref. [ 47 , 48 ]) highlight the importance of addressing data security and privacy issues and device-level vulnerability in the context of smart cities. Furthermore, coordinating and integrating various domains to create a cohesive and integrated smart city ecosystem is a complex challenge. Smart cities should strategically use networked infrastructure and associated data-driven technologies to produce a smart economy, smart government, smart mobility, smart environment, smart living, and smart people. More so, the social, ethical, political, legal, and regulatory challenges facing smart cities have shown to be difficult to deal with. To address these challenges, a multidimensional approach focusing on technology, citizens, and institutions is necessary. This includes developing robust technology infrastructure, implementing effective governance models, and actively involving citizens in decision-making processes. Overall, while smart cities hold great potential for advancing sustainable urban development, they pose significant challenges that require a comprehensive approach to build successful and inclusive ecosystems.

2.1.2. Eco-cities

The concept of eco-cities refers to “an urban environmental system in which input (of resources) and output (of waste) are minimized,” [ 49 ]. With their ubiquity today, eco-cities vary in the strategies and solutions they prioritize in response to environmental challenges in different urban contexts. They are widely diverse in conceptualization, implementation, and development. Thus, there is no definitive definition of an eco-city but rather a collection of concepts, ideas, and ambitions [ 32 ]. Broadly, eco-cities are urban areas designed and developed with a strong focus on environmental sustainability and ecological balance [ 50 ]. They aim to minimize their ecological footprint and promote a harmonious relationship between humans and nature. They prioritize energy efficiency, renewable energy sources, waste reduction, green spaces, sustainable transportation, and resource conservation. They strive to create a sustainable living environment that supports the well-being of citizens and the surrounding ecosystems while fostering social inclusivity and economic prosperity. The ultimate goal of eco-cities is to create resilient, low-carbon, and environmentally friendly urban spaces that contribute to a more sustainable future ([ 51 , 52 ]).

2.1.3. Smart eco-cities

Eco-cities manifest themselves into different models based mainly on applying the principles of urban ecology or combining the strategies of sustainable cities and the solutions of smart cities. For the latter, the most prominent of those models are smart eco-cities, which integrate IoT and Big Data technologies with environmental technologies for achieving urban sustainability (e.g., Ref. [ 35 , [53] , [54] , [55] , [56] ]). Smart eco-cities refer to urban environments that integrate advanced technologies, data analytics, and intelligent systems to enhance sustainability, efficiency, and quality of life while prioritizing environmental well-being. Accordingly, they leverage data-driven technologies and solutions to promote the use of renewable energy, Biomass Combined Heat Power (BCHP), sustainable transportation (walking, cycling, car sharing, biogas cars), eco-cycle waste management, green infrastructure, urban metabolism, sustainable buildings, smart grids, and sustainable urban planning strategies to minimize environmental impact, conserve resources, and foster sustainable and resilient urban environments (see Refs. [ 53 , 57 ] for illustrative case studies).

Unlike smart cities, smart eco-cities go beyond technology-driven approaches and emphasize environmental sustainability and ecological balance as central pillars. They focus on the integration of sustainable practices and advanced technologies, incorporate nature-based solutions into urban planning and design, and engage communities in environmental stewardship. They aim to create harmonious urban environments that not only leverage technology but also prioritize preserving and enhancing natural resources, biodiversity, and ecosystem services. In essence, they strive for a more holistic and nature-centric approach to urban development, promoting long-term sustainability and resilience and preserving natural resources for future generations to create environmentally friendly and livable urban communities.

2.1.4. Smarter eco-cities

The concept of smarter eco-cities, as specific to this study, describes smart eco-cities that integrate AI and AIoT technologies and solutions with environmental technologies and strategies to maximize the performance of their sustainable systems and integrate them with smart systems given the clear synergies in their operation. This integration is intended to produce combined effects greater than the sum of the separate effects of these systems in terms of boosting the benefits of environmental sustainability. Smart city systems include smart grids, smart traffic lights, smart mobility, smart buildings, smart waste management, and smart environmental monitoring. Smarter eco-cities represent an evolution or advancement of smart eco-cities. They prioritize environmental sustainability, employing cutting-edge technologies and innovative approaches to energy, waste, water, transportation, and urban planning to create more resilient, sustainable, and technologically advanced urban environments. They also emphasize a more holistic and comprehensive approach to urban development by integrating the environmental, social, and economic dimensions of sustainability. Accordingly, they aim to balance environmental preservation, social equity, and economic prosperity. They leverage the advanced technologies and solutions offered by smarter cities, notably AI and AIoT, to optimize urban systems and address complex challenges in a more integrated and intelligent manner. They are characterized by — as derived based on the synthesized studies:

• • The innovative potential of AI and AIoT technologies for sustainability;
• • The enhanced sustainability outcomes enabled by the applied AI and AIoT solutions;
• • The synergies between smart city systems and eco-city systems;
• • The optimized performance and efficiency of smart eco-city systems; and
• • The improved practices of urban management and planning.

In sum, while smart cities focus on technology-driven urban development, smart eco-cities place a stronger emphasis on achieving environmental sustainability through IoT and Big Data technologies. Smarter eco-cities go a step further by incorporating social and economic dimensions, leveraging emerging AI and IoT technologies for a more holistic approach to urban development.

2.2. Data-driven technologies

2.2.1. iot, computing models, and big data.

The term “IoT” describes the collective network of physical objects embedded with sensing, processing, communication, and actuating technologies and capabilities that enable them to exchange data with other devices over the Internet or other networks. These objects, often called “smart devices,” can range from everyday items to complex systems like city infrastructure. IoT allows these devices to communicate and interact with each other, collect data, transfer data, and perform automated tasks, leading to enhanced efficiency, performance, and sustainability in various urban domains. More and more IoT devices are connected worldwide daily and feeding vast amounts of data into analytical systems. It is estimated that 2.5 quintillion bytes of data are being generated globally daily, which will rise to 463 exabytes by 2025 [ 58 ].

Edge, fog, and cloud computing are three interconnected paradigms that play a pivotal role in the IoT ecosystem in smart cities and smart eco-cities in the context of environmental sustainability [ 53 ]. Edge computing involves processing data at or near the data source, often within the device or a nearby gateway. It aims to reduce latency and improve real-time responsiveness by executing computations locally. It is particularly useful for applications that require quick decision-making and low-latency interactions, such as autonomous vehicles. Fog computing extends the concept of edge computing by creating a hierarchical architecture that includes multiple edge devices and gateways. Fog nodes are strategically placed in proximity to data sources to perform intermediate processing, data filtering, and preliminary analytics before transmitting relevant data to the cloud. This approach optimizes bandwidth usage and enhances system performance, making it suitable for scenarios involving distributed data sources and resource-constrained devices. Cloud computing involves using centralized remote servers to store, manage, and process data. It offers vast computational resources and storage capabilities, making it suitable for complex data analytics, machine learning, and large-scale processing. Cloud computing allows data to be accessed and analyzed from anywhere with an internet connection, making it ideal for applications that require extensive computation and storage capabilities. These computing paradigms collaborate to create a holistic IoT ecosystem that efficiently manages data processing and analysis across different levels of the network infrastructure.

Big Data refers to extremely large and complex datasets that cannot be easily managed, processed, or analyzed using traditional data processing techniques. They are huge in volume, high in velocity, diverse in variety, exhaustive in scope, fine-grained in resolution, and relational, among others. Big Data analytics involves using advanced techniques and technologies to extract valuable insights, patterns, and correlations from large and complex datasets. This process typically involves data collection, storage, processing, analysis, and visualization. In summary, IoT and Big Data are interconnected concepts revolutionizing how we collect, analyze, and utilize data. IoT enables the connectivity of smart devices, allowing them to generate vast amounts of data, while Big Data provides the means to manage, analyze, and derive meaningful insights from this data. Together, they have the potential to drive innovation, improve decision-making, and create new opportunities in diverse fields, including urban development [ 59 ].

2.2.2. AI models and techniques

AI is often described as mimicking human intelligent behavior by creating computers or machines capable of its simulation. The working definition for this study describes AI as “any device/system that perceives its environment and takes actions for its goals” [ 60 ]. Broadly, an artificially intelligent machine can learn by acquiring information on the surrounding environment [ 61 ], improving performance with knowledge from experience, and performing complex tasks in a way that is similar to how humans solve problems. The capabilities of AI systems involve data analysis and learning from external data using Natural Computing (NC) and ML (e.g., Ref. [ 62 , 63 ]); emulating human cognitive functions using Computer Vision (CV), Fuzzy Logic (FL), Natural Language Processing (NLP) (e.g., Ref. [ 61 ]); and dealing with the complexities of human thinking and emotion (e.g., [ 64 ]) using decision support, strategic planning, sequential actions [ 65 ], self-learning, and self-improvement [ 66 ]. Concerning NC, it simulates natural phenomena and utilizes natural material as computational media in computers to optimize ML algorithms [ 62 ]. Evolutionary computing (EC) is also used for continuous optimization and in complex optimization problems involving many variables. It is extensively applied to optimize ML models [ 67 ]. For example, evolution and ecology as biological phenomena inspire optimization algorithms [ 68 ]. ML can be based on FL in terms of imitating human reasoning and cognition using 0 and 1 as extreme cases of truth with various intermediate degrees, characterizing the space between black-or-white scenarios, or using fuzzy c-means clustering to separate each data point into different clusters based on probability score attribution. CV applies ML to take information from visual data by recognizing patterns and making meaningful decisions based on that information. In addition, ML overlaps, intersects, or can be used as a tool for different AI models, such as CV, FL, and NLP.

For example, NLP enables computers to understand, analyze, manipulate, and generate human language. NLP plays a significant role in smart cities by enabling efficient and effective communication between humans and smart systems. NLP techniques analyze and understand human language in various forms, allowing for intelligent interactions and decision-making. In smart cities, NLP can be applied in multiple domains, such as smart planning, smart governance, smart mobility, and smart services. By leveraging NLP, smart cities can enhance communication channels, improve service delivery, and gain valuable insights from citizen feedback, leading to more responsive and citizen-centric urban environments. Tyagi and Bhushan [ 69 ] explore the potential of NLP in optimizing Information and Communication Technology (ICT) processes for building smart cities. The study analyzes the architecture, background, and scope of NLP and presents its recent applications in various domains. The authors highlight NLP's role in advancing smart cities and discuss the open challenges in its implementation, aiming to emphasize NLP as a key pillar in building smart cities.

Furthermore, AI involves many techniques that have gained traction over the past few years as part of AI and AIoT applications for environmental sustainability, climate change, and smart cities. These techniques include Artificial Neural Network (ANN), Support Vector Machine (SVM), Linear Regression (LR), Decision Trees (DT), Random Forests (RF), Adaptive Neuro-Fuzzy Inference System (ANFIS), Batch-Normalization (BN), Convolutional Neural Networks (CNNs), Deep Neural Networks (DNNs), and Genetic Algorithm (GA). As regards ML, among the supervised learning techniques used for regression, classification, or both are LR, Generalized Linear Models (LGM), DT, RF, SVM, ANN, and Bayesian Networks (BN). As to DL, it is a biological neural network or brain-inspired type of ML that uses DNN, CNNs, and Recurrent Neural Networks (RNNs) algorithms. Thus, it emulates the way humans gain certain types of knowledge by collecting, analyzing, and interpreting large amounts of data and making decisions in a faster and easier manner. DL techniques leverage neural networks comprising three fundamental layers: the input layer, hidden layers, and the output layer. These layers play a crucial role in acquiring data representation and establishing connections across multiple levels of abstraction.

2.2.3. AIoT and its system pillars: A data science cycle perspective

To manage and analyze the dynamic and relational data generated via IoT increasingly requires powerful computational and analytical capabilities. This has led to the emergence of AIoT, a technological framework that optimizes the efficiency of IoT operations, improves human-machine interactions, advances data management and analytics models, and enhances decision-making processes. AIoT involves connecting and combining IoT devices and sensors with AI models and techniques to enable advanced analysis, enhanced real-time insights, intelligent decision-making, and autonomous behavior. It acts through control and interaction to respond to the dynamic environment, a process where ML/DL has shown value in enhancing control accuracy and facilitating multimodal interactions [ 13 ]. The synergy between AI and IoT through Big Data enables smarter and more efficient applications across various domains, driving innovation and enabling transformative solutions. IoT produces Big Data, which in turn requires “AI to interpret, understand, and make decisions that provide optimal outcomes” [ 70 ] pertaining to a wide variety of practical applications for urban systems in different context of urbanism [ 1 , 71 ]. In other words, IoT enables data-driven AI analytics to optimally accomplish complex tasks and extract useful knowledge in the form of applied intelligence. The resurgence of AI is driven by the abundance and potency of Big Data, thanks to enhanced computing storage capacity and real-time data processing speed.

AIoT enables the utilization of AI to incorporate intelligence and decision-making capabilities into IoT systems and applications. The AI/AIoT-driven system consists of five pillars: (1) sensing, (2) perceiving, (3) learning, (4) visualizing, and (5) acting. This is illustrated in Fig. 2 from a conceptually generic perspective, implying that this system can be tailored to various applications depending on their characteristics, requirements, and objectives. For example, Zhang and Tao [ 13 ] present the progress of AIoT research from four perspectives: (1) perceiving, (2) learning, (3) reasoning, and (4) behaving in connection with smart transportation, smart buildings, and smart grids. This entails empowering smart things with human-like cognitive and behavioral abilities to bring them closer to reality, which is essential to system operation.

The five pillars of an AI/AIoT-driven system: 1-sensing in charge of collecting raw data, 2-perceiving in charge of extracting semantically meaningful information from raw data, 3-learning in charge of learning to predict patterns, 4-visualizing in charge of communicating key insights, and 5-acting in charge of taking action to achieve a certain goal.

An AI/AIoT-driven system is characterized by the ability to process raw data to extract useful insights to enable better decisions and/or take action. This involves different interrelated computational capabilities and processes. Machine perception is the capability of the system to use input data from sensors (e.g., vision, audio, proximity, position, tactile, photoelectric, infrared, light, and ultrasonic) to deduce different facets of the world, e.g., object detection/tracking, action recognition, image classification, semantic segmentation, language recognition, and pose estimation. There are different sensory information that provides patterns to the system for it to generate perceptions. Overall, machine perception aims to translate these data into meaningful information, thereby recognizing and interpreting these data by capturing the sensory information to relate to the real world. The acquisition of sensory data from the surrounding environment and their correct interpretation are key inputs to the learning process. The common state-of-the-art method for learning is based on ML, which allows the system to learn from experience without explicitly being programmed. Learning starts with collecting and preparing data (e.g., sensory, numbers, human pictures, object images, records, texts) to be used as training data, building the ML model to be trained on these data, supplying the data, and letting the system train itself to find patterns or make predictions. The outcome is an ML model that can be used with different sets of data and can use the data for predictive, descriptive, prognostic, and prescriptive functions. For example, the latter function involves suggesting what actions to take.

Further, there are three subcategories of ML: (1) supervised learning (trained with labeled data sets by humans to identify objects or things), (2) unsupervised/semi-unsupervised learning (finding patterns in unlabeled data such as behaviors or trends), and (3) reinforcement learning (trained on trial and error to take the best action based on the right decision such as autonomous driving or automating routine functions). Ullah et al. [ 72 ] illustrate different supervised, unsupervised, and reinforcement learning algorithms. These add to transduction learning, multitasking learning [ 73 ], federated learning [ 74 ], transfer learning [ 75 ], and few-shot learning [ 76 ]. Moreover, to enable the system to learn from the interpreted data and the performed computations to generate repeatable outputs and reliable decisions commonly requires using large datasets to train ML models, thereby, the relevance of DL, which can be leveraged to improve the learning of the system and enable it to adapt to varied situations to enhance its performance. DL has attracted increased attention and has proven useful for improving the intelligence of AIoT applications to handle dynamic and complex environments in the context of unsupervised and reinforcement learning methods [ 13 ].

ML is about leveraging data to improve performance on complex tasks by building decision-making models (e.g., Ref. [ 14 , 72 , 77 , 78 ]). Speaking of tasks, Mitchell [ 79 ] conceives of ML as “a computer program learning from experience ‘E’ with respect to some class of tasks ‘T’ and performance measure ‘P,’ if its performance at tasks in ‘T’ as measured by ‘P,’ improves with experience E.’” The workhorses of the decision-making process are ML models and algorithms, given their role in systematically extracting useful knowledge from data (patterns, correlations, predictions, forecasts, etc.) that can support decision-making. For example, by using DL algorithms, the system can conduct real-time analysis of video streams, identify objects, and detect events with absolute precision using CV models for real-time traffic monitoring or analysis of traffic conditions. The recent development of ML is associated with its ability to apply complex mathematical calculations to colossal amounts of data in a repetitive and faster manner. Accordingly, the system renders the decision-making process more data-driven, accurate, clearer, and faster thanks to ML. It makes decisions based on the perceived patterns in the data it receives. Examples of decisions in this regard may include automating a waste management process, enhancing an energy operation, optimizing a planning function, improving an environmental strategy, adjusting a policy, and evaluating risk.

As the decision-making process is based on data-driven insights, data visualization becomes important in terms of conveying complex data in such a way as to make it easier for humans to better comprehend and react to these data. Data visualization entails using specialized algorithms to generate graphical representations or visual displays of data using such elements as charts, infographics, maps, images, animations, and other metaphors. It enables decision-makers to gain and pull insights more rapidly by exploring, monitoring, and interpreting data. Examples of data visualization include city dashboards, cityScore, city metabolism, and situation rooms.

Finally, the last process of the system is acting to maximize a certain goal. To interact with the environment and humans, the system should be able to reason/make inferences and behave. Acting is associated with actuation mechanisms, which enable communication in a smarter eco-city environment and perform output functions. A wide range of actuators constitutes an integral part of the sub-systems of smarter eco-cities for operations, functions, and services. Actuation aims to execute actions to optimize different smart systems, such as power grid, building, transport, traffic, street lighting, waste management, and water distribution. The optimization occurs through adding, minimizing, adjusting, and transferring resources. In this respect, actuation is central to AIoT applications for monitoring things, controlling things, ranging things, operating things, repairing things, evaluating things, and assigning things, to name a few. Functions in smart cities “enable the actuation mechanisms to be employed directly on the IoT-enabled smart devices” [ 80 ]. For a detailed review of smart city actuators, the reader might be directed to Ref. [ 81 ]. However, developing more response systems and actuators is required to improve the engineering applications of AI and AIoT and enable their implementation in emerging smarter eco-cities concerning the performance and behavior of their physical systems.

The rationale for adopting a data science type of cycle as to the pillars of the AI/AIoT system is to emphasize the iterative and data-driven nature of the system in line with most of the synthesized studies reporting on the relationship between AIoT environmental sustainability, climate change, and smart cities. By incorporating data science principles, we seek to highlight the importance of leveraging large volumes of heterogeneous data in AIoT applications. This approach enables the extraction of meaningful insights and the development of predictive models, leading to enhanced decision-making processes and improved system performance. Highlighting the data science cycle provides a holistic AI/AIoT ecosystem perspective. It emphasizes the continuous feedback loop, where data are collected, analyzed, and fed back into the system to optimize its functionality and adaptability. This iterative process aligns with the dynamic nature of AI and AIoT applications, where data from various sources continuously flow and contribute to the intelligence and effectiveness of the overall system.

3. A review of related literature studies

In this section, we present a survey of the existing work conducted in the field of emerging smarter eco-cities and their applied AI and AIoT solutions. This survey aims to provide a comprehensive overview of the current state of research, highlighting the key findings, contributions, and trends in the field. By examining the existing literature, we aim to identify gaps and opportunities for further exploration in developing and implementing smarter eco-cities. This survey serves as a foundation for our comprehensive systematic review, enabling us to synthesize and analyze the findings in a structured and rigorous manner.

It was not until more recently that the literature on AI and AIoT applications for environmental sustainability and climate change started to grow and extend across many domains and disciplines. Several reviews have been performed on AI and AIoT in improving or advancing the different areas of environmental sustainability and climate change ( Table 1 ).

Table 1

A set of literature review studies on AI solutions for environmental sustainability and climate change.

3.1. AI for environmental sustainability and climate change

The review studies on energy conservation and renewable energy contribute to understanding various approaches and strategies for achieving energy efficiency and promoting renewable energy sources in different contexts. These studies highlight the importance of technological advancements, policy frameworks, and behavioral changes in achieving sustainable energy practices. In the field of water resources conservation, review studies contribute to the knowledge on water management practices, including efficient water use, water conservation strategies, and the impact of climate change on water resources. These studies emphasize the need for integrated water resource management and sustainable water use practices to address water scarcity and ensure long-term water sustainability. The research review on waste management demonstrates the significant contributions of AI in waste management, ranging from enhanced waste sorting to optimized collection routes, predictive maintenance, and Decision Support Systems (DSS). These advancements can improve resource efficiency, reduce environmental impact, and promote sustainable waste management practices. The review studies on biodiversity and ecosystem services provide insights into the importance of biodiversity conservation and the role of ecosystem services in sustaining human well-being. These studies emphasize the need for conservation measures, habitat restoration, and the integration of ecosystem services into decision-making processes for sustainable development. In the context of sustainable transportation, studies shed light on various aspects of sustainable transportation, including electric vehicles, intelligent transportation systems, and multimodal transportation options. These studies highlight the potential of sustainable transportation solutions in reducing CO 2 emissions, improving air quality, and enhancing urban mobility. Finally, in the domain of climate change adaptation and mitigation, review studies contribute to understanding the challenges and opportunities in addressing climate change impacts. These studies explore adaptation strategies, mitigation measures, and policy frameworks to reduce GHG emissions and build resilience to climate change. Collectively, these review studies provide valuable insights into various aspects of environmental sustainability and climate change and contribute to the knowledge base in their respective fields. They highlight the importance of adopting holistic approaches, integrating multiple disciplines, and considering technological and policy dimensions to address environmental challenges and promote sustainable practices.

3.2. AI for smart cities

Some reviews have been conducted on the link between AI, IoT, and Big Data in smart cities from a more general perspective (e.g., Ref. [ 1 , 106 , 107 ]), broadly addressing different domains beyond environmental sustainability without providing technical details. Allam and Dhunny [ 106 ] focus mainly on the role of AI in building smart cities, addressing metabolism, governance, and culture, and identifying the strengths and weaknesses of AI. The study contributes to understanding the intersection between Big Data, AI, and smart cities. It explores the potential of utilizing these technologies in the context of smart cities, highlighting their impact on various aspects, such as resource optimization, urban planning, and transportation. It provides insights into the challenges and opportunities associated with integrating Big Data and AI in smart city initiatives. Bibri et al. [ 1 ] examine the research trends and driving factors of environmentally sustainable smart cities and their converging AI, IoT, and Big Data technologies. The authors show that environmentally sustainable smart cities have experienced rapid growth during 2016–2022, driven by both the digitalization and decarbonization agenda and the rapid advancement of data-driven technologies. The study highlights the importance of addressing the environmental costs and ethical risks associated with these technologies. The findings provide insights for scholars, practitioners, and policymakers developing data-driven technology solutions and implementing environmental policies for smart cities. Navarathna and Malagi [ 107 ] focus on the role of AI in smart city analysis. The study examines the application of AI techniques in analyzing the vast amount of data generated by smart city systems. It explores how AI can enhance decision-making processes, optimize resource allocation, and improve the overall efficiency and sustainability of smart cities. It contributes to understanding how AI can be leveraged to address smart city development challenges and complexities.

3.3. AIoT: theoretical foundations and practical applications

Few review studies have been carried out on AIoT as an emerging technological area. Mukhopadhyay et al. [ 108 ] highlight the importance of sensors in IoT systems and their integration with AI. The authors emphasize the need for efficient, intelligent, and connected sensors to make smart decisions and communicate collaboratively. They also mention the emergence of advanced AI technologies that enable sensors to detect performance degradation, identify patterns, and promote innovation. The focus is on sensors, smart data processing, communication protocols, and AI to enable the deployment of AI-based sensors for future IoT applications. Shi et al. [ 11 ] focus on the convergence of IoT and AI. The study compares two AI forms, knowledge-enabled AI and data-driven AI, highlighting their respective advantages and disadvantages. It surveys recent progress in the integration of AI throughout the IoT architecture, covering the sensing, network, and application layers. Zhang and Tao [ 13 ] explore the concept of AIoT and its potential to empower IoT. The study presents a comprehensive survey on AIoT, showcasing how AI techniques, particularly DL, can enhance IoT speed, intelligence, sustainability, and safety. It discusses the AIoT architecture within the context of cloud computing, fog computing, and edge computing. It also highlights promising applications of AIoT and outlines the challenges and research opportunities in this field. Both studies contribute to the understanding of the convergence of IoT and AI, with the first study focusing on the general significance and progress in the integration and the second study specifically exploring the concept of AIoT and its implications for IoT. Mastorakis et al. [ 12 ] take a broader perspective on the convergence of AI and IoT, covering various topics related to AI methods in IoT, including research trends, industry needs, and practical implementation. Their work balances theoretical concepts and real-world applications through case studies and best practices. It serves as a comprehensive resource for researchers and practitioners interested in the integration of AI and IoT. Both contributions provide insights into the integration of AI and IoT technologies and the understanding of AIoT and its applications in specific contexts and the broader IoT landscape. Together, these studies provide valuable insights into the advancements, challenges, and potential applications of AIoT, offering a foundation for further research and development in this area.

The previous review studies on AI and environmental sustainability, AI and climate change, AI and smart cities, and AIoT have laid a strong foundation for understanding the opportunities, benefits, and challenges in creating environmentally sustainable and technologically advanced urban environments. However, with the rapid advancement of AI and AIoT technologies, there is a need to explore the specific intersection of these technologies and existing smart eco-cities and their underlying multifaceted dimensions. The emerging field of AI and AIoT presents new possibilities for tackling the complexity of ecological degradation and the climate crisis in urban areas. By conducting a comprehensive systematic review of emerging smarter eco-cities and their leading-edge AI and AIoT solutions, we can bridge the gap between these solutions and the existing research on environmental sustainability, climate change, and smart cities within the defining context of smarter eco-cities. Overall, this review is the first of its kind and seeks to bring new insights into the flourishing field of smart eco-urbanism and extend the knowledge of its diverse domains by synthesizing a plethora of studies from multiple sources and disciplines.

4. Materials and methods

A systematic literature review addressed the three questions guiding the study and, hence, achieved its specific objectives. It involves retrieving, mapping, aggregating, configuring, and critically evaluating studies published to address and discuss the research topic of smarter eco-cities as an interdisciplinary field [ 109 ]. It allows for the mining of relevant information from the continuously expanding corpus of publications [ 110 ]. As illustrated in Fig. 3 , the study consists of nine key stages: (1) research focus and scope definition, (2) literature search, (3) screening and selection, (4) data extraction, (5) critical evaluation, (6) synthesis and analysis, (7) interpretation and narration, (8) existing gaps and areas requiring further investigation, and (9) summary and manuscript preparation.

Flow diagram outlining the process of conducting the systematic review.

Regarding stages 2 and 3, we followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) approach for literature search and selection [ [111] , [112] ]. Fig. 4 shows the four-phase flowchart literature search and selection process related to this approach. Among the available pool of academic research databases, SCOPUS was selected given its broad coverage of 455 high-quality peer-reviewed studies related to the topic on focus that meets strict standards for rigor. This online platform is one of the most reliable and trustworthy academic literature sources. To retrieve the scholarly literature, we developed a broad-based search string covering the different topics of the study and the associated links. Accordingly, the search string included: “smart eco-cities,” “smart cities AND internet of things,” “smart cities AND artificial intelligence OR machine learning OR deep learning,” “smart cities AND environmental sustainability,” “environmental sustainability AND artificial intelligence OR machine learning OR deep learning,” “climate change AND artificial intelligence OR machine learning,” “artificial intelligence of things AND environmental sustainability”, “artificial intelligence of things AND smart cities,” “artificial intelligence of things AND climate change,” “artificial intelligence AND smart eco-cities,” “blockchain AND environmental sustainability,” and “blockchain AND artificial intelligence.” These were used to search against the title, abstract, and keywords of articles to produce initial insights. We then refined and narrowed the reading scope, focusing on the documents providing definitive primary information. Accordingly, titles and abstracts from these documents were screened to select those focused on the relationships between smart cities, smart eco-cities, environmental sustainability, climate change, AI and AIoT technologies, and IoT and Big Data technologies. After excluding overlaps, 230 documents remained in the database. These were checked, and 12 papers were excluded as they did not include information on the relationships in question. Afterward, we explored citation tracking or reference chaining techniques to uncover additional relevant sources. Accordingly, the reference sections of the remaining papers were checked, and 17 other relevant papers were added to the final database, which included 235 documents in total. This was considered reliable when conducting a systematic review [ 113 ].

The PRISMA flowchart for literature search and selection. Adapted from [ [111] , [112] ].

The reviewed articles were published in prominent journals and conferences in urban planning, sustainable urban development, computing, and emerging technologies. Among these outlets were “Sustainable Cities and Society,” “International Conference on Smart Sustainable Cities,” “IEEE Transactions on Sustainable Computing,” “Environmental Modeling and Software,” “Cleaner Production,” “Environment and Urban Systems,” “Renewable and Sustainable Energy Reviews,” “Applied Energy,” “Sustainability,” and “Technological Forecasting and Social Change.” These outlets showcased the relevance and significance of the research in advancing the understanding and implementation of AI and AIoT solutions in the context of smart eco-cities.

The literature search was conducted in late March 2023 and returned 455 documents covering 2015 to 2023. The starting year was selected because it marked the approval of the 2030 Agenda for Sustainable Development by the United Nations General Assembly as an international policy framework for the 17 SDGs. The full period, 2015–2023, captures the multifaceted nature of the topic of smarter eco-cities from the perspective of AI and AIoT technologies concerning environmental sustainability, climate change, and smart cities. This is determined by an earlier bibliometric study conducted by Bibri et al. [ 1 ], highlighting several urban trends and events highly relevant to the current study.

Data extraction and synthesis are crucial steps in conducting a systematic review. These steps involve extracting relevant information from the included studies and synthesizing the data to identify themes and patterns. Concerning stage 4, we developed a structured Excel spreadsheet that specifies what information needs to be extracted from these studies, following a deductive approach to content analysis. This information included study characteristics, methodological approaches, technological and sectoral domains, AI models and techniques applied in environmental sustainability and climate change, applied AI and AIoT solutions in smart cities, linkages between smart cities and smart eco-cities, use cases and applications, and knowledge gaps, among others. In terms of methodological approaches, for example, studies were qualitative based on descriptive analysis and literature review, mixed-methods, and quantitative approaches based on modeling and simulation regarding AI and AIoT. Moreover, we evaluated the quality and relevance of the selected documents through a critical appraisal process. This involved analyzing the methodologies used in these documents to assess the strengths and weaknesses of the research approaches. During data extraction, we carefully read and analyzed each included study to identify and record the relevant information.

As regards stage 5 and 6, the primary focus of this study was to identify, make, establish, and project interconnections between the different dimensions of smarter eco-cities based on the synthesized studies. The synthesized analysis involved integrating the extracted data from multiple studies to identify patterns, trends, similarities, and differences across studies to generate findings or conclusions. The synthesis approach applied seamlessly integrated configurative, aggregative, and narrative synthesis as qualitative analytical approaches. Themes were derived from the research objectives and findings from the reviewed studies. The synthesis was performed based on these themes using an integrated approach. Specific to this study, this approach attempted to strike a balance between theoretical, empirical, and practical issues. Accordingly, it included evidence from case studies, exploratory studies, observational studies, experimental studies, theory-building studies, statistical modeling studies, and review studies. The findings of the synthesized studies were merged based on a set of specified conceptual and descriptive categories. This process entailed integrating and fusing information from the multiple studies reporting on the different dimensions of emerging smarter eco-cities and related direct and indirect linkages. This categorization evolved as more precise themes were identified and revised (combined, separated, refined, or discarded). From the identified categories, themes were organized to offer new interpretations beyond the synthesized studies' findings using three different — yet complementary — approaches to synthesis ( Fig. 5 ).

A framework for unified evidence synthesis and its characteristics.

Configurative synthesis involves interpretation during the synthesis process to identify the big picture and construct the overall meaning, i.e., thematic synthesis [ 114 ]. It identifies common themes across the studies and developing a conceptual framework to explain the interconnections and variations observed in the findings. It aims to provide a deeper understanding of the research evidence by exploring the relationships and context in which the findings occur. On the other hand, aggregative synthesis involves summarizing and combining multiple research studies to produce an overall summary of the findings [ 115 ], where the interpretation is performed after the synthesis process to frame the findings, i.e., thematic summary. It aims to provide a comprehensive overview and evaluation of the research evidence. In other words, it adds and leverages evidence to make statements based on particular conceptual positions [ 114 ]. Overall, configurative synthesis goes beyond the aggregation of the research findings and focuses on understanding the underlying patterns, arrangements, or relationships among the research findings [ 116 ]. As a form of storytelling, narrative synthesis involves summarizing, explaining, and integrating the research findings from individual studies through a narrative approach. It focuses on providing a descriptive and interpretive account of the research evidence, often using textual descriptions [ 117 ]. It allows for investigating the similarities and differences between multiple studies and exploring their relationships [ 118 ]. It aims to combine diverse perspectives and findings from multiple studies to generate a coherent narrative highlighting the key concepts, themes, and implications emerging from the research. In this respect, it may also involve identifying common themes or patterns across the studies and providing an overall narrative of the findings.

To present the results, we consolidate all pertinent information concerning emerging smarter eco-cities as an interdisciplinary field. This consolidation encompasses a wide spectrum of theoretical, empirical, and practical evidence and pertains to the various dimensions of smarter eco-cities, highlighting their synergies in producing and enhancing environmental sustainability benefits. Fig. 6 provides a structured and navigable representation of the results.

An overview of the main conceptual categories identified and their relationships.

5.1. The relationship between data-driven technologies, environmental sustainability, smart cities, and smart eco-cities

5.1.1. on the early adoption of iot and big data technologies in smart cities in the field of environmental sustainability.

To become environmentally smarter and more sustainable, both smart cities and eco-cities have undergone large-scale digital transformation enabled by the convergence of IoT, Big Data, and AI technologies. This has occurred at varying degrees and in different periods, given the specific focus of these two paradigms of urbanism in terms of strategies, solutions, and policies. Accordingly, in the early 2010s, numerous studies addressed the role of ICT in tackling the challenges of environmental sustainability in the realm of smart cities in various domains, notably:

• • Power grids: to deliver energy and manage its production, consumption, and distribution to reduce costs and increase the reliability of energy supply (e.g., Ref. [ 119 , 120 ]).
• • Environmental management: to manage natural resources and related infrastructure to improve environmental sustainability (e.g., Ref. [ 119 , 121 , 122 ]).
• • Transportation management: to optimize transport efficiency and manage mobility by taking into account traffic conditions and energy usage (e.g., Ref. [ 121 , [123] , [124] , [125] ]).
• • Waste management: to collect, recycle, reuse, recover, and dispose of different types of waste (e.g., Ref. [ 126 , 127 ]).

ICT enables the implementation of smart grids, which use advanced sensors, communication networks, and data analytics to optimize energy production, consumption, and distribution. This helps reduce costs, improve energy efficiency, integrate renewable energy sources, and enhance the reliability and resilience of the power grid. Moreover, ICT tools and systems facilitate the monitoring, analyzing, and managing of natural resources and related infrastructure. With sensors, remote sensing technologies, and data analytics, environmental parameters such as air quality, water quality, and waste management can be monitored and controlled in real time, enabling more effective environmental sustainability practices. Also, ICT applications contribute to optimizing transport efficiency and managing mobility in smart cities. Intelligent transportation systems, traffic management systems, and real-time data analysis help monitor traffic conditions, optimize routes, and improve energy usage. This leads to reduced congestion, better transportation planning, and reduced energy consumption and GHG emissions. Furthermore, ICT solutions are utilized to improve waste management processes. These include systems for waste collection, recycling, and disposal and technologies for waste monitoring, sorting, and tracking. By optimizing waste management operations and promoting recycling and resource recovery, ICT enables more sustainable and efficient waste management practices. ICT has played a vital role in transforming cities into smart and sustainable environments by enabling key advancements to achieve greater environmental sustainability and resilience in urban areas.

Concurrently, smart cities started to focus on embedding the next–generation of ICT into everyday objects and city structures and systems as part of the early deployment of IoT (e.g., Ref. [ 128 , 129 ]), paving the way for merging digital technologies with urban infrastructures and coordinating and integrating these through digital instrumentation and hyper-connectivity. One of the comprehensive theoretical and empirical studies on smart cities and IoT and Big Data conducted by Batty et al. [ 128 ] defined some goals concerning the development of a new understanding of environmental issues and the identification of critical problems relating to transport, energy, mobility, risks, and hazards. The authors additionally identified some challenges in using management, control, and optimization processes to connect smart city infrastructures to their operational functioning and planning.

During the period 2012–2015, smart cities gained traction as a model for sustainable urban development, with great potential to improve environmental sustainability based on IoT and Big Data Technologies (e.g., Ref. [ [130] , [131] , [132] , [133] , [134] , [135] , [136] ]). This traction stimulated a debate on how innovative data-driven IoT technologies could efficiently manage natural resources and mitigate environmental impacts in response to the rapid pace of urbanization and its potential effects on jeopardizing the sustainability of smart cities. Consequently, IoT and Big Data technologies gained further momentum in the pursuit of environmental sustainability across the different domains of smart cities, especially transport, mobility, energy, waste, pollution control, air quality, and planning (e.g., Ref. [ [137] , [138] , [139] , [140] ]). Subsequently, they became essential to the functioning of smart cities (e.g., Ref. [ 40 , [141] , [142] , [143] , [144] , [145] ]). This was manifested in the processes and practices of smart cities becoming “highly responsive to a form of data-driven urbanism” [ 146 ]. IoT and Big Data technologies provide the ability to monitor urban operations, functions, and structures using advanced forms of decision-making in urban intelligence functions for design and planning to improve environmental sustainability.

5.1.2. The influence of the IoT and Big Data Technologies of smart cities on the materialization of smart eco-cities

It is until around 2014 that smart cities started to have a significant impact on eco-cities for environmental sustainability (e.g., Ref. [ 57 , [147] , [148] , [149] , [150] , [151] ) thanks to the adoption of IoT and Big Data technologies as advanced forms of ICT. [ 54 ]) traces the evolution of smart cities and eco-cities over the past two decades regarding how their conceptual trajectories have converged under “smart eco-cities” from the mid-2010s onwards. The author highlights how this new paradigm of smart, sustainable urbanism is set to leverage the potential of IoT, Big Data, and digital infrastructures to integrate urban and green visions and policies. Smart eco-cities became widespread around 2016/2017 (e.g., Ref. [ 35 , 152 ]) as “a potential niche where environmental and economic reforms can be tested and introduced in areas which are both spatially proximate … and in an international context … through networks of knowledge, technology and policy transfer and learning” ([ 56 ], p. 1). Ever since, IoT and Big Data technologies and their applied solutions have become instrumental in the functioning of smart eco-cities concerning transport, mobility, energy, waste, pollution control, air quality, and planning.

To expand on the relationship between smart cities and eco-cities from an empirical perspective, Bibri and Krogstie [ 53 ] examine and compare the eco-city of Stockholm and the smart city of Barcelona, focusing on the innovative potential of IoT and Big Data technologies for advancing the goals of environmental sustainability. The authors show that smart grids, smart meters, smart buildings, smart environmental monitoring, and smart urban metabolism are the main data-driven IoT solutions adopted to enhance the performance of data-driven smart cities and smart-eco-cities under what they term “environmentally data-driven smart sustainable cities.” They also demonstrate the clear synergy between the eco-city and smart city solutions as to producing “combined effects greater than the sum of their separate effects — concerning energy efficiency and conservation improvement, environmental pollution reduction, renewable energy adoption, and real-time feedback on energy and material flows.” Pasichnyi et al. [ 153 ] propose, as part of case study research, a novel data-driven smart approach to the strategic planning of retrofitting building energy that allows a holistic city-level analysis and assesses change in total energy demand from large-scale retrofitting. Similar to Stockholm City [ 53 ], the energy transition model of Barcelona as one of the leading smart cities in Europe, aims to produce a 100 % certified renewable energy supply plan through smart energy [ 154 ]. Many other eco-cities, mainly from Europe and China, have adopted IoT and Big Data technologies in the domains of energy, transport, waste, water, and planning (e.g., Refs. [ 35 , 54 , [54] , [55] , [56] ]). A recent comprehensive state-of-the-art review conducted by Bibri [ 27 ] on smart eco-cities reveals that the newly planned and ongoing eco-city project developments are increasingly trialing innovative smart technologies to improve several aspects of environmental sustainability and climate change. This entails leveraging the advantages of eco-cities and smart cities and capitalizing on the synergies between their approaches and solutions, ultimately empowering eco-cities to enhance their environmental performance. Besides, IoT and Big Data technologies will fundamentally and irrevocably transform the landscape of eco-urbanism in terms of how eco-cities will be monitored, understood, analyzed, managed, planned, and governed.

5.1.3. The influence of policy instruments and government initiatives on the materialization of smart eco-cities

The materialization of smart eco-cities is strongly influenced by, in addition to technological advancements and environmental concerns, policy instruments. Technological advancements provide the tools and capabilities for these cities to integrate and optimize various urban systems for sustainability. Environmental concerns drive the need for these cities to adopt smart and eco-friendly solutions. However, policy instruments play a crucial role in shaping and guiding the development of smart eco-cities. Policy frameworks, regulations, and incentives create an enabling environment for implementing sustainable practices and innovative technologies and ensure the effective integration of smart solutions into urban environments. The effective use of policy instruments is essential in harnessing the full potential of technological advancements and addressing environmental challenges to realize the vision of smart eco-cities. Indeed, governance and smart eco-cities are deeply intertwined in a self-reinforcing relationship. To put it differently, governance is at the core of smart eco-cities (e.g., Refs. [ 35 , 53 , 56 , 57 , 155 ]), and its key function is to make and implement policy. One of the key roles of urban policy — as a set of plans, laws, rules, regulations, and actions — lies in aligning and mobilizing the stakeholders involved in the governance of smart eco-cities. Smart eco-cities require targeted policies to drive progress in various environmental sustainability areas and effectively implement innovative data-driven solutions. Overall, policy instruments play a crucial role in the materialization of smart eco-cities by providing a framework for planning, implementing, and regulating various initiatives. Among the policy instruments that facilitate the materialization of smart eco-cities [ 35 , 55 , 56 , [154] , [155] , [156] ] are:

• • Government regulations: The government enacts regulations that mandate specific sustainability standards and requirements for smart eco-city development, e.g., setting energy efficiency targets for buildings, enforcing waste management practices, and promoting renewable energy integration.
• • Financial incentives: Governments and local authorities provide financial incentives to encourage the adoption of smart technologies and sustainable practices, including tax incentives, grants, and subsidies for smart eco-city projects that incorporate smart solutions.
• • Public-private partnerships: Governments can partner with technology and energy companies, research institutions, and industry stakeholders to develop and implement sustainable solutions. These partnerships leverage expertise, resources, and funding to implement smart applications successfully.
• • Open data initiatives: Governments can promote sharing data collected from various sources, such as sensors and IoT devices, to foster innovation and facilitate evidence-based decision-making. These initiatives enable researchers, businesses, and policymakers to access and analyze information to develop smart solutions and monitor the performance of smart eco-city projects.
• • Standards and certification programs: Establishment of these programs ensures the interoperability, reliability, and safety of smart solutions in smart eco-cities. Standards organizations and certification bodies can define technical requirements, data privacy guidelines, and cybersecurity protocols to promote trust and facilitate the adoption of advanced technologies in smart eco-cities.

These policy instruments create a supportive environment for the materialization of smart eco-cities. They provide guidance, incentives, and regulations that drive sustainable development, promote technological innovation, and enhance the overall quality of life for residents. As revealed by Joss and Cowley [ 304 ], based on a comparative case study analysis, policy is found to exercise a strong shaping role in what sustainable development for cities is understood to be, which helps explain the considerable differences in priorities and approaches across countries.

5.2. The rise of AI and AIoT in environmental sustainability, climate change, and smart cities: solutions, use cases, and applications

This subsection is concerned with the solutions, applications, and use cases in the realm of AI and AIoT technologies and innovations. In this context, a solution represents a comprehensive approach encompassing software, hardware, processes, and strategies to solve a particular problem. It is implemented in real-world use cases, which depict practical scenarios illustrating how data-driven technologies work in action. On the other hand, an application refers to specific software designed to perform tasks, often utilizing both AI and AIoT models and algorithms. While these terms can overlap, they offer distinct perspectives: applications focus on software functionality, solutions provide holistic problem-solving approaches, and use cases offer practical insights into real-world implementations, collectively advancing data-driven technologies and innovations.

The use and application of AI and AIoT in environmental sustainability, climate change, and smart cities have dramatically increased from 2016 onward. This is seen as a sequel to the wide adoption of IoT and Big Data technologies and solutions in the various domains of smart cities for advancing environmental sustainability, as documented by many studies (e.g., Ref. [ [137] , [138] , [139] , [140] , [143] , [157] ]). As presented below, the empirical, theoretical, and literature research on AI and AIoT solutions, applications, and use cases involve the different areas of environmental sustainability, climate change, and smart cities. These areas are organized into two main periods: 2016–2019 and 2020–2023, which were identified based on an earlier bibliometric study conducted by Bibri et al. [ 1 ]. This study explores the key research trends and driving factors behind the emergence of environmentally sustainable smart cities and maps their thematic evolution over time. It demonstrates the rapidly growing trend of this emerging paradigm of urbanism that markedly escalated during 2016–2022 due to the accelerated digitalization and decarbonization agendas — due to COVID-19 and the rapid advancement of data-driven technologies. Accordingly, the two periods were derived based on the accelerated digitalization of smart cities prompted by COVID-19 in early 2020 and what this entails in terms of harnessing digital technologies for addressing environmental issues (e.g., Ref. [ [158] , [159] , [160] ]), thereby the relevance of subdividing the full period into two distinct sub-periods. Also, several other studies in early 2020, as mentioned earlier, emphasized the increasing recognition of the complexity of environment degradation and climate change challenges and the growing need for more innovative, advanced, and immediate solutions based on emerging data-driven technologies to tackle them. As to the starting year of 2016, it has marked the materialization of smart eco-cities (e.g., Ref. [ 27 , 35 , 152 ]), as discussed in Section 4 .

5.2.1. The first period: 2016–2019

5.2.1.1. environmental sustainability, empirical ai research.

The following empirical studies have significantly contributed to the advancement of emerging smarter eco-cities by applying AI and ML models and techniques to various aspects of environmental sustainability.

• • Water resources conservation: ML techniques, such as ANN, SVM, FL, ANFIS (an ANN-based on FL), LR, and Key-Nearest Neighbors (KNN), have been applied to predict stream flow and examine water quality parameters (e.g., Ref. [ 161 , 162 ]). Some studies have adopted ML techniques and DSS, such as ANN, DT, GA, FL, and ANFIS, to analyze water chemistry and assess water quality (e.g., Ref. [ 163 , 164 ]). These models and techniques have also been applied to hydro-meteorological forecasting and leak detection [ [165] , [166] , [167] ].
• • Energy conservation and renewable energy: Energy-related studies have harnessed ML and DSS with techniques such as ANN, FL, SVM, DT, Evolution Strategies (ES), Evolutionary Computing (EC), BN (an algorithmic method that makes the training of DNN faster and more stable), and ANFIS for tasks including energy operation, production, distribution, maintenance, and planning (e.g., Ref. [ 92 , 161 , [168] , [169] , [170] ]).
• • Sustainable transportation: ML has been applied to traffic forecasting using ANN, DT, and time series models [ 171 , 172 ]. Additionally, transport-related studies have employed ML and DSS with ANN, DT, NC, FL, SVM, LR, and time series models (e.g., Ref. [ 173 , 174 ]).
• • Biodiversity conservation and ecosystem services: ML techniques, including FL, GA, ARIES, and BN (an algorithmic method used for modeling networks in ecosystems), have been used to address different aspects of biodiversity conservation and assess ecosystem services (e.g., Ref. [ 98 , [175] , [176] , [177] , [178] , [179] ]).

The insights gained from the studies conducted on water resources conservation can inform effective water resources management strategies, aiding in the conservation and sustainable use of water in existing smart eco-cities. By utilizing AI algorithms, energy conservation and renewable energy studies have contributed to energy conservation efforts and the integration of renewable energy sources within existing smart eco-cities. As to sustainable transportation, related studies have provided valuable insights for better transportation planning and management, facilitating more efficient and sustainable mobility within existing smart eco-cities. The research on biodiversity conservation and ecosystem services has contributed to a deeper understanding of the relationships between biodiversity, ecosystem services, and the sustainability of existing smart eco-cities. The findings and insights from these studies can inform evidence-based decision-making and promote sustainable practices in smart eco-city development.

IoT's role within the AIoT framework

IoT, as a fundamental component of AIoT, significantly contributes to the above areas of environmental sustainability by providing a robust technical framework for data collection, analysis, and decision-making.

• • Data acquisition: IoT devices equipped with sensors and actuators collect real-time data on environmental parameters. This extensive data acquisition network forms the foundation for environmental monitoring.
• • Data transmission: IoT enables seamless data transmission to centralized platforms, such as cloud computing, ensuring that data from distributed sources is readily available for processing and analysis. Connectivity protocols ensure efficient and secure data transfer.
• • Big data handling: IoT generates vast amounts of data and AI algorithms, including ML and DL, process these data to identify patterns, anomalies, and trends, providing insights into environmental conditions.
• • Predictive analytics: AI-driven predictive models utilize historical and real-time IoT-driven data to forecast environmental changes, which enables proactive decision-making and timely responses.
• • Automation and control: IoT's integration with AI allows for autonomous control of systems and processes. For instance, smart grids can adjust energy distribution based on real-time demand and renewable energy availability, promoting energy efficiency and conservation.
• • Resource optimization: AI algorithms optimize resource allocation and utilization, ensuring minimal waste and maximum efficiency in such areas as energy consumption, water usage, and transportation.
• • Remote monitoring: IoT-enabled devices can be remotely controlled and monitored, reducing the need for physical interventions and minimizing human impact on sensitive ecosystems.

Overall, IoT's technical capabilities within the AIoT framework enable comprehensive data collection, analysis, and automation, fostering environmentally sustainable practices and informed decision-making across various domains.

Theoretical and literature AI research

Similarly, a range of theoretical and literature review studies have made notable contributions in the context of emerging smarter eco-cities across various domains:

• • Water resources conservation: Several studies have focused on water resources conservation by applying ML and DSS, employing techniques such as ANN, SVM, FL, GA, and ANFIS. Key studies in this area include those by Mehr et al. [ 180 ], Oyebode and Stretch [ 90 ], Sahoo et al. [ 181 ], and Valizadeh et al. [ 182 ].
• • Energy conservation and renewable energy: Efforts to conserve energy and promote renewable sources have been supported by ML techniques such as LR, ANN, SVM, NC, and EC. Researchers and scholars like Akhter et al. [ 82 ], Alsadi and Khatib [ 183 ], Das et al. [ 83 ], Dawoud, Lin, and Okba [ 184 ], Khan et al. [ 84 ], Youssef, El-Telbany, and Zekry [ 185 ], and Wang and Srinivasan [ 186 ] have made contributions in this field of study.
• • Sustainable transportation: Related initiatives have been advanced by applying ML, CV, and DSS utilizing ANN, DT, NC, SVM, FL, and time series models. Key studies in this area include those by Jiang and Zhang [ 187 ] and Liyanage et al. [ 101 ].
• • Biodiversity and ecosystem services: Biodiversity-related studies have explored modeling competition and population dynamics, often employing ML techniques like cellular automata [ 188 ]. Additionally, species conservation efforts have benefited from ML and DSS involving SVM, ANN, GA, and FL, as demonstrated by the work of Salcedo-Sanz, Cuadra, and Vermeij [ 97 ]. On the other hand, ecosystem services assessment has leveraged ML methodologies such as ARIES, as exemplified in the research by Ochoa and Urbina-Cardona [ 189 ].

The research on water resource conservation has provided insights into effective water resource management strategies. Regarding energy conservation and renewable energy, the studies in this area have contributed to optimizing energy operations, promoting energy efficiency, and fostering the adoption of renewable energy technologies. The research on sustainable transportation has provided insights into traffic management, transportation planning, and improving the overall efficiency and sustainability of transportation systems. Concerning the studies in biodiversity conservation they contribute to our understanding of biodiversity and the sustainable management of ecosystems. Overall, these theoretical and literature review studies collectively contribute to the advancement of smarter eco-cities, and their outcomes provide insights into the development of evidence-based approaches for building more sustainable and resilient smart eco-cities.

5.2.1.2. Climate change: empirical AI research

In empirical AI research on climate change, notable attention has been directed toward specific areas, including scenario analysis, marine resources management, and disaster management and resilience. This focus has yielded significant contributions to advancing smarter eco-cities:

• • Scenario analysis: Studies have employed various AI and ML models and techniques. Noteworthy works have performed scenario analysis based on ML, EC, and FL using ANN, BN, SVM, GA, and neuro-fuzzy (e.g., Ref. [ [190] , [191] , [192] ]); analysis across sustainability elements based on ML using ANN (e.g., Ref. [ 193 ]); and CO2 emission [ [194] , [195] , [196] ] and natural disaster (e.g. Ref. [ 197 , 198 ], based on ML and FL using ANN, SVM, EC, and neuro-fuzzy. Studies on ocean and cryosphere (e.g., Ref. [ 199 ]) and atmospheric forecasting (e.g., Ref. [ 200 , 201 ]) have applied ML using ANN. Furthermore, scenario analysis focuses on developing and analyzing different scenarios of future climate conditions based on various parameters, data inputs, and assumptions. It considers a range of potential future outcomes for climate change, including different levels of GHG emissions, temperature increases, sea-level rise, and other climate-related factors. In scenario analysis, various modeling techniques are used, including Generative Adversarial Networks (GANs), DL (CNNs and RNNs) and NLP, to create and analyze these scenarios. The goal is to provide insights into the potential consequences of different climate pathways and to help inform decision-making and policy development. Scenario analysis is valuable for understanding the range of possible future outcomes and their associated risks and impacts. It allows stakeholders to consider various “what-if” scenarios and plan accordingly for a range of potential climate-related challenges.
• • Disaster management and resilience: The domain of disaster resilience has witnessed significant AI-driven contributions in prediction and forecasting and resilient infrastructure and urban planning. The studies carried out by Cheng and Hoang [ 305 ], Choubin et al. [ 202 ], and Ji et al. [ 203 ] have advanced the field by employing different AI models and techniques. Among these are ML in predictive modeling, pattern recognition, and damage assessment; NLP in sentiment analysis and information extraction; CV in image analysis and object recognition; AI-driven Geographic Information Systems (GIS) in spatial analysis and mapping and visualization; reinforcement learning in resource allocation optimization; and DSS in dynamic resource allocation.
• • Marine resources management: AI research has played a pivotal role in marine resources management, encompassing water pollution monitoring, pollutant tracing in water quality, pollution reduction and prevention strategies, acidification mitigation, and habitat and species protection. These endeavors have harnessed various AI models and techniques, as demonstrated by the works of Lu et al. [ 204 ] and Wang et al. [ 205 ], to address challenges related to the sustainable use and conservation of marine ecosystems. Some of these models and techniques include ML, DL (e.g., CNNs and RNNs), GA, and ML-based Species Distribution Models (SDMs), NLP, time series forecasting, Autonomous Underwater Vehicles (AUVs) and Remotely Operated Vehicles (ROVs), and DSS.

This body of research underscores the critical role of AI in addressing climate change challenges, highlighting its potential to inform smarter eco-city initiatives and promote sustainable urban development. Specifically, the research on scenario analysis has provided insights into climate change impacts, vulnerability, and adaptation strategies; the interdependencies and interactions between different aspects of sustainability; understanding and managing CO2 emissions; and enabling the development of strategies for reducing carbon footprints. Regarding natural disaster analysis, the key insights gained pertain to predicting and forecasting natural disasters, resilient infrastructure planning, and environmental hazard detection. Disaster resilience studies have enhanced disaster resilience, enabling effective preparedness, response, and recovery measures. The studies on marine resources management have provided new perspectives on tackling different relevant challenges and enhancing the sustainable management of marine resources. Overall, these empirical studies provide valuable insights and advancements in understanding climate change challenges in the context of emerging smarter eco-cities.

5.2.1.3. Smart cities

During the first period, AI research on environmental sustainability was expanding to include smart cities based on various AI models, including ML, DL, CV, NLP, and robotics (e.g., Ref. [ 107 , [206] , [207] , [208] , [209] ]). It also involved using ML in empirical studies (e.g., Ref. [ 174 , 210 ]). Concerning climate change in the urban context, some studies addressed mitigation concerning urban planning, mobility, and land use (e.g., Ref. [ 211 , 212 ]). For AI-powered IoT architecture, research tended to focus mainly on the theoretical aspects of ML and data analysis (e.g., Ref. [ [213] , [214] , [215] ]), cognitive AI (e.g., Ref. [ 216 , 217 ]), knowledge-based DSS [ 218 ], and knowledge-based AI [ 174 ]. Regarding the latter, expert systems are associated with the capability of AI to solve complex problems, e.g., assessment of climate change impacts, and to aid decision-making by relying on specific knowledge derived from databases [ 17 ]. AIoT has been deployed in multiple ways to help users efficiently manage energy to reduce cost as well as energy producers optimize their equipment for better service delivery [ 219 ]. It also found important applications in vehicles and transportation, especially self-driving or autonomous vehicles. These are embedded with several sensing instruments (e.g., cameras, Light Detection and Ranging (LIDAR), and radar) and thus generate massive amounts of data [ 220 ]. LIDAR is a remote sensing technology that uses laser light to measure distances and generate precise, three-dimensional representations of objects as well as landscapes.

5.2.2. The second period: 2020–2023

During the second period, 2020–2023, the empirical, theoretical, and literature research on AI increased and expanded across environmental sustainability, climate change, and smart cities. This increase and expansion pertain to energy conservation and renewable energy, water resources conservation, sustainable transport and mobility, biodiversity conservation, pollution control, climate adaptation and mitigation, and disaster management and resilience. The rapidly growing body of research in environmental sustainability, climate change, and smart cities has made further valuable contributions and advancements in the context of emerging smarter eco-cities, enhancing and extending the knowledge gained for the strategies and solutions for addressing and overcoming environmental challenges.

5.2.2.1. Environmental sustainability

While the main areas of AI and AIoT research in environmental sustainability continued to attract attention, academic interest slightly decreased compared to the previous period. This is likely attributed to COVID-19 taking priority in research during the second period, inducing scholars in smart cities and environmental sustainability to investigate the link between COVID-19 and CO 2 emissions reduction and air quality improvement. Worth pointing out is that, during 2020–2021, COVID-19-related publications received 20 % of all citations, and 98 of the 100 most-cited publications were associated with COVID-19 [ 1 ]. There was a shift during this period in academic interest from focusing on environmental sustainability challenges toward focusing on the massive deployment and implementation of digital technologies and applied solutions offered by smart cities. However, the areas of environmental sustainability that continued to attract AI research during COVID-19 include:

• • Energy conservation and air quality efforts encompass pollution reduction and prevention, pollution monitoring, pollutant filtering and capture, air quality prediction, and early hazard warning, as well as the promotion of clean and renewable energy sources (e.g., Ref. [ 21 , 86 , 87 , 99 , [221] , [222] , [223] , [224] ]) and environmental quality control [ 225 ].
• • Sustainable transportation initiatives include the evaluation of energy consumption in household transportation through ML models [ 226 ], energy planning, transportation connectivity, urban traffic management, assessment of transport network capacity, urban traffic surveillance, and optimization of commuting corridors and jobs-housing balance using techniques, such as GA, EC, ANN, Spatial DNA, and reinforcement learning (e.g., Ref. [ 17 , 99 , [227] , [228] , [229] , [230] ]).
• • Clean water security and water resource conservation efforts involve various aspects such as water quality management, water supply quantity optimization, water control, water treatment, and sanitation (e.g., Ref. [ 93 , 93 , 231 , 232 ]). Several AI models and techniques have been applied to address challenges related to clean water security and water resource conservation, including ML, DL, ANN, CNNs, RNNs, GA, Particle Swarm Optimization (PSO), NLP, and DSS.
• • Biodiversity and conservation endeavors revolve around enhancing and protecting natural capital, preserving ecosystem health, safeguarding habitats, restoring ecosystems, maintaining forest landscape visual quality, protecting species, conserving biological diversity, preventing marine pollution, and ensuring the preservation of marine resources (e.g., Ref. [ 99 , [233] , [234] , [235] , [236] , [237] , 238 ]). Among the AI models and techniques being used in biodiversity and conservation efforts are ML, DL, FL, SVM, CNNs RNNs, GA, NLP, ARIES, AI-driven GIS, and AI-powered drones.

In the pursuit of creating more intelligent and sustainable urban environments, a wealth of research has emerged to address the unique challenges emerging smarter eco-cities face. These studies, spanning diverse domains, have made further contributions to shaping the trajectory of these cities. The studies on energy conservation and air quality studies significantly enhance their environmental performance and efficiency. The research in sustainable transportation research plays a crucial role in promoting eco-friendly and efficient urban transport systems. Studies on clean water security and resource conservation are pivotal for ensuring sustainable water resource management. The research on biodiversity and conservation is instrumental in supporting the development of eco-friendly and sustainable environments. These advancements can reshape the landscape of urban development models beyond smarter eco-cities in response to the growing wave of urbanization, making cities more intelligent and environmentally conscious.

5.2.2.2. Climate change

Many more studies were conducted during the second period, indicating increased scholarly interest in climate change and its relation to AI and AIoT. The digitalized transformation triggered by COVID-19 significantly contributed to climate actions [ 158 , 159 ]. Consequently, AI and AIoT technologies gained strong traction because of the accelerated digitalization prompted by COVID-19.

Mitigation and adaptation

In the face of escalating environmental challenges, the exploration of innovative technologies has become paramount to addressing the urgent concerns of climate change. This narrative dives into the realm of climate change mitigation and adaptation, shedding light on the pivotal role that AI and AIoT play in reshaping sustainable urban development in this regard. Climate change has accelerated the need for proactive measures, particularly in urban areas where high energy consumption contributes significantly to GHG emissions. Extensive research conducted during the second period highlights the urgency of addressing climate change as cities contribute significantly to CO 2 emissions through high energy consumption. It emphasizes the potential of AI to mitigate climate change by integrating knowledge, design strategies, and innovative technologies. It further discusses AI applications in transportation, urban energy, water use, and waste management, showcasing their impact on reshaping urban planning and design. Overall, it demonstrates that implementing AI-driven solutions can improve the sustainability of future cities and contribute to climate change mitigation efforts. Ivanova, Ivanova, and Medarov [ 239 ] acknowledge the growing influence of AI across various domains and predict its continued expansion in the coming decades. The authors emphasize the prevalence of narrow AI, specialized neural networks designed to solve specific problems, in technical fields. They focus on applying narrow AI to investigate the impact of climate change on transport infrastructure, providing guidelines for data collection and AI modeling. They highlight the controllability and capabilities of narrow AI, underscoring their potential for studying climate change's effects on transportation systems.

Furthermore, in connection with the first period, one of the areas that received more focus in the theoretical and empirical research on AI applications for climate change is disaster resilience and management, including early warning systems, resilience and planning, and simulation and prediction (e.g., Ref. [ 19 , 240 , 241 ]). However, Leal Filho [ 18 ] report on— in a systematic review and survey questionnaire — all studies conducted during 2020–2022 on the relationship between AI and climate change and its opportunities for adaptation and mitigation, covering several areas that have attracted research on AI applications. Among the themes studied by the authors in relevance to the current study while expanding on those mentioned in the first period are:

• • Large-scale urbanization impacts under climate change scenarios.
• • Eco-services and tradeoffs model valuation for ecosystem services quantification.
• • Water utilization management using and combining Blockchain and AI.
• • AI for disaster response, digital response, and disaster management.
• • IoT-Based smart tree management.
• • AI and ML for wildfire evacuations, wildfire prediction and prevention, wildfire susceptibility mapping, human-caused wildfire occurrence, risk-reduction strategies for floods and droughts, conservation planning under climate-changing patterns, and green-roof irrigation optimization.
• • ML for flood prediction and protection.
• • AI for improving resilience and preparedness against flood events impact.
• • ANN for drought tolerance determination.
• • Evolutionary Neural Network (ENN) for forecasting carbon emissions, energy demand, and wind generation.
• • ANFIS for modeling climate change impact on wind power resources.
• • DL for modeling sub-grid processes in climate models.
• • ML for water security improvement and water demand forecasting in cities.
• • ML for adaptation policy.
• • AI for sustainable development.

Some of these themes are linked to smart cities and smart eco-cities for renewable energy, water, biodiversity, climate, planning, and policy within the framework of SDG 11: Sustainable Cities and Communities, SDG 9: Clean and Affordable Energy, and SDG 13: Climate Action. Moreover, AI can be a critical change agent because it enables climate change mitigation through carbon neutrality in energy production, distribution, transportation, buildings, construction, and others [ 242 ]. Some studies proposed benchmark datasets with additional modeling components for better climate change prediction [ 243 ]. For example, Samadi [ 20 ] notes that the convergence of AI and IoT has the potential to accurately predict floods and accelerate the convergence of AI models and techniques to advance flood analytics research. The author discusses the workflow of an AIoT prototype, namely Flood Analytics Information System (FAIS), which integrates ML, NLP, CNNs, and others.

In the context of climate change mitigation and adaptation, IoT as a core component of AIoT brings critical technical aspects to the forefront. As to data sensing and collection, IoT devices equipped with sensors monitor relevant environmental parameters. This continuous data stream forms the basis for understanding climate change patterns and trends. In terms of network connectivity, IoT ensures seamless data transmission and communication between devices and central data repositories. Robust communication protocols and networks, such as 5G/6G, facilitate rapid data sharing. Concerning data analysis, AI algorithms process vast datasets from IoT sensors, identifying climate change patterns and trends. ML/DL models and algorithms enable predictive analytics for anticipating changes and their potential impacts. IoT-connected weather and climate monitoring stations, coupled with AI-driven forecasting models, enable the development of early warning systems for extreme weather events and natural disasters. Moreover, IoT devices track energy consumption in real-time as part of climate change mitigation approaches. AI algorithms analyze these data to optimize energy use, identify areas for conservation, and promote the integration of renewable energy sources to reduce environmental impacts. Also, IoT-enabled infrastructure, such as smart buildings and resilient urban planning, enhances adaptation efforts. In this regard, sensors monitor structural integrity and climate-related risks, facilitating adaptive responses. Furthermore, IoT sensors in ecosystems track changes in flora and fauna behavior and health. AI algorithms aid in assessing the impact of climate change on biodiversity and guiding conservation efforts. Additionally, IoT-connected satellites and drones equipped with AI-enabled remote sensing technology provide critical data for monitoring various environmental changes. Lastly, regarding real-time decision support: IoT systems provide real-time climate data and actionable insights to decision-makers, allowing for adaptive strategies and informed policy development. In sum, IoT's technical capabilities within AIoT are instrumental in climate change mitigation and adaptation efforts. They enable data-driven decision-making, resource optimization, and resilience-building, all crucial components in addressing the challenges posed by climate change.

5.2.2.3. Smart cities: transportation, energy, waste, environmental management, and the SDGs

Compared to the first period, AI, AIoT, and Blockchain technologies have proliferated and expanded, attracting more research interest, especially with their applications in smart cities. This also implies that smart cities are embracing AI and AIoT solutions developed initially for the different areas of environmental sustainability and climate change as separate fields.

AI and AIoT applications

AI and AIoT applications span many domains of smart cities in the field of environmental sustainability. In a systematic literature review on smart cities and AI, Yigitcanlar et al. [ 244 ] found that the key contributions of AI to environmental sustainability areas include:

• • optimizing energy production and consumption via domotics (home automation),
• • predicting the risks of climate change via ML algorithms and climate models,
• • monitoring changes in the natural environment via remote sensing with autonomous drones, and
• • operationalizing transport systems via mobility-as-a-service (MaaS).

AI applications related to the latter theme also include the management of transport systems of cities in terms of shared autonomous mobility-on-demand [ 245 ], autonomous cities [ 246 ], and autonomous vehicles (e.g., Ref. [ [247] , [248] , [249] ]). Concerning the latter, the deployment of AIoT at the network edge and secure trust models offer potential solutions for the real-time processing of sensor data to enable fast response to complex scenarios, such as obstacle avoidance and velocity adaptation [ 10 ]. Moreover, Zhang and Tao [ 13 ] synthesize several studies on the application of AIoT using DL in smart transportation (e.g., traffic participants, traffic infrastructures, connected logistics, and in-car driver behavior monitoring) as well as smart grids (grid fault diagnosis, building management and optimization, load monitoring and scheduling, and cyber-attack detection). To solve the load forecasting problem in energy management, Han et al. [ 250 ] propose a DL–based framework to predict future energy consumption in smart residential homes and industries, where the IoT network is connected to smart grids to maintain energy demand and supply activities effectively. The experimental results demonstrate the ability of the approach to predict energy consumption with high accuracy. El Himer et al. [ 251 ] address the role of AIoT in providing new opportunities in distributed energy resources (DER), focusing on AIoT applications in renewable energy sources, such as solar and wind. An AIoT system developed by Puri et al. [ 16 ] generates energy from different sensors, such as piezoelectric sensors, including from stress caused by human body weight, heat generated by the movement of the human body, and sunlight. The authors built and validated the data collected from the sensors with ANN and ANFIS models to predict generated power output and demonstrated that their system could produce accurate results in predicting the power generated from renewable resources.

Broadly, AIoT can be used in smart cities to analyze and track how different consumers and residents use energy to make decisions on where and what kind of renewable energy sources could be used, as well as where energy is being wasted and how it can be directed for other uses or conserved. Sleem and Elhenawy [ 252 ] discuss the contribution of AIoT to the development of smart buildings and their functionality, as well as its benefits for reducing energy consumption and costs, improving occupant comfort and productivity, and increasing safety and security. The authors also address the challenges associated with deploying AIoT and emphasize the potential of AIoT-empowered smart buildings to contribute to sustainable urban development and improve the quality of life. Furthermore, AIoT applications are converging in smart cities. Seng et al. [ 10 ] review and discuss several dimensions of AIoT applications. These are more relevant to this study — energy and smart grids, industry and smart buildings, vehicles and smart transportation, and robotics and computer vision.

Furthermore, AI and AIoT have been instrumental in developing advanced waste collection systems that optimize several parameters and maximize efficiency. Fang et al. [ 94 ] provide a comprehensive review of the application of AI in waste management, including waste-to-energy, smart bins, waste-sorting robots, waste generation models, waste monitoring and tracking, waste logistics, waste disposal, waste resource recovery, waste process efficiency, waste cost savings, and improving public health. The authors highlight the benefits of AI in waste logistics in terms of reducing transportation distance and time savings, as well as improving waste pyrolysis, carbon emission estimation, and energy conversion. They also emphasize the role of AI in increasing efficiency and reducing waste identification and sorting costs in smart cities. Nasir and Aziz Al-Talib [ 95 ] discuss the challenges in waste classification and the potential of AI and image processing techniques to address them. They acknowledge the limitations of current waste classification models driven by DL and highlight the need for improvements in accuracy and runtime to achieve precise results. They argue that accurate waste classification is crucial for multiple reasons, including enabling recycling and resource recovery, safeguarding the environment and human health, and minimizing waste management costs. The core idea distilled from the study is that waste is the byproduct of various human activities, encompassing domestic, agricultural, and industrial sectors. Different types of waste exist, including non-biodegradable, hazardous, industrial, municipal solid, and agricultural waste. Solid waste can take hundreds of years to decompose, posing environmental risks. Mounaded et al. [ 96 ] focus on applying AI techniques in municipal solid waste (MSW) management. They emphasize the use of ANN in various MSW-related problems and highlight the challenges related to data reliability and the absence of clear performance baselines for assessing AI approaches. Overall, these studies contribute to understanding how AI can revolutionize waste management by improving waste logistics, classification, and treatment processes. They highlight the potential benefits and challenges of implementing AI in the field, providing valuable insights for future research and practical applications.

In particular, the need to overcome the constraints and complexities associated with conventional approaches (e.g., RFID, GPS, GIS), especially the status and waste level in bins, has driven the development and implementation of various advanced techniques. These include PSO (e.g., Ref. [ 253 ]), ANN (e.g., Ref. [ 254 ]), and Backtracking Search Algorithm (BSA) (e.g., Ref. [ 255 ]) for waste collection optimization. GA and nearest neighborhood search algorithms have also been used for waste vehicle routing (e.g., Ref. [ 256 ]). However, these techniques still lack precision and require a long execution time. Therefore, new techniques are needed to deal with cost and emission issues and consider bin capacity, waste weight inside the bin, collection frequency, vehicle capacity and maintenance, and trip rate [ 257 ]. Further, however, AI models can be applied to predict equipment failures in waste management facilities. AI models can identify potential issues in advance by analyzing data patterns, allowing for timely maintenance and minimizing downtime. Also, AI models can enable powerful DSS for waste management. These systems integrate various data sources, including weather conditions, population density, and waste composition, to provide insights and recommendations for effective waste management strategies.

Blockchain and AI and IoT applications

Blockchain technology is gaining widespread popularity across various domains, including energy, environmental conservation, and urban development, owing to its capacity to decentralize data and processes while ensuring robust security measures. In essence, Blockchain is an open-source, peer-to-peer, distributed ledger system that encompasses multiple transactions and their associated data organized within a chain of interconnected blocks within a decentralized, peer-to-peer, and openly accessible network, using technologies such as AI, ML, and Big Data (e.g., Ref. [ 258 ]). These blocks are subject to cryptographic validation by the network itself. According to Parmentola [ 259 ], Blockchain is a rapidly evolving approach that enables the recording, sharing, updating, and synchronizing of information and transactions across multiple data ledgers or databases within a distributed and openly accessible network of diverse participants. Consequently, it fosters enhanced collaboration and interaction among various organizations and individuals participating in the network. Moreover, it is distinguished by its core attributes, including anonymity, transparency, auditability, permanence, persistence, and decentralization, which collectively translate into improved operational performance, efficiency gains, and cost reductions [ 258 , 260 ].

In more recent years, Blockchain has become an innovative technology and solution for smart cities in environmental sustainability. It has been used by many governments to improve environmental sustainability. Integrating blockchain into renewable energy sources can unlock energy sustainability by facilitating the development of a decentralized and democratized energy system while aiding in improved climate governance through its attributes of transparency, global decentralization, and collaborative capabilities [ 261 ]. As demonstrated in a recent bibliometric study on environmentally sustainable smart cities, Blockchain is linked to the challenges, services, and resources of smart cities and AI, IoT, and big data analytics [ 1 ]. For the latter, in a use case of a developed “blockchain-based carbon emission rights verification system,” AI and Big Data are used to learn proven data [ 262 ]. Miao et al. [ 263 ] propose a blockchain and AI-based architecture for the natural gas intelligent IoT to address the supply chain failure of existing centralized energy supply architectures because of their overwhelming numerous requests that could cause pressure, temperature, and natural gas load to exceed safety limits. Also, based on multi-sensor-driven (or IoT-based) AI tools, blockchain platforms can optimize circular economy loops ([ 264 , 265 ]), allowing carbon footprint reduction and solid waste disposal control and thereby contributing to sustainability transitions [ 266 ]. Xiao et al. [ 267 ] propose using Blockchain for intelligent driving edge systems. The approach utilizes a double auction mechanism to optimize the satisfaction of users and service providers for edge computing and shows potential for better performance for resource utilization.

The value of Blockchain technology lies in storing data on green energy production activities related to environmental degradation and air pollution; enabling new means of green energy production, supply chain and logistics, real-time data collection and analysis for timely decision-making pertaining to green and low-carbon processes; and monitoring EV charging systems [ 259 ]. Moreover, Blockchain informs consumers and users about the use of less-efficient appliances. It enables them to improve their consumption behavior and thus reduce their carbon emissions [ 268 ], in addition to monitoring compliance with environmental standards by utilizing product traceability that can decrease resource inefficiencies and losses at different supply chain stages [ 269 ]. Also, Blockchain-based initiatives have been designed to provide credible trading services for polluters. They have also been used to tokenize carbon credits [ 269 ] and monitor carbon emissions. In a recent review of Blockchain applications in sustainable and smart cities, Makani et al. [ 270 ] provide a detailed account of how Blockchain technologies contribute to various urban domains, including transportation, smart grids, smart operational management, and smart homes. As regards the trading of renewable energy by local energy producers based on cryptocurrencies, Blockchain contributes to energy supply diversification, supply disruption risk reduction [ 269 ], and renewable energy promotion through microgrids and other alternative models [ 259 ]. These suggest the potential for innovative approaches that support localized and community-driven renewable energy production. By improving the alignment of energy supply and demand, these approaches can strengthen energy security and resilience, offering new avenues to enhance sustainable energy prospects. Concerning carbon emissions monitoring, the combination of Blockchain and IoT provides reliability for data and establishes measurement criteria homogeneity about registration systems and measurement tools, respectively. These solutions “minimize registration errors and eliminate fraud arising from the accounting and measurement of gas emissions” [ 269 ].

Regarding the role of AI in preventing and reducing marine pollution [ 271 ], Blockchain monitors water pollution changes and preserves marine resources [ 259 ]. It is also used in rewards schemes for residents of coastal areas using tokens of cryptocurrencies, which “can later be redeemed for credit to collect and share data on environmental conditions of water bodies” that can aid in enhancing decisions and designing regulations [ 269 ]. Similar reward schemes can raise public awareness and increase public participation in waste management and recycling by developing models to reward active users. Further, implementing Blockchain and AI as smart city technologies has several co-benefits associated with water management. Contributions in this regard relate to water provision efficiency, wastewater management, ensuring water security, groundwater monitoring, environmental awareness, promoting peer-to-peer trading of water rights, conserving water resources, and tackling the nexus between various natural resources in urban areas [ 259 , 269 , 271 , 272 ]. In particular, Blockchain and AI technologies could optimize water management in water-stressed urban areas by facilitating autonomous water distribution and management systems These minimize loss and control quality.

Overall, Blockchain technology, combined with AI and IoT, is crucial in advancing environmental sustainability. It provides a decentralized and transparent platform for securely recording and verifying transactions, data, and information. Integrating AI and IoT enables efficient data collection, analysis, and decision-making processes, leading to improved resource management, reduced environmental impact, and enhanced sustainability practices. The combination of blockchain, AI, and IoT allows for the development of innovative solutions such as smart grids, decentralized energy systems, waste management, and carbon footprint tracking. By fostering trust, traceability, and accountability, blockchain enhances the implementation of sustainable practices and facilitates the transition toward a greener and more sustainable future.

The sustainable development goals (SDGs)

The potential benefits of smart cities in catalyzing the transition to SDG 11 through advanced technologies and data-driven approaches are evident. Iris-Panagiota and Egleton [ 3 ] explore the role of AI within smart, sustainable cities, emphasizing its contributions to urban planning, management, and development. Zaidi et al. [ 273 ] analyze the trajectory of AI in smart sustainable cities research, pinpointing publication trends and research hotspots, including digital innovation, intelligent data systems, smart energy efficiency, and AI-IoT data analytics nexus. Yigitcanlar and Cugurullo [ 26 ] explore the sustainability of AI within the context of smart, sustainable cities, generating insights into emerging urban AI and the potential symbiosis between AI and smart, sustainable urbanism. The study reveals that AI applications have become integral in urban services, managing various aspects of urban life, such as transport systems, infrastructure, and environmental monitoring. The increasing adoption of AI is expected to continue, impacting the three dimensions of urban sustainability. Vinuesa et al. [ 271 ] reveal the potential of AI to advance 134 targets across all goals while hindering 59 targets. Collectively, these studies enrich the understanding of AI's role within smart, sustainable cities, an overarching umbrella term for smarter eco-cities — its applications in urban planning, the AI research landscape in smart cities, the integration of AI and IoT in urban contexts, the alignment of AI with SDG objectives, and the status of smarter eco-cities. Nonetheless, the trade-offs of smart cities — privacy, cybersecurity, digital divide, technology misuse, and legal frameworks — demand attention considering the use of AI and its integration (e.g., Ref. [ 1 , 3 ]). The imperative lies in devising measures that amplify social and economic priorities in smart city planning and development toward rendering smart eco-cities smarter and more sustainable.

6. Discussion: challenges, open issues, and limitations

6.1. smart city ai, iot, and big data technologies as key factors impacting the dynamics of existing smart eco-cities.

AI, IoT, and Big Data technologies are transforming how smart cities — and hence smart eco-cities — function by optimizing their processes, enhancing their practices, augmenting their solution capabilities, and improving their environmental sustainability performance. Since the mid-2010s, the data-driven technologies and solutions of smart cities as changing elements have gradually impacted the dynamics of eco-cities toward becoming smart in their approach to environmental sustainability by integrating their core domains with smart city domains. This will continue in the same direction as the AI, IoT, and Big Data technologies and solutions of smart cities become more advanced and integrated with sustainable technologies and strategies to provide innovative approaches that can demonstrate the ability to tackle more complex challenges. This, in turn, means making smart eco-cities smarter in their pursuit of achieving environmental sustainability thanks to the increasing use of AI and AIoT applications. The essence of AIoT revolves around the need to harness and leverage the power of smart city technologies and solutions, given the clear synergies in their operation concerning the optimization, efficiency, management, and planning processes of smarter eco-cities. This entails integrating their systems, coordinating their domains, and coupling their networks, creating many new opportunities that could be realized in environmental sustainability.

At the technical level, AI empowers the analysis of the colossal amounts of data generated via the IoT infrastructure in smart cities [ 14 ] and hence smart eco-cities, largely using ML for decision-making processes. Regarding AI-enabled sustainable smart cities, ML models can grow over time, detect invisible anomalies and alterations, exhibit various behaviors on different runs for the same input, and help provide real-time feedback for transport management, pollution control, energy management [ 2 ], and water management systems. Intelligent machines can “learn from experience, adjust to new inputs, and perform human-like tasks” ([ 274 ], p. 63) to “interpret external data correctly, to learn from such data, and to use those learnings to achieve specific goals through flexible adaptation” ([ 63 ], p. 17). Overall, AI can provide unsurpassed ways of automating or autonomizing the repetitive, complex, cognitively demanding, and time-consuming tasks associated with the operational functioning and planning of smarter eco-cities.

Concerning environmentally smart sustainable urbanism as an underlying paradigm of smart eco-cities, it is increasingly becoming a powerful societal framework for the transition toward environmental sustainability. This lies in developing joint actions for preserving the environment based on analyzing large-scale databases, understanding the complexity of climate change and modeling and simulating its potential impacts, improving the health of ecosystems, and enabling high integration of renewable energy and smart energy [ 271 ], enhancing smart renewable energy infrastructures in smart cities [ 275 ], optimizing energy consumption and production, developing more environmentally efficient transport systems, enhancing environmental monitoring (e.g., Ref. [ [245] , [276] , [277] ]), and strengthening low-carbon energy systems by supporting circular economies and smart eco-cities [ 271 ]. In particular, more than 250 studies applied AI to energy conservation and renewable energy during 2015–2019 [ 17 ]. The focus on the potential of AI and AIoT for energy can be justified by its pivotal role in the transition to smart eco-cities. This occurs through integrating large shares of renewable energy with smart energy through additional flexibility and decarbonizing other key emitting sectors, notably manufacturing, industry, transport, and buildings. Addressing the energy crisis and reducing fossil fuels will mitigate the impacts of climate change and make adaptation easier [ 278 ].

However, the dynamics of smarter eco-cities should evaluate AI and AIoT technologies as key components initiating changes in different domains. Investments in large-scale AI and AIoT as digital ecosystems are expected to positively impact smarter eco-cities that may involve feedback mechanisms, resulting in further adoption of these ecosystems and additional future investments. In other words, AI and AIoT technologies are likely to benefit from providing innovative applications in response to the need to overcome environmental sustainability challenges. Accordingly, they may exhibit positive feedback in that the more their solutions are implemented, the more likely they will be further implemented, thanks to network effects, learning, adaptation, and coordination. While the relationship between outcomes, investments, and implementations is expected to advance the transition of smart eco-cities toward environmental sustainability, stating a strong causal relationship resulting from such linkages needs to be more accurate. For this reason, coupled with other complex intertwined internal and external factors, understanding the dynamics of smarter eco-cities remains a daunting and uncertain challenge. This can be justified and elucidated in what remains of this discussion.

6.2. Environmental challenges and costs of AI and AIoT technologies

AI, IoT, and Big Data technologies pose significant challenges when making smart eco-cities environmentally smarter. Therefore, paying attention to both the opportunities and threats of AI and AIoT technologies is necessary. Smarter eco-cities must be environmentally friendly, thereby minimizing the negative impacts resulting from the wide use and increasing adoption of the applied solutions of AI and AIoT technologies. These enabling, integrative, and constitute technologies are embedded into a much wider socio-technical landscape involving a complex set of intertwined and heterogenous factors and actors. There is a risk of a mismatch between the environmental goals of smarter eco-cities and the opportunities offered by AI and AIoT technologies. This is due to their indirect, direct, rebound, and systemic effects, which are generated through their development, design, use, application, and disposal (see Ref. [ 279 ] for a detailed discussion). In particular, the indirect effects are expected to be exacerbated the most due to the increasing demand for AI and AIoT applications. The operation of AI and AIoT applications requires a lot of energy to power the IoT infrastructure, data processing platforms, cloud and edge computing, high-speed wireless networks, and large-scale AI systems. Concerning the latter, large data centers, which provide massive computational resources required by AI research, design, and development, are associated with significant energy consumption and, thus, carbon footprint, compromising the efforts supporting energy reduction and climate action [ 271 ]. According to current estimates, the global electricity demand for advanced ICT could increase to 20 % compared to around 1 % today [ 271 ]. AI involves establishing heavy energy dependency due to the intensive use of innovative technologies, increasing environmental impacts [ [280] , [281] ], extending car traveling distance, and causing urban sprawl [ 249 ]. The latter two relate to exurbanization, a process whereby upper-class or affluent dwellers move from urban areas to rural areas to maintain an urban life or live in high-end housing through advanced technology or long-distance commuting.

However, the high energy requirements for AI and AIoT applications in the case of the use of non-renewable sources of energy will undermine the efforts to achieve the environmental targets of smarter eco-cities. Concerning the direct effects, building smarter eco-cities requires deploying urban operating systems, urban operations centers, and urban dashboards [ 53 ] and thus massive amounts of natural resources for developing, installing, and maintaining AI and AIoT ecosystems. In addition, IoT and AI production, distribution, service, and disposal produce vast amounts of e-waste, unsustainable materials, and toxic pollution [ [63] , [282] ]. Almalki et al. [ 282 ] discuss, in a recent comprehensive review, the capabilities and potentials of IoT to respond to the needs of smart cities while highlighting the challenges for future research on smart city data-driven IoT applications, with a focus on their risks to environmental sustainability in terms of energy consumption, toxic pollution, e-waste, and others. All in all, the applied AI and AIoT solutions for smarter eco-cities are challenged by the effects of high energy-intensive structures, undermining the efforts deployed to avoid the overexploitation of primary resources to achieve carbon neutrality.

The green growth of AI, IoT, Big Data technologies, green computing, and eco-friendly design is critical to mitigating the risks of the mismatch between the environmental goals of smarter eco-cities and the opportunities offered by AI and AIoT technologies. This is consistent with the environment being intrinsic to SDG 11 in terms of recognizing the need to apply the most innovative technologies to make critical urban infrastructure resource-efficient, low-emission, and resilient by reaching the targets related to energy, climate, transport, waste, and water, as well as integrated policy and planning. The positive impacts of adopting sustainable approaches to the development, use, application, and disposal of AI and AIoT technologies lie in creating eco-friendly environments that are healthier and more livable in smarter eco-cities while accelerating their digital transformation. In this respect, Almalki et al. [ 282 ] analyze the various techniques and strategies for enhancing the quality of life and well-being by making cities greener, smarter, safer, and more sustainable. Bibri [ [157] , [279] ] sheds light on the innovative role of advanced ICT as a potential remedy for mitigating its carbon footprint and thus advancing environmental sustainability goals, enabling the transition from smart cities to environmentally smarter cities. In this line of thinking, Almalki et al. [ 282 ] note that the smart things enabled by IoT in smart cities “become smarter to perform their tasks autonomously” while communicating “among themselves and humans with efficient bandwidth utilization, energy efficiency, mitigation of hazardous emissions, and reducing e-waste to make the city eco-friendly and sustainable.” Here comes the role of AI in green computing concerning smarter eco-city sensor integrated transportation systems, energy systems, building systems, waste systems, environmental monitoring systems, and so on. Green computing is key to decreasing carbon emissions and energy consumption to fulfill the environmental goals of sustainability in smart cities.

It is essential to focus on “reducing pollution hazards, traffic waste, resource usage, energy consumption, providing public safety, life quality, and sustaining the environment and cost management” to make smart cities eco-friendly [ 282 ]. This, in turn, means that AI and AIoT solutions should be carefully implemented in combination with sustainable and eco-friendly design principles, energy-efficient policy instruments, and other relevant measures. This is to ensure that the efficiency gains enabled by AI and AIoT solutions lead to reducing energy use and carbon footprint. Almalki et al. [ 282 ] provide practical insights into the data-driven IoT-based eco-friendly and sustainable cities research field. Wang and Liao [ 283 ] explore the intersection of eco-design with AI and Big Data. In doing so, they identify automation and control systems and computer science among the leading application disciplines. The authors argue for the necessity of more concerted efforts “to advance both the theoretical and empirical research on the nascent topic among researchers, funding bodies, policy-makers, and industry professionals given that the notion of eco-design of AI and Big Data applications is expected to be pertinent and relevant for designing greener strategies, products, and services for green digital transformation.” As a justification for a more consolidated green approach to AI, most attempts at using AI applications to enhance urban efficiencies have struggled, if not failed, to accomplish the transformative changes to smart cities due to “the short-sighted, technologically determined, and reductionist AI approaches being applied to complex urbanization problems” [ 4 ].

6.3. Technical and computational challenges of AI and AIoT technologies

Inherent to AI models and systems are several technical and analytical challenges. These include, as indicated by Nishant et al. [ 17 ] in a study conducted on AI for environmental sustainability, the overreliance of ML models on historical data, the uncertainty surrounding how humans behave in response to AI-based interventions, and the difficulty in measuring the effects of intervention strategies. Most new AI systems rely on Big Data and fail to demonstrate self-ideation or self-creation. AI must develop new concepts and models of intelligence cognition beyond ML/DL. This could offer novel solutions for environmental sustainability and climate change, feeding into new models of smarter eco-cities that may reduce decision biases due to the incompleteness and uncertainty of the data collected and aggregated in real-time. Concerning the over-reliance on Big (historical) Data in smart cities, Batty et al. (Ref. [ 128 ], p. 507) note that the prospect of real-time data collection and aggregation to deal with urban changes at any spatial or temporal scale “is a long way off and will never be reached … but what it does promise is an ability to have a real-time view of change at different spatial scales and over different time scales. This will change both the models we can build and how these technologies can inform the decision process with simulations and decision support being telescoped across space and time.” While this may be relevant to climate change in modeling and simulation, human-related variables are, to some extent, unpredictable and dynamically changing. This implies that more and varied types of data need to be collected and aggregated, new and more extensive sources of data to be explored, and new and more advanced tools for handling various velocities of data to be created. Significantly, historical datasets tend to be of limited value about climate periods and cycles, which makes it difficult to make precise predictions or decisions. A deterministic approach is difficult to adopt in climate change, as it is impossible to estimate or determine potential outcomes precisely. Non-deterministic ML models are more relevant to transport management, energy management, water management, waste management, and pollution monitoring, where they can detect anomalies and alterations and help provide real-time feedback. Furthermore, the variance-bias tradeoffs associated with ML [ 284 ] have implications for climate change solutions due to the bias and oversimplification inherent in predicting future climate change scenarios.

In addition, Kuguoglu et al. [ 15 ] investigate the reasons behind the failure of many smart city initiatives that rely on AIoT to scale up. Through a combination of literature study and expert interviews, the study identifies various factors contributing to the lack of scalability. These factors include resource and capability constraints, overlooking the importance of comprehensive change, and the influence of different factors at different stages of implementation. Regarding the technical challenges related to DL-based AIoT, they include [ 13 ]:

• • Multimodal heterogeneous data processing, transmission, and storage pertaining to the massive numbers of heterogeneous sensors and the vast data streams of different formats, sizes, and timestamps they generate.
• • Limited computational and storage resources in relation to using DL models for real-time data stream processing and low latency.
• • Computational scheduling in AIoT architecture and related intense computation. This entails meticulous coordination across cloud centers, fog nodes, and edge devices, factoring in variables like data type, volume, network bandwidth, processing latency, performance accuracy, data security, privacy specific to the application scenario, and unbalanced data flow and user demands over time.
• • Labeling unlabeled big data for DL in AIoT in terms of managing the time-consuming and financially demanding nature of this process while ensuring high-quality results.
• • Data monopoly, where access to proprietary data is restricted due to vested interests, poses challenges to achieving equitable access to extensive proprietary datasets.

6.4. Challenges and considerations of explainable AI and interpretable ML

Explainable AI (XAI) and Interpretable ML (IML) encounter significant challenges in the context of AI and AIoT solutions for smart cities, environmental sustainability, and climate change, particularly in the evolving landscape of smarter eco-cities. XAI and IML are interconnected concepts aiming to enhance the transparency, comprehensibility, and credibility of AI models for various stakeholders involved in smarter eco-city development. While XAI focuses on explaining the decision-making process of AI systems, IML specifically concentrates on creating ML models that produce easily interpretable outcomes. XAI encompasses diverse approaches, including IML techniques, to explain AI decisions, aligning with the broader aim of enhancing explainability in AI. Both XAI and IML play pivotal roles in creating AI and AIoT systems that foster accountability, trustworthiness, and effective human-AI interaction, which is vital for making informed decisions in the context of smarter eco-cities. Nonetheless, several challenges and issues arise in this context (e.g., Ref. [ [285] , [286] , [287] , [288] , [289] , [290] , [291] , 292 ]), including, but are not limited to:

• • Complexity and interpretability: Applying AI and AIoT solutions to complex challenges in smarter eco-cities can lead to intricate models, hindering their interpretability. Ensuring these complex systems generate transparent decisions amid intricate environmental and climate data is crucial.
• • Black-box models: Many advanced AI models (e.g., DNN) are considered black boxes, lacking transparency in decision-making. This lack of insight can hinder trust in and adoption of smart city systems, especially when critical decisions are at stake.
• • Bias and fairness: Bias in AI models, derived from biased training data, can perpetuate existing inequalities in resource allocation and exacerbate environmental disparities. Overcoming these biases and ensuring fair outcomes is a daunting task. Biased decision-making becomes evident in real-time and predictive analytics, hindering the pursuit of environmental sustainability.
• • The trade-off between accuracy and interpretability: Striking a balance between accurate predictions and clear explanations is essential, as more complex models might offer better predictions but sacrifice interpretability.
• • Interdisciplinary nature: Addressing environmental sustainability and climate change requires expertise from diverse fields. Ensuring that AI and AIoT solutions are interpretable to domain experts, policymakers, and citizens across various disciplines is a challenge, as technical terminology can create barriers to communication.
• • Data privacy and security: While explaining AI decisions and promoting transparency, care must be taken not to compromise sensitive or private information about individuals, potentially compromising their privacy.
• • Dynamic and evolving environments: Smart cities and smarter eco-cities are dynamically changing environments, necessitating adaptable and robust methods for interpreting AI decisions.
• • Education and adoption: Educating stakeholders, including policymakers, city planners, city managers, and citizens, about the benefits and limitations of AI and AIoT solutions, building trust and confidence, and encouraging adoption are critical factors in realizing smarter eco-cities.

The real challenge of XAI lies in granting substantial power to smarter eco-city systems without simultaneously enabling them to explain the intricate decision-making processes to different groups of domain experts. AI and ML models and algorithms assume control over decision-making by analyzing generated data, predicting outcomes, and maximizing value based on certain criteria. This reduces the rich complexity of urban life and the unpredictability of urban dynamics and systems to narrow quantitative and unitary languages, potentially disregarding the significance of cultural, ethical, social, and political values. As a result, technological advancements may pose difficulties in achieving the status of smarter eco-cities due to the mechanistic way of perceiving these complex systems. Therefore, a recent wave of research has started to focus on XAI to address some of the concerns posed by the application of AI in various domains. Mayuri, Vasile, and Indranath [ 290 ] present several applications of XAI/IML and methods to make AI/ML models explainable/interpretable. Ghonge [ 287 ] addressed several case studies and use cases of XAI as well as its impacts and challenges in smart city applications. Javid et al. [ 288 ] comprehensively delve into the landscape of XAI in smart cities, focusing on current and future developments, trends, enabling factors, use cases, challenges, and solutions. The authors outline research projects, standardization efforts, lessons learned, and technical hurdles.

XAI and IML methods are pivotal for the sustainable advancement of AI and AIoT solutions, allowing society to foster trust in the environmental and social-economic aspects of sustainability. These methods explain accuracy, fairness, transparency, accountability, and human-centeredness outcomes in AI and AIoT-powered decision-making, addressing ethical and governance concerns. These principles hold substantial relevance for data-driven decision-making in smarter eco-cities, thereby the need for creating explainable/interpretable models, techniques, and tools. Collaborative efforts among AI/AIoT experts, environmental scientists, urban planners, and policy-makers are essential to ensure the effective contribution of AI and AIoT technologies to environmental sustainability and climate change mitigation in the evolving landscape of smarter eco-cities. Also, future research endeavors will play a pivotal role in realizing transparent, effective, and ethically sound applications of XAI and IML methods within AI and AIoT solutions, advancing environmental sustainability in smarter eco-cities while ensuring equitable outcomes for all stakeholders.

6.5. Ethical and Societal Challenges of AI and AIoT technologies

AI technology's ethical and humanistic issues and risks are subject to long-standing intellectual and philosophical debates. The development, deployment, and adoption of AI technology raise these concerns, irrespective of the environmental benefits of its applied innovative solutions, depending on the application domain. Against the backdrop of this study, the use of AI involves making biased decisions, exacerbating privacy and cybersecurity, and limiting public trust [ 244 , 293 , 294 ]. Most of these challenges also apply by extension to AIoT, e.g., AIoT security for smart cities, AIoT and intrusion detection, and AIoT and trust recommendation [ 10 ]. Koffka [ 9 ] addresses critical concerns, including security and privacy, interoperability, and ethics, underscoring the importance of a responsible AIoT ecosystem. Furthermore, using AI entails devaluating human abilities, deepening information asymmetries, undermining equal power relations, and causing system failures. Regarding the latter, increasing public awareness of this type of risk is crucial before launching large-scale AI deployments in a society increasingly dependent on AI technology [ 271 ]. This indeed is arcane in that its actual functionalities and mechanisms are understood by only a group of people, despite being already part of the everyday life of many of us [ 6 ]. Therefore, given that AI as a disruptive technology will greatly transform smart eco-cities, it needs to earn public trust regarding how people perceive it. Also, AI technology needs to gain the trust in the minds of government agencies and public organizations investing in it [ 295 ]. The challenges posed by AI generally involve gaps in ethical standards, including safety, fairness, transparency [ 293 , 296 ], socio-economic equality, cultural diversity, and social inclusion. For example, safety is a key topic in ethical and legal debates over autonomous systems [ 297 , 298 ]. Most ethical issues raised by AI and AIoT applications relate to the difficulties in explaining AI models or interpreting ML algorithms, which resemble black boxes, with some being the hardest for humans to comprehend. There is a need for developing new methods for Explainable AI (XAI) or Interpretable ML (IML) that allow humans (designers, engineers, researchers, city planners, city managers, regulators, and policymakers) to understand and trust the decisions or predictions that AI models and systems make in terms of their potential biases and expected impacts. Ghonge [ 287 ] addressed several case studies and use cases of XAI as well as its impacts and open challenges in smart city applications.

The underlying ethical gaps of AI technology call for designing and implementing appropriate regulatory frameworks to address the counterproductive outcomes emanating from the penetrative patterns of AI (e.g., Ref. [ 63 , 244 , 299 , 300 ]) in urban life domains in emerging smarter eco-cities. Especially, early in the decision-making process of its deployment — when the opportunity for effective inputs and informed choices is greatest. This pertains to developing “responsible and ethical AI” before it is too late [ 297 , [301] , [302] ]. There is a warrant for this as the integration of AI with IoT and Big Data is speeding up the pace of advancements and innovations in both AI and AIoT systems, particularly the exponential rise of their computational power, paving the way for them to gain more and more power of the automation and autonomization of smart cities, with profound implications for smarter eco-cities. While it is possible to automate certain urban processes and practices concerning environmental sustainability and climate change, it is necessary to carefully plan and implement them to avoid generating fully automated or autonomous smarter eco-cities based on mechanical decisions. In this regard, it is essential to address and overcome the regulatory challenges pertaining to the use of AI and AIoT applications to advance environmental sustainability. Vinuesa et al. [ 271 ] emphasize the need for regulatory insight and oversight to support the development of AI-based technologies for sustainable development.

The realization of the common good of AI and AIoT technologies remains highly improbable when AI systems operate solely according to the algorithms designed and implemented by powerful corporations driven by ambitions for power, profit, and extensive reach and influence. These tech giants pursue various trajectories and explore uncharted possibilities, raising concerns about the potential consequences of the large-scale implementation of AI and AIoT systems. There is an urgent demand for well-regulated and responsible AI and AIoT systems that prioritize safeguarding public and civic values within the context of smart eco-cities. Such systems must be designed to ensure that the broader benefit to society takes precedence over corporate interests and unchecked advancements. Indeed, civic values and public values play vital roles in the functioning of both civil society and government. These values are the moral compass that guides individuals, communities, and public institutions in their pursuit of a just, inclusive, and prosperous society. In a civil society, civic values, such as social justice, freedom, tolerance, compassion, and tolerance, are the cornerstones of a harmonious and fair community. They inspire individuals to engage in civic activities, advocate for their rights, and work collectively to address societal issues. Civic engagement is fostered by these values, encouraging citizens to participate in public discourse and actively contribute to the betterment of society. At the same time, public values are fundamental to the proper functioning of government. These principles, including accountability, transparency, integrity, inclusivity, public participation, and environmental stewardship, ensure that public institutions operate in the best interests of the people they serve. However, for example, in relation to environmental stewardship, the crucial aspects of environmental protection, justice, and preservation are often sidelined when unregulated economic interests drive urban development.

While technological advancements, such as AI and AIoT systems, have the potential to enhance the efficiency and effectiveness of both civil society and government, it is essential to recognize that certain core functions should not be outsourced to these systems. The decision-making processes guided by civic and public values require the nuanced judgment and ethical considerations that only humans can provide. AI and AioT systems can be valuable tools, but they should support and complement the efforts of individuals and institutions rather than replace or overshadow the importance of these foundational values. The interplay between civic and public values and emerging technologies should be carefully managed to ensure that they continue to serve as the moral and ethical foundations of our society and government. However, Kassens-Noor and Hintze [ 303 ] argue that the adoption rate of AI technology, coupled with policy regulations and unforeseen events, has the potential to transform bustling metropolises into deserted ghost cities. The complete advancement of AI and AIoT may signify a decline in moral and societal values, raising concerns about the potential demise of the human race. Nevertheless, despite the enticing conceptual and discursive benefits (which relate to both ideas, theories, and perceptions, as well as the ways in which they are discussed and communicated) of transitioning cities into eco-cities, formidable obstacles have hindered large-scale implementations since the early 1990s, not to mention the development of smart(er) eco-cities. One of the most significant challenges confronting urban transformations lies in the significant costs, risks, and uncertainties associated with the incorporation of AI and AIoT into the realm of eco-urbanism.

6.6. Methodological limitations

It is important to acknowledge the methodological limitations of this comprehensive systematic review to allow the readers to assess the reliability and validity of the findings and understand the potential implications for future research and practice. These limitations, which arose from the various aspects of the review process, include:

• • Search strategy: Despite efforts to conduct a thorough literature search, some relevant studies may have been missed. Limitations in database selection, search keywords, language restriction, and inclusion/exclusion criteria could influence the breadth and depth of the included studies for synthesis. Moreover, as the study relied mainly on peer-reviewed documents, there is a risk of excluding a large part of grey literature and stakeholder input and not gaining extensive insights on emerging smarter eco-cities. This area of research is still evolving, and most of its existing work drew mainly from the fields of environmental sustainability, climate change, and smart cities with respect to AI and AIoT technologies and solutions. Even these fields are associated with a paucity of knowledge and a sparsity of empirical evidence due to the burgeoning nature of these technologies and solutions.
• • Publication bias: The study relied on published literature, and there is a risk of publication bias, where studies with positive results are more likely to be published, while studies with negative findings may be overlooked. This can affect the comprehensiveness and representativeness of the systematic review. It is common for researchers and journals to preferentially publish studies based on the direction or significance of their findings. This can lead to an overrepresentation of studies with positive results, while studies with negative results may be less likely to get published and thus be included in the systematic review. In addition, there is a language bias in that studies published in English are more accessible and commonly included in systematic reviews, leading to the potential exclusion of valuable evidence. Especially, English language was one of the inclusion criteria applied in the study. Funding sources are another bias regarding the studies funded by industry having a higher likelihood of being published — if they produce favorable results.
• • Data extraction and synthesis pertain to the complexities and challenges of extracting data from selected studies and synthesizing their findings. One of the primary issues posed in the study was the heterogeneity across studies. This included variations in study designs (e.g., methodologies) that complicated the synthesis of findings and reporting formats (differences in the presentation of results) that affected the extraction and synthesis process. These differences can make it difficult to directly compare and combine data from different studies, thus obtaining a comprehensive overview of the evidence and drawing robust conclusions from the systematic review.

7. Suggestions for future research

Smarter eco-city scholars, practitioners, and policymakers have a new opportunity to foster sustainable development practices based on a new paradigm of solution-thinking grounded in a deeper understanding of the interplay between techno-scientific and socio-political solutions. Developing this multi-faceted change process is one of the most critical challenges of sustainable urban development to achieve the status of smarter eco-cities. This emerging area of research is empirically under-researched, theoretically under-developed, and critically under-thought to allow for large-scale implementations. This means that a plethora of problems and questions need to be addressed and answered to guide the development of smarter eco-cities and, hence, large-scale AI and AIoT deployments for the common good. Some gaps in our knowledge of emerging smarter eco-cities follow from our results and discussion. These gaps span a broad set of topics that are significant to investigate or critically engage with and that can be approached from various perspectives in the form of suggestions for future research.

Considering the transformative potential of AI and AIoT technologies in reshaping smarter eco-cities, a comprehensive investigation is warranted to unravel the dimensions, opportunities, benefits, and challenges inherent in this emerging urban paradigm. Given the nascent stage of research at the intersection of environmental sustainability, climate change, and smart cities within the context of AI and AIoT solutions, the following avenues are crucial:

• • Identify key drivers: Delve into the multifaceted drivers that underpin the evolution of smart eco-cities, encompassing social, economic, institutional, and political factors beyond technological and environmental aspects.
• • Evaluate effectiveness: Prioritize assessing the real-world effectiveness and scalability of AI and AIoT applications in smarter eco-cities, determining their actual impact on achieving SDGs.
• • Explore long-term benefits: Probe the long-term benefits and opportunities offered by AI and AIoT technologies in fostering sustainability practices in smarter eco-cities. This includes exploring their potential for resource optimization, energy efficiency, waste reduction, and enhancing the quality of life for citizens.
• • Overcome barriers and risks: Confront challenges and mitigate risks associated with AI and AIoT implementation, such as privacy concerns, data security, and governance frameworks.
• • Promote responsible AI practices: Investigate guidelines and best practices for the responsible design, deployment, and governance of AI and AIoT solutions. Ensuring fairness, transparency, and accountability in decision-making processes is essential.
• • Integrate disciplines: Foster interdisciplinary research merging environmental sustainability, climate change, and smart cities with AI and AIoT. This approach unravels intricate relationships and facilitates a comprehensive understanding of the complex interactions and interdependencies between these domains, and enables the development of integrated and holistic solutions
• • Formulate policies and frameworks: Develop robust policy and governance frameworks that facilitate the ethical and transparent use of AI and AIoT technologies and support their adoption and implementation in sustainable urban development. This includes examining regulatory mechanisms, standards, and guidelines to ensure transparency, accountability, and ethical use of these technologies.
• • Advance XAI and IML methods: Develop solutions for interpretability in intricate models, user-centric model training, tailored solutions for smarter eco-cities, resilience and adaptability enhancement, and ethical implications. This entails developing XAI techniques for elucidating complex AI models in AIoT systems, exploring IML integration to engage users in refining models, utilizing XAI and IML to tackle distinct environmental challenges, fortifying AIoT systems' resilience against dynamic urban scenarios, and scrutinizing the ethical and societal implications of deploying XAI and IML in AIoT solutions to ensure equity and transparency.
• • Promote community engagement: Explore ways to involve local communities, stakeholders, and citizens in the design, implementation, and monitoring of AI and AIoT solutions for smarter eco-cities. Their active participation can lead to more inclusive and effective outcomes.
• • Quantify environmental impact: Develop methodologies to quantify the environmental impact of AI and AIoT solutions in smarter eco-cities. This involves assessing factors like energy consumption, carbon footprint, and resource utilization to understand the overall sustainability gains.
• • Conduct lifecycle analysis: Assess the sustainability of AI and AIoT technologies across their entire lifecycle, from production to disposal. This holistic approach can reveal potential environmental hotspots and guide improvements.
• • Promote citizen participation: Explore ways to empower citizens with AI-augmented information and tools that enable them to participate in environmental conservation and sustainable behaviors actively.
• • Enhance behavioral insights: Investigate how AI can leverage behavioral insights to encourage environmentally friendly behaviors among citizens, such as energy conservation and waste reduction.
• • Perform case studies: Conduct in-depth empirical inquiries and practical applications of AI and AIoT in actual smart eco-city projects. These real-world insights illuminate challenges, best practices, and valuable lessons for effective integration.

By addressing these knowledge gaps and pursuing these research avenues, the field of smarter eco-cities can advance its understanding, implementation, and impact. This will contribute to developing more sustainable and technologically advanced urban environments that benefit society and the environment.

8. Conclusion

As disruptive technologies, AI and AIoT lay the foundational technological infrastructure essential for constructing the digital ecosystem of emerging smarter eco-cities to amplify and sustain their contributions to environmental sustainability goals. This pursuit involves enhancing the efficiency and effectiveness of their operations, functions, strategies, and policies in alignment with the environmental targets of SDG 11. Within this context, it is important to acknowledge the immense potential of AI and AIoT technologies to develop robust intelligent systems generating profound insights for decision-making processes. However, these technologies cannot serve as a universal remedy or panacea for the wicked problems characterizing smarter eco-cities as complex systems. In this study, we aimed to provide a comprehensive systematic review of emerging smarter eco-cities and their leading-edge AI and AIoT solutions for environmental sustainability, employing a unified approach to evidence synthesis. The study's key findings concerning the five research questions are outlined as follows:

Interlinked foundational underpinnings of smarter eco-cities: The study showed that the fundamental concepts underpinning smarter eco-cities are intricately interconnected and build one on another on various scales. The key underlying urbanism paradigms, namely smart cities and eco-cities, serve as the foundation for integrating data-driven technologies and environmental solutions. Data-driven technologies enable real-time monitoring, analysis, and decision-making, while environmental solutions focus on optimizing resource efficiency and minimizing ecological footprint. Data-driven insights enhance the effectiveness of environmental strategies, ultimately contributing to creating more resilient, livable, and environmentally friendly urban environments.

The materialization of smarter eco-cities: The study identified several intertwined factors contributing to the materialization of smarter eco-cities as an emerging paradigm of urbanism, including the growing need for sustainable development, advancements in technology, environmental considerations, policy instruments, and government initiatives, and the recognition of the potential of data-driven technologies in addressing complex environmental challenges.

The primary AI and AIoT solutions harnessed in the development of emerging smarter eco-cities: The study identified many applied solutions of AI and AIoT technologies, demonstrating their role in urban planning, management, and development. These solutions encompass energy conservation and renewable energy, sustainable transportation management, traffic control, water resources conservation, waste management for efficient resource utilization, biodiversity and ecosystem services, environmental monitoring and control, climate change adaptation and mitigation, and disaster resilience and management.

The benefits and opportunities of AI and AIoT technologies in fostering sustainability practices in emerging smarter eco-cities: The study identified the opportunities and benefits offered by AI and AIoT technologies in the context of environmental sustainability. Combined, these opportunities and benefits included optimized resource management, increased energy efficiency, enhanced waste management, improved transportation, and mobility management, reduced environmental impacts, increased resilience to environmental challenges, enhanced decision-making in urban management and planning, and the potential for creating more sustainable and technologically advanced urban environments. Identifying opportunities — favorable circumstances and possibilities — helps understand the potential areas where AI and AIoT solutions can bring about positive changes and contribute to the overall development of smarter eco-cities. Benefits — positive outcomes and advantages — highlight the tangible and intangible gains that can be achieved by adopting AI and AIoT solutions.

Challenges and barriers arising in the implementation of AI and AIoT solutions for the development of emerging smarter eco-cities: The study identified and evaluated the key challenges pertaining to environmental costs, privacy concerns related to data collection and usage, cybersecurity risks related to interconnected systems, public trust, and social acceptance, limited technical expertise and knowledge, the lack of robust regulatory frameworks to ensure ethical and responsible AI and AIoT deployment and the requirement for addressing the social issues to ensure equitable and transparent use of AI and AIoT technologies.

Overall, the study highlighted the significance of AI and AIoT technologies in advancing the transition toward environmental sustainability in smarter eco-cities. While these technologies provide new and largely expanded opportunities to understand better and prevent environmental problems, they pose significant challenges that must be addressed and overcome to successfully implement smarter eco-cities. Therefore, it is important to emphasize the need for interdisciplinary research, policy support, and collaboration among different stakeholders to overcome these challenges and maximize the benefits of these technologies.

The synthesized evidence presented in this study has significant implications for researchers, practitioners, and policymakers involved in designing, managing, and planning smarter eco-cities. It offers valuable insights into the various dimensions of emerging smarter eco-cities and identifies best practices that can inform decision-making processes. The systematic review serves as a knowledge repository, guiding stakeholders in understanding the current state of research, identifying gaps, and shaping future strategies for sustainable urban development. Firstly, by identifying the core conceptual underpinnings of emerging smarter eco-cities and the intricate interconnections between them (RQ1), coupled with the intertwined factors propelling the materialization of smarter eco-cities (RQ2), the study encourages interdisciplinary collaboration among researchers and practitioners from various fields and foster new research avenues and practice pathways, leading to a more thorough understanding and focused improvement of emerging smarter eco-cities. Secondly, by identifying the key applied solutions of AI and AIoT technologies (RQ3), the study can inform researchers, practitioners, and policymakers about the technological advancements and innovative approaches employed in fostering sustainable urban development practices. Furthermore, exploring the potential opportunities and benefits offered by AI and AIoT technologies in this regard (RQ4) can provide valuable insights for decision-makers and urban planners seeking to leverage these technologies for achieving the SDGs, especially SDG 11. Lastly, identifying challenges and barriers in implementing AI and AIoT solutions in emerging smarter eco-cities (RQ5) can inform policymakers and stakeholders about the potential obstacles and open issues that need to be addressed when integrating these technologies into urban development strategies. Overall, the research, practice, and policymaking implications of this study encompass a wide range of areas, including urban planning, technology implementation, sustainability practices, and policy development, facilitating informed decision-making, and promoting the advancement of smarter eco-cities.

Ultimately, the findings of the systematic review contribute to the broader goal of creating smarter eco-cities that prioritize environmental sustainability, resource efficiency, and human well-being. The invaluable insights gained accordingly will empower stakeholders to make strategic choices, implement innovative solutions, and drive positive change in urban planning and management. By leveraging the potential of AI and AIoT, policymakers, urban planners, researchers, and practitioners can work together toward creating smarter, more resilient, more livable, and environmentally conscious cities that meet the needs of present and future generations. To sum up, AI and AIoT technologies will offer unprecedented capabilities to rise to many of the grand environmental challenges, but how these technologies will be used and what other possible directions this use might take is up to all of us, especially the research community, and for the time to tell.

CRediT authorship contribution statement

Simon Elias Bibri : Conceptualization, Methodology, Formal Analysis, Investigation, Data Curation, Visualization, Software, Writing - Original Draft, Writing - Review & Editing. John Krogstie : Conceptualization, Writing - Review & Editing. Alexandre Alahi and Amin Kaboli : Writing - Review & Editing. All authors read and approved the published version of the manuscript.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this article.

Acknowledgments

This project has received funding from the European Union's Horizon 2020 research and innovation program under the Marie Sklodowska-Curie grant agreement No. 101034260.

Abbreviations

• Open access
• Published: 21 May 2024

A scoping review on bovine tuberculosis highlights the need for novel data streams and analytical approaches to curb zoonotic diseases

• Kimberly Conteddu   ORCID: orcid.org/0000-0002-3883-4137 1 ,
• Holly M. English 1 ,
• Andrew W. Byrne 2 ,
• Bawan Amin 1 ,
• Laura L. Griffin 1 ,
• Prabhleen Kaur 3 ,
• Virginia Morera-Pujol 1 ,
• Kilian J. Murphy 1 ,
• Michael Salter-Townshend 3 ,
• Adam F. Smith 4 , 5 , 6 &
• Simone Ciuti 1  

Veterinary Research volume  55 , Article number:  64 ( 2024 ) Cite this article

215 Accesses

Metrics details

Zoonotic diseases represent a significant societal challenge in terms of their health and economic impacts. One Health approaches to managing zoonotic diseases are becoming more prevalent, but require novel thinking, tools and cross-disciplinary collaboration. Bovine tuberculosis (bTB) is one example of a costly One Health challenge with a complex epidemiology involving humans, domestic animals, wildlife and environmental factors, which require sophisticated collaborative approaches. We undertook a scoping review of multi-host bTB epidemiology to identify trends in species publication focus, methodologies, and One Health approaches. We aimed to identify knowledge gaps where novel research could provide insights to inform control policy, for bTB and other zoonoses. The review included 532 articles. We found different levels of research attention across episystems, with a significant proportion of the literature focusing on the badger-cattle-TB episystem, with far less attention given to tropical multi-host episystems. We found a limited number of studies focusing on management solutions and their efficacy, with very few studies looking at modelling exit strategies. Only a small number of studies looked at the effect of human disturbances on the spread of bTB involving wildlife hosts. Most of the studies we reviewed focused on the effect of badger vaccination and culling on bTB dynamics with few looking at how roads, human perturbations and habitat change may affect wildlife movement and disease spread. Finally, we observed a lack of studies considering the effect of weather variables on bTB spread, which is particularly relevant when studying zoonoses under climate change scenarios. Significant technological and methodological advances have been applied to bTB episystems, providing explicit insights into its spread and maintenance across populations. We identified a prominent bias towards certain species and locations. Generating more high-quality empirical data on wildlife host distribution and abundance, high-resolution individual behaviours and greater use of mathematical models and simulations are key areas for future research. Integrating data sources across disciplines, and a “virtuous cycle” of well-designed empirical data collection linked with mathematical and simulation modelling could provide additional gains for policy-makers and managers, enabling optimised bTB management with broader insights for other zoonoses.

1 Introduction

Emerging infectious diseases represent a significant public health concern as they become more prevalent worldwide [ 1 , 2 , 3 ]. It is estimated that about 60% of emerging infectious diseases are zoonotic, 72% of which have been estimated to originate from wildlife [ 2 , 3 ]. In 2019, thirteen different zoonoses had confirmed cases in humans within the European Union [ 4 ]. This has likely been accelerated by exponential growth in global population size and mobility with associated increases in urbanisation and concurrent loss of natural habitats. It has also led to increasing occurrences of human-wildlife interactions (e.g., improper waste disposal, intentional feeding of wildlife, movement of wildlife to human-dominated areas) and, therefore, exposure to zoonotic diseases [ 5 , 6 , 7 ]. Contact between humans, livestock and other captive animals, and wildlife species is only expected to keep increasing, leading to concerns about increased incidences of zoonotic disease transfer [ 6 , 8 , 9 ]. The question, however, remains of how to best track and manage emerging diseases.

A critical example is Zoonotic Tuberculosis (zoonotic TB), which was estimated in 2016 to be linked to 147 000 human cases and 12 500 deaths worldwide [ 10 ]. Zoonotic TB is driven mainly by Mycobacterium bovis (i.e., the causative agent of Bovine Tuberculosis—also as bovine TB or bTB), which is transmitted by several wildlife hosts and livestock. Britain and Ireland, as well as many other countries worldwide [ 10 ], have been increasingly impacted by bTB, resulting in significant economic loss. In Ireland, for instance, 4.89% of cattle herds tested positive for bTB in 2023, leading to the humane killing of 28 868 cattle [ 11 ]. This is in addition to the economic costs associated with the national bTB eradication program with €92 million spent in 2018 alone [ 12 ]. Similar trends can be observed in the UK, with £70 million spent annually for bTB prevention and control [ 13 ]. This disease also raises welfare concerns for wildlife hosts, especially considering its high prevalence in the wild. Badgers ( Meles meles ), for example, have been shown to have a bTB prevalence exceeding 40% in hotspot areas in Ireland [ 14 ], and red deer in Spain have been estimated to have a prevalence of up to 50% [ 15 ].

Bovine TB eradication is prioritised by governments and researchers due to the significant health concerns and economic (trade) impacts. Despite decades of control efforts in several countries, the pathogen has successfully avoided eradication. There are complex reasons as to why this is the case [ 16 ], but a primary factor relates to its complex dynamics of transmission and maintenance across differing hosts and the environment. Therefore, new thinking may be required to further investigate if disease control can be driven toward eradication. Detecting gaps in the current bTB literature is an essential step required to identify target areas for future research and to further hone government eradication strategies.

One way in which this may be addressed, and which requires assessment as to its prevalence in the literature, is through multidisciplinary, coordinated collaborations between the public health sector, veterinarians, ecologists and wildlife managers. The importance of interdisciplinary approaches is highlighted by the interlinked nature of human, animal and ecosystem health, which led to the concept of “One World One Health™” [ 17 , 18 ]. Despite such multidisciplinary efforts, the effect of stressors (i.e., direct and/or indirect disturbances such as hunting, habitat loss, and more broadly habitat and climate change) on animal ecology within human-dominated landscapes and the potential emergence of zoonotic disease is still understudied [ 1 ]. For example, we are aware that human-driven changes in the environment can modify interactions between hosts, change host and vector densities, and alter host longevity and movement [ 19 , 20 ]. A study by Castillo-Neyra et al. showed that rabies transmission was spatially linked to water channels, which act as ecological corridors connecting multiple susceptible populations and facilitating pathogen spread and persistence [ 20 ]. However, with cities expanding and providing urban corridors to wildlife, pathogen persistence could become even more of an issue [ 20 ], confirming the importance of studying the effect of human perturbations on animal ecology and related implications in disease ecology.

Additionally, transmission of different zoonoses often involve multiple agents including humans and a diverse range of wild and domestic animals. In order to understand the processes behind their transmission, it is essential to clearly disentangle the role of each agent involved [ 19 ]. Due to the complexity of disease transmission and the maintenance of infection within multiple wildlife hosts, for example between bovine and badger populations in the case of bTB, the individual components of the transmission chain are often studied separately. This can limit our understanding of the subtle underlying effects explaining disease emergence and transmission. Therefore, a holistic approach is essential to develop a complete picture of the transmission dynamics of zoonotic diseases like bTB [ 19 ]; for example, recent research on rabies has shown how empirical data can be used to elucidate epidemiological dynamics [ 21 ].

However, even in cases where empirical data is used, there may be limited power, which can impact results and interpretation. In these cases, evidence from empirical data can now be boosted by mathematical simulations, which are powerful tools for predicting disease transmission trajectories [ 22 ]. Simulations of disease transmission through compartmental models (e.g., the Susceptible, Infectious, and/or Removed (SIR) model and its variations) have been used in a variety of disease systems, including the recent COVID-19 pandemic. COVID-19, however, is exceptional in the level of global concern garnered and resultant significant investment in funding. This meant that large empirical datasets were also made readily accessible, which made direct complex modelling possible [ 23 ]. Other zoonoses are typically more difficult to model this way due to the lack of empirical data on disease transmission and associated hosts [ 24 ]. Mathematical simulations, using for example SIR models, therefore create opportunities to also model these zoonoses. In addition, such simulations allow us to undertake experiments that are currently logistically unfeasible, too costly, too complex or on “unobservable” phenomena [ 22 , 25 ].

As mentioned, lack of information on associated hosts and transmission pathways is often a limiting factor in modelling zoonoses and may potentially also be an issue in bTB research. Studying interactions between and within host species, as well as the role played by each host in the transmission chain, can enable us to better understand zoonotic disease dynamics. While simulations can achieve much, it is important to note that interactions amongst wild animals are heterogeneous by nature and vary significantly between different populations as well as individuals. Therefore, it is important to account for this variability to understand the mechanisms behind transmission and subsequently be able to predict and control disease spread [ 8 ]. This can be achieved by using network modelling, where heterogeneous contacts between animals can be used to simulate disease transmission [ 8 , 24 ], for example using social network analysis (SNA) [ 8 ]. SNA can be beneficial for disease management since it enables us to identify “super-spreaders” (i.e., highly connected individuals) which can then be targeted for vaccination, allowing for a dramatic reduction in transmission [ 1 , 8 ]. In addition, new research is looking at integrating SNA with molecular epidemiology (phylodynamics) to better estimate transmission pathways and direction of transmission between individuals [ 26 ].

Finally, it is of key importance that models of disease risk and distribution consider variances across space and time [ 27 , 28 ], which enables us to identify disease clusters [ 5 ] and model host abundance [ 29 ]. As ecological processes occur at different scales (from single study sites to macroecological scales), the spatial scale used for disease distribution modelling is crucial in understanding how these processes exacerbate the spread of zoonotic diseases, such as bTB [ 30 , 31 ]. Large spatial scales (i.e., global, continental) can examine the broader picture and disentangle how host abundance and abiotic factors influence disease prevalence [ 19 ]. Smaller spatial scales (i.e., country, region) can be used to examine population dynamics and pathogen genetic diversity at the local level [ 19 ]. Temporal patterns are important to consider as many zoonotic diseases show seasonal variations (e.g., Zoonotic enteric diseases such as Salmonella spp, Escherichia coli , Giardia spp) as well as daily variations (i.e., due to the circadian rhythm of microbes and pathogens as well as chronobiology of wildlife hosts) in their infection patterns [ 32 , 33 , 34 ]. It is of key importance that any gaps in bTB research pertaining to factors discussed above be identified, in order to inform future research direction.

Here, we aimed to uncover empirical and methodological gaps in the peer-reviewed literature on bTB. Our intention is to use bTB as an example of a complex multihost zoonotic disease for which recent developments with sampling design, animal monitoring tools and technology, and mathematical modelling has helped to fill the gaps in knowledge and improve our understanding and ability to combat zoonotic diseases more generally.

To achieve our goal, we developed a scoping review of bTB multihost epidemiology focusing on 18 research questions (reported in Table  1 and conceptually summarised in Figure 1 ) regarding the type of study, whether, which and how wildlife species have been monitored, what kind of sampling designs and methodological approaches have been used, and whether epidemiological empirical data have been collected. We then gathered data from the peer-reviewed literature on the mechanisms driving inter- and intraspecies bTB transmission, looking in particular at novel and multi-disciplinary approaches. Our goal is that our work will spark renewed discussion on how to monitor and deal with zoonotic diseases, direct future research, and stimulate focused funding efforts (Figure 2 ).

Key host species and topics of interest we screened for in the bovine tuberculosis scientific literature published between 1981 and 2022 . bTB host species include cattle as well as a range of wild species: badger, wild boar, cervid species (with the following species identified in the literature screened: white-tailed deer, red deer, fallow deer, roe deer, wapiti elk, sika deer and muntjac deer), brush-tailed possum and wild buffalo. The circles on the outside illustrate the key information sought in peer-reviewed papers dealing with bTB, which has been expanded and clarified in Table  1 : type of data collected by researchers; whether spatial analyses were carried out (i.e., in cattle and or wildlife); what type of spatial and temporal scales were considered; whether environmental variables were taken into account (i.e., environment in the farm, environment around the farm and/or weather variables); whether the methodological approach captured the direction of disease transmission; whether the study used common epidemiological modelling techniques (i.e., compartmental models, transmission rates), or whether the study included intra/interspecies interactions in their methodology (i.e., what type of interactions did they look at - e.g., direct and/or indirect, what type of equipment was used to get interactions data and what methodology was used to analyse the data); finally, if human perturbations (i.e., forest felling, culling, vaccination) were taken into account when looking at variables affecting bTB spread, and management solutions to offset the spread of bTB, if any. Animal silhouettes were downloaded from PhyloPic [ 134 ]. Cattle, cervid, brushed-tailed possum and wild boar silhouettes are under: CC0 1.0 Universal (CC0 1.0) Public Domain Dedication. Buffalo silhouette is by Jan A. Venter, Herbert H. T. Prins, David A. Balfour & Rob Slotow (vectorized by T. Michael Keesey) under: Attribution 3.0 Unported (CC BY 3.0) [ 135 ]. Badger silhouette is by Anthony Caravaggi under: Attribution-NonCommercial-ShareAlike 3.0 Unported (CC BY-NC-SA 3.0) [ 135 ]

Cascade diagram of the process used in the selection of relevant papers.

We conducted a scoping review (as per PRISMA guidelines) [ 35 ] by sourcing peer-reviewed papers using Web of Science (Clarivate, 2021 Online Version) focusing on bovine tuberculosis, and more specifically its most common cause, Mycobacterium bovis , in cattle and several key wildlife hosts. The search terms and list of articles have been summarised in Additional file 1 . We identified 3531 potentially relevant papers (i.e., the search included all years of publication) which were uploaded and screened for duplicates using EndNote (Clarivate, Version 20.1.0.15341)(Figure  2 ). Relevant articles were then selected using a PEO (Population, Exposure, Outcomes) eligibility criterium structure [ 36 ]. The aim of the PEO is to identify articles of interest by selecting the “Population” (i.e., the subject being affected by the disease/health condition) for a particular “Exposure” (i.e., a disease/health condition) and either a particular “Outcome” or “Themes’’ to examine [ 36 , 37 ]. The PEO eligibility criterium was chosen since it was in line with the recommendations given for scoping reviews that target literature on etiology and risk factors, such as a particular disease. We decided to use a modified version of the PEO framework structure which also includes themes of interest as potential “Outcomes” [ 37 ], as summarised in Table  2 to aid reproducibility. All papers that did not meet the eligibility criteria listed in Table  2 were removed (Figure 2 ). The papers were screened by one researcher who coded 18 variables (stored in an excel spreadsheet) to answer the questions of interest summarised in Table  1 . The results were then imported and plotted using ggplot2 in R version 4.1.1 [ 38 ].

Our results are based on 532 peer-reviewed papers published between 1981 and 2022. The study location of the papers was representative of 6 continents and 52 different countries (Figure 3 ). The continent with the highest number of studies on bTB is Europe ( n  = 303, 169 of which were from the UK), significantly higher than those carried out in much larger continents such as Africa, Asia, and both Americas (Figure 3 ). We screened all papers for 18 different variables (addressing our 18 questions, see Table  1 ) which we summarised in the following section under the heading: 3.1 general characteristics (Sub-headings: “Study species and wildlife species”; “Management and data type”), 3.2 data analysis (Sub-headings: “Spatial analysis, spatial scale and temporal scale”; “Farm environment and human perturbations”), and 3.3 epidemiological analysis (Sub-headings: “Intra- and interspecies interactions”, “Direction of transmission and compartmental models”). Note that most plots presented below have a sample size of n  = 532, corresponding to the number of papers screened, with a few exceptions where this sample size is higher (for example, in relation to temporal scale included in the study, if a paper reported multiple temporal scales, therefore contributing to multiple levels of a category) or lower (for example, in relation to epidemiology, where variables of interest were analysed only in the subset of papers describing studies that included epidemiological interactions).

World map showing number of papers screened per country . Number of papers per continent: Europe (303), Africa (68), Oceania (60), North America (53), South America (29), Asia (26).

3.1 General characteristics

3.1.1 study species and wildlife species.

We found that 41% of bTB papers focused on cattle only, whereas 30% of them included both cattle and wildlife species and 29% targeted only wildlife species (Figure 4 A). Among those papers reporting wildlife data, we found that the European badger attracted most research effort (50% of wildlife studies), followed by cervid species (28%: 13% red deer, 11% white-tailed deer, 5% fallow deer, 3% roe deer, 2% wapiti elk, from hereinafter referred to as simply elk, and < 1% of studies including sika and muntjac deer), wild boar (18%), brushed tailed possum (17%) and buffalo (4%) (Figure 4 B).

Species, data and study type . Number of papers screened and reporting data on A study species type (whether the study was on cattle and/or wildlife), B wildlife species, C management (whether a paper investigated potential management solutions and their efficacy), D and data type. Animal silhouettes were downloaded from PhyloPic [ 134 ]. Cattle, cervid, brushed-tailed possum and wild boar silhouettes are under: CC0 1.0 Universal (CC0 1.0) Public Domain Dedication. Buffalo silhouette is by Jan A. Venter, Herbert H. T. Prins, David A. Balfour & Rob Slotow (vectorized by T. Michael Keesey) under: Attribution 3.0 Unported (CC BY 3.0) [ 135 ]. Badger silhouette is by Anthony Caravaggi under: Attribution-NonCommercial-ShareAlike 3.0 Unported (CC BY-NC-SA 3.0) [ 135 ].

3.1.2 Management and data type

Our results highlighted that only 25% of the studies dealt with management solutions (Figure 4 C). Management strategies mainly included culling (18%) or vaccination (6%), with 5% looking at other strategies (e.g., fencing, sterilisation). We also found that most papers gathered original empirical data (79%), and papers only using simulations were limited (4%), with a remaining 17% of papers combining empirical data and simulations (Figure 4 D).

3.2 Data analysis

3.2.1 spatial analysis, spatial scale and temporal scale.

We found that the majority of papers did not include any spatial analysis. Those that did focused on spatial patterns in wildlife (30%, Figure 5 B) slightly more than cattle (28%, Figure 5 A). Among the 149 papers that investigated spatial analysis in cattle, 58% looked at bTB risk and probability of infection; 16% looked at cattle interactions with wildlife, 13% analysed the spatial distribution of bTB positive biological samples, 11% investigated cattle movement outside the farm. Interactions between farm animals and cattle movement inside the farm were included in 5% and 1% of papers, respectively. Among the 161 studies which investigated spatial behaviour in wildlife (Figure 5 B), analysis was undertaken using a variety of methodologies; direct observations (36%), satellite GPS telemetry (19%), spatial patterns predicted by future scenarios modelling or mathematical simulations (19%), genetic samples (11%), camera traps (7%), proximity loggers (4%) and indirect observations (e.g., faecal samples for population density estimations; 1%).

Spatial and temporal analysis . Number of papers screened and reporting data on A spatial analysis of cattle (whether the study included any type of spatial analysis), B spatial analysis of wildlife, C spatial scale, and D temporal scale.

We also found that most papers included spatial scales at the regional level or smaller (72%), with less than 4% papers looking at national and/or international spatial scales (Figure 5 C). In regard to temporal scales, 36% of the studies considered interannual variability, whereas 17% tackled intra-annual variability. Thirty-six percent of the studies did not analyse any intra- or interannual temporal variability (Figure 5 D). Only 4% of the studies looked at fine-scale variability (e.g., days), whereas in a few instances the year of study was not reported at all (4%). Finally, 12% of papers included predictions for temporal patterns into future scenarios.

3.2.2 Farm environment and human perturbations

When looking at farm characteristics, 50% of the studies included some type of herd data (e.g., herd size, bTB history), with 44% not including any type of in-farm environmental variables (Figure 6 A) and 24% of papers incorporating other types of farm characteristics. These included environmental conditions on the farm (e.g., natural habitats, land-fragmentation; 13%), farm location in respect to other farms (11%) and farm location in respect to wildlife (6%).

Environmental variables . Number of papers screened and reporting data on A in-farm environment (whether the paper analysis included variables explaining environmental characteristics inside the farm), B outside farm environment, C human perturbations (whether the paper analysed the effect of human disturbances on bTB transmission dynamics), and D weather variables.

Environmental conditions outside the farm were included in 33% of the papers’ data analysis (Figure 6 B). These studies mainly looked at habitat characteristics around the farm (e.g., wildlife presence, natural habitats), with two papers also including variables focusing on habitat variation (e.g., forest clearfell, new artificial plantations). We also looked at weather variables (e.g., temperature, rainfall) and observed that 4% of papers included these as part of their analysis (Figure 6 D). Finally, 25% of the papers screened included human perturbation variables with the vast majority looking at the effect of vaccination and culling on transmission dynamics (Figure 6 C).

3.3 Epidemiological analysis

3.3.1 intra- and interspecies interactions.

We found that most papers (69%) did not include an analysis on interactions, with 25% of papers looking at intraspecies transmission and 14% at interspecies transmission (Figure 7 A). Among the interaction studies, 33% included direct interactions, 17% included indirect interactions and 51% included both (Figure  7 B). In addition, interaction data were mostly collected using simulations (39%), followed by technological tools (29%; e.g., GPS, proximity loggers, camera traps), and direct observations (23%), with genetic sampling used in 7% of papers (Figure 7 C). The methodology used to analyse interaction data also varied between papers with 28% of papers using differential equations (e.g., SIR models, discreate models), 19% social network analysis, 18% linear models (i.e., including generalized mixed models as well as simple linear regressions) and 38% using a variety of statistical techniques (e.g., t-test/ANOVA, stochastic models) (Figure 7 D).

Interaction analysis . Number of papers screened and reporting data on A interactions (inclusion of interaction analysis i.e., intra- and/or interspecies interactions), B interaction type, C the way interactions were monitored, and D interaction data analysis statistical approach.

3.3.2 Direction of transmission and compartmental models

We found that a limited proportion of the papers (8%) included direction of transmission in their analysis (Figure  8 A). We also found that epidemiological modelling techniques (e.g., compartmental models and transmission rates) were adopted in 15% of the studies (Figure 8 B).

Epidemiological analysis . Number of papers screened and reporting data on A direction of transmission (whether this was analysed in the paper, e.g., transmission across species), B epidemiological modelling (i.e., papers included compartmental models and/or transmission rates in the analysis).

4 Discussion

In this review we found that there has been significant research focusing on the badger-cattle bTB episystem. We acknowledge, however, that we also found a very limited number of studies on other episystems [ 39 , 40 , 41 ]. Our spatially-explicit overview of bTB research efforts (Figure 3 ) highlights how the badger-cattle episystem has been the focus of most research done to date, highlighting a huge amount of money and research effort on bTB transmission dynamics across Europe and particularly in Britain and Ireland. However, there has been far less attention given on other multi-host episystems of countries in southern Africa, Asia and both South and North America. We believe we have more to learn from these chronically understudied systems.

Our scoping review found a limited number of studies focusing on management solutions and their efficacy, with very few looking at modelling exit strategies [ 42 , 43 ]. This is due to the paucity of studies using mathematical simulations, not only to better understand and predict possible outputs of management solutions, but also to explore long-term bTB dynamics under different scenarios (e.g. [ 44 , 45 ]). Only a small number of studies have looked at the effect of human disturbances on the spread of bTB in wildlife host species, and this knowledge gap needs to be tackled as we are aware that human perturbations may exacerbate zoonotic outbreaks and spread [ 46 , 47 , 48 ]. Most of the studies we reviewed have focused on the effect of badger vaccination and culling on bTB dynamics with only three studies looking at how other human perturbations may affect these dynamics [ 49 , 50 , 51 ]. Additionally, only two focused on the effect of habitat change (e.g., clearfell forest operations) on bTB breakdowns [ 52 , 53 ]. Finally, we observed that there is only a few studies looking at the effect of weather variables (i.e., rainfall, soil humidity, temperature etc.) on bTB spread or risk [ 54 ]. This is especially important when considering wildlife-cattle transmission since it is now thought to occur also through environmental sources [ 55 ].

We have carefully evaluated the outcome of our scoping review, and in the following sections we have summarised data types and methodological approaches which, we believe, could contribute to gaining further insights into bTB epidemiology. Based on our review, we have identified a significant gap when it comes to prediction and simulation models, which would be a useful tool for managers to assess disease risk under different land use and climate change scenarios. Another major gap is the lack of integration between empirically-informed tactical (short-term decision support) and strategic (larger spatial scales and longer term) models being used concurrently in single studies (though we do note that there are exceptions, for instance Brooks-Pollock et al. [ 56 ]). Future research should include compartmental models fitted across space, linked via meta-populations and/or real-landscape multi-host episystems; or agent-based models (ABMs) with empirical data feedback loops. We describe such modelling approaches and their prerequisites in the following sections, beginning with data and monitoring programs, and we continue with recent advances in technology, mathematical tools and analytical solutions.

4.1 Empirical data and long-term monitoring programs: involving stakeholders and setting up fixed long-term monitoring stations across large spatial scales

As good quality data is required to generate informed strategies on wildlife interventions, we need reliable data sources to model spatial distribution and abundance of the host species involved in transmission. In reference to the badger-cattle bTB episystem in western Europe, both badgers (with several examples among the literature: [ 57 , 58 ]), and cattle [ 59 ] have been extensively monitored. However, in some populations, it is possible that deer and wild boar may also play a role in the local spread and maintenance of infection [ 60 ]. In Britain and Ireland, the significance of deer as a wildlife host impacting bTB epidemiology has been uncertain [ 61 ]. However, recent research is starting to uncover the role deer may play at local scales where conditions favour the transmission between badgers-deer-cattle [ 62 ]. There could be opportunities to gather data in collaboration with hunters (as has occurred in France [ 63 ] and Spain [ 64 ], for example) to have access to a high number of deer samples, within and across countries (large spatial scales) and across years (long-term temporal scales). Involving stakeholders like hunters may provide the unique opportunity to collect pictures of clearly infected animals (e.g., small to large white, tan, or yellow lesions on the lungs, rib cage, or in the chest cavity) - to be submitted via smartphone applications (see [ 65 ]). These stakeholders may also gather biological samples to be collected by government officials at ad hoc collecting centres. This type of information would boost opportunities for monitoring the dynamics of the disease across multiple spatio-temporal scales and in relation to bTB occurrence in the two other hosts in the system (badgers and cattle). The ability to involve stakeholders across large spatial scales (e.g., hunters, farmers, foresters) may help to establish systematic, relatively inexpensive, and long-term monitoring programmes. These can provide species presence-only and presence-absence data for Bayesian species and disease distribution models (described in Sect. " Modelling and mathematical simulations: social network analysis Bayesian species distribution models, and agent based models "), allowing managers to access up to date risk scenarios. This approach can also highlight hotspots of disease outbreak that could drive focused longitudinal studies using satellite telemetry on multiple species simultaneously. This would enable us to better disentangle species overlaps and contact rates [ 66 , 67 ]. The role of stakeholders/citizen scientists in this bTB example could be confirming infection, which is almost never inexpensive, although there is the hope that cheaper field tests will be released in the next decade. The veracity of the data collected and level of engagement from stakeholders/citizen could also be a problem which needs to be taken into consideration. For the time being, a well distributed number of samples could be collected from hunters to cover large areas systematically and limit the costs required for testing.

When it comes to establishing long-term monitoring programs, fixed long-term sampling stations across large-spatial scales can capture wildlife population spatio-temporal dynamics. This can, on one hand, provide data on occurrence, relative density, and spatio-temporal overlaps of the host species and, on the other hand, gather key empirical data required to parameterise mathematical simulations. Camera traps are a popular and effective tool for estimating state variables of wildlife populations [ 68 ]. For ungulates, they have successfully been used to understand temporal behaviour (e.g., diel activity patterns, [ 69 ]), spatial behaviour (e.g., occupancy, [ 70 ]), and abundance (e.g., density, [ 71 ]). Camera traps have been used for quantifying temporal and spatial overlap of wild ungulates with domestic animals in open systems [ 72 , 73 ] with varying results [ 74 ]. Kukielka et al. demonstrated their use in identifying hotspots of indirect wildlife–livestock overlap for the prevention of bTB crossover [ 72 ]. For wildlife, especially ungulates, camera traps offer powerful monitoring solutions not only to measure abundance and spatial overlap, but also to understand behavioural dynamics that may align closely with disease risk. An example is the use of camera traps to individually recognise animals, which has been shown to be possible in a recent study by Hinojo et al. [ 75 ]. They demonstrated how roe deer ( Capreolus capreolus ) antler shapes could be used to identify distinct individuals. This data could be used to obtain better estimates of abundance as well as to build wildlife social networks (which will be discussed in more detail in Sect. " Modelling and mathematical simulations: social network analysis, Bayesian species distribution models, and agent based models ") and therefore provide information on contact rates between and within species. The parameters from these analyses would be useful as an input for mathematical simulations to help better understand disease transmission dynamics in wildlife populations.

The use of camera traps as well as satellite telemetry can be quite challenging to use in developing countries since they can be extremely expensive (satellite telemetry, in particular) as well as difficult to use when collecting data in remote locations (camera traps, in particular). In addition, the invasive nature of satellite telemetry - which requires trapping animals - often makes it hard to collect data from enough individuals from an ethical, logistical and administrative points of view. Therefore, to improve our understanding of episystems in developing nations, advances in non-invasive diagnostic techniques and eDNA (i.e., a genetic sampling technique that uses environmental sources - such as water and soil - to extract genetic information used for biosecurity and biomonitoring purposes) are essential [ 76 , 77 , 78 , 79 ]. An example of a widely used non-invasive sampling technique is faecal sampling [ 76 , 77 , 78 , 79 , 80 , 81 ]. Faecal samples are a relatively inexpensive way of monitoring diseases and health status in wildlife species. It is also possible to collect a high number of samples in a short period of time, which is especially important for long-term monitoring programs of wildlife hosts. Collecting eDNA can be even faster and is especially useful for long term spatio-temporal dynamics of infectious pathogens at the wider scale, which can improve the monitoring of zoonoses at the country and continental level [ 77 ].

However, timing is key when monitoring diseases as infectious pathogens can mutate and be rapidly transmitted between wildlife, humans, and domestic animals, with potentially devastating impacts on human health and animal welfare. Therefore, novel and rapid genetic techniques, such as culture-free pathogen genetic sequencing [ 82 ], can greatly benefit disease surveillance by decreasing the time needed to sequence pathogens and, consequently, the time needed to make essential ecological management decisions and activate public health responses. In addition, these new sequencing technologies can be very useful during wildlife field studies in isolated areas since they can be rapidly deployed and need limited laboratory equipment for processing [ 82 ]. In addition, when monitoring zoonosis such as bTB and collecting related data (invasively or not) it is important to recall the characteristics of the bacterium itself, Mycobacterium bovis . For example, different lineages exist across the globe [ 83 ] with different strains potentially showing different evolutionary [clock] rates. This greatly affects the rate at which the bacterium needs to be monitored among countries, and we believe that faster sequencing technologies will be of great help in tracking the evolution and spread of different lineages, informing adaptive management of bTB (and zoonosis in general) at the local level.

4.2 Recent advances with technology can help to gather data for mathematical simulations: interindividual variability within animal populations and human socio-economic factors matter and should be taken into account

Animal-attached sensors, i.e., biologging [ 84 , 85 ], can allow us to disentangle animal behaviour and the movement patterns that promote disease transmission. GPS units are the most widely used of these sensors, providing data on animal space use. Proximity sensors can detect when two or more sensor-equipped animals interact and can be used to detect direct encounters which may result in disease transmission. Collars with both GPS units and proximity sensors have been used concurrently on badgers and cattle uncovering that, while badgers show a habitat preference for cattle pastures, there were rare to no direct contacts between the two species [ 86 , 87 ]. This indicates that environmental transmission may play an important role in the case of bTB [ 87 ]. As such, proximity sensors allow insights which are not obtainable through investigating shared space use alone. When the disease state of an individual is known, proximity sensors can also provide information on if and how the duration of exposure to said individual affects transmission rate to other members of the population [ 88 ]. Other biologging sensors, including accelerometers, magnetometers, and gyroscopes, are used to classify distinct behaviours from logged datasets [ 85 ]. Behaviour classification allows activity budgets to be built so that behaviours which increase the likelihood of acquiring or transmitting pathogens can be detected and mapped in the landscape. Accelerometers have also been used to compare micro-movements in diseased and healthy animals, with diseased animals exhibiting differences in posture, gait dynamism (e.g., the “bounce” in subsequent walking steps) and energy levels [ 89 ]. Monitoring such micro-movements in cattle could act as a warning sign to test herds for bTB when signs of illness are detected, e.g. by adapting existing systems in place to monitor lameness through accelerometry [ 90 ]. These effects of disease on the internal state of animals yield important insights into how disease status impacts animal movement patterns and therefore disease spread.

Biologging and satellite telemetry monitoring can, on one hand, provide answers aimed at understanding the transmission dynamics within multi-host disease systems [ 87 , 91 , 92 ] and, on the other hand, provide highly valuable empirical data that are strongly needed by parameter hungry mathematical simulations [ 88 ]. However, when tracking animals, special care should be taken to understand the behaviour of those animals that we are monitoring, and specifically whether we are following a bolder subset of the overall population that are easier to trap. This applies also to where we study animals which will provide empirical data for mathematical simulations, because behaviour and movement ecology may vary significantly depending on the level of human disturbance. We are aware that tracking multiple individuals of multiple species can be expensive and not accessible unless large amounts of funding is available. However, recent technological advances with satellite telemetry using LoRaWAN transmission technology [ 93 , 94 ] have been developed to monitor livestock at affordable prices (e.g. less than 100 euros for 1 GPS unit), opening up new opportunities for extensive monitoring programmes in wildlife, within and across species.

The concept of One Health has highlighted the role that human activities play in the spread of zoonotic diseases [ 95 ]. For example, urbanisation, improper waste disposal, and the intentional feeding of wildlife have been shown to result in wildlife movement into human-dominated areas [ 7 ], which may facilitate disease transfer to humans and other animal communities [ 96 ]. However, evidence has shown that only a select proportion of individuals within wildlife populations will engage in interactions with humans [ 97 ] or utilise these human-dominated areas [ 7 , 98 ]. Individual variation in movement patterns [ 99 ], sociability [ 100 ], and immunological defence [ 101 ], among others, impacts disease transmission and spread [ 102 ]. There is also evidence that certain behavioural types have higher infection rates than others (e.g. [ 103 , 104 ]), although the causal direction may be difficult to determine since infections also alter host behaviour [ 103 , 105 ]. Regardless, to gain a more complete understanding of disease spread, future studies should incorporate this individual variation. These studies often utilise direct behavioural observations, since these are an invaluable data source that can be used to determine which individuals in a known population are more likely to engage in close-contact interactions with humans [ 97 ] or access human areas (e.g., farmland) [ 106 ]. This can provide us with information on which individuals in a population may be at “higher risk” of transferring disease to humans or to other animal populations.

Nevertheless, considering human behaviour is also fundamental in infectious disease transmission. The One Health definition has changed in 2022 accordingly and now it includes the importance of society and its diversity in values and beliefs in effectively fighting zoonoses [ 107 , 108 ]. Collaboration between scientific disciplines is not enough to improve current and emerging infectious disease transmission. It is fundamental that community members and expertise at every level, from village to continent, be included if we wish to equitably improve human health and animal welfare [ 107 ]. In this way we may also improve the effectiveness of disease management solutions by tailoring them to communities instead of trying to use the same solutions in different areas without taking into account socio-economical differences.

4.3 Modelling and mathematical simulations: social network analysis, Bayesian species distribution models, and agent based models

Social network analysis (SNA) is a powerful tool in uncovering the causes and consequences of disease transmission within animal communities [ 109 , 110 ]. In the past decade SNA has mainly focused on understanding contact and transport networks of cattle and livestock movements, as well as wildlife movements [ 111 , 112 , 113 ]. Nonetheless, it could be expanded to better unravel the dynamics of disease transmission between wildlife populations and livestock [ 110 ]. Unlike in domestic cattle, the movements and interactions of wildlife can be challenging to track. As a result, a small proportion of individuals are typically monitored using biologging and satellite telemetry, as discussed earlier. Recent advances in statistical analysis of social networks have paved the way to obtain better inferences from limited data [ 114 , 115 , 116 ]. The first step is to identify the network metrics affecting disease transmission dynamics that best suits the disease system under study (e.g., transitivity, betweenness centrality) [ 114 , 115 ]. Using global metrics of a social network, for example, can help estimate potential changes in the overall structure of the wildlife population. A commonly used global metric when studying disease transmission dynamics is transitivity, which represents the tendency of a population to cluster together and is considered to be negatively correlated with disease transmission rates [ 113 ]. Local network metrics, on the other hand, can help in understanding social characteristics at the individual level. A type of local metric is betweenness centrality, which represents the tendency of an individual to serve as a bridge between one part of the community and another (i.e., a community in SNA is a group of nodes, for example individual animals, with denser connections between each other compared to other nodes in the network), helping the selection of individuals to be vaccinated/removed from the population.

Once we have selected the metrics to use, they can be tested via pre-network permutations of available observations to ensure that the available data sufficiently captures non-random interactions among the animals. However, when using small samples for SNA we also must be careful on what we infer from it. Recent research [ 115 ] has shown that estimates may be inaccurate, or “noisy”, at low sample sizes. Therefore, stable metrics with respect to low sample sizes should be identified before making inferences. Research on data collected for wild ungulates [ 115 ], for example, shows that the betweenness centrality values of smaller samples remain well correlated with those in larger samples, indicating that this metric can be used even when the social network is built using a small sample of the population. Similar correlation analysis can be done for other network metrics, mainly in cases of limited data availability for disease transmission. Whenever limited animals from a population are monitored, confidence intervals around the network metrics should also be obtained to make informed decisions using statistical evidence.

Using the methodologies discussed above (see Silk et al. and Kaur et al. for more details, [ 113 , 115 ]) we now have the possibility of analysing all telemetry data collected thus far on species involved in bTB transmission (e.g., badgers, wild boar; but also applicable to species from other disease systems) to test hypotheses on disease transmission dynamics. For example, we can now use these statistical techniques to better understand behavioural patterns of wildlife species, as well as comparing networks overtime and how wildlife behaviour can be affected by perturbations in the environment (e.g., climate change, land-use change or other type of anthropogenic factors) even when only limited data is available [ 115 ]. In addition, it will help in collecting future data since these methodologies can be used to estimate the minimum number of individuals needed in order to reliably build a social network, which can vary enormously depending on the scope of the project as well as the wildlife species of interest. This will, for example, help in answering specific questions regarding the role of deer species in bTB transmission by simultaneously collecting telemetry data on badgers and deer species in Ireland.

Knowing the distribution and abundance of wildlife vectors (i.e., a living agent that carries and transmits pathogens - e.g. HIV, Covid-19, bTB - to other living beings) is also essential when aiming to reduce zoonotic risk [ 117 , 118 ]. To that aim, Species Distribution Models (SDMs) can be used to produce models of the distribution and abundance of species based on occurrence data [ 119 ]. In recent years spatial modelling has undergone a conceptual and technical revolution. New modelling techniques within Bayesian [ 120 ] and Machine Learning frameworks [ 121 ] allow us to develop spatially explicit models of animal abundance and distributions with unprecedented accuracy, and the improvement of computational power allows computers to rise to the challenge and cope with the high computational demands of these models. The flexibility of the new techniques allows us to use different types of data (e.g., individual tracking data, survey data, and even citizen science data) and combine them in what are called Integrated Species Distribution Models (ISDMs), while still taking into account the different observational processes of each type of data, to produce accurate models even in data scarce systems [ 122 ]. In addition, these new techniques also allow for the calculation of uncertainty in a spatially explicit manner, which will help us evaluate the quality of the models and better interpret the results. Bayesian ISDMs using INLA (i.e., Integrated Nested Laplace Approximation) [ 123 ] were used to model the distribution of red, sika and fallow deer in Ireland, which are vectors of bTB [ 65 ]. The models produced, for the first time, relative abundance and distribution maps for each species, which will be an essential tool for deer population management and thus towards bTB eradication. They are already being used to determine high sika-density areas for a pilot study on the effect of deer on biodiversity, which will provide further management tools for the overabundant deer populations in Ireland. In addition, hierarchical Bayesian models are also the basis of a new project aimed at modelling European badger sett distribution, badger density, and their body condition. These three models will be linked to bTB infection in badgers and outbreaks in cattle, in an attempt for the first time to link badger spatial ecology to bTB management and eradication in Ireland (V. Morera-Pujol 2023, personal communication).

Agent-based simulations are another useful modelling approach, or complementary tool to traditional methods, when data is limited/not available; helping elucidate transient effects of wildlife disease transmission in human-dominated landscapes [ 25 ]. These models serve as a computational laboratory that allow researchers to plug-in available real-world data and parameterise both agents (for instance, a badger) and the environment (for instance, a mosaic of natural habitats and farms). This enables researchers to empirically test if animal behaviour in response to landscape change or management interventions modulates disease risk dynamics over time and space [ 124 ]. Recent technological advancements have bolstered agent-based simulations allowing for high-resolution spatio-temporal models that incorporate geographic information systems (GIS) data to create hyper realistic environments, and machine learning algorithms to introduce cognition and applied decision making for agents. Furthermore, process-driven agent-based models (e.g., disease transmission) can be integrated into larger mechanistic agent-based models (e.g., ecosystem scale epi-dynamics) for increasingly higher-resolution models that reduce uncertainty and overly-theoretical parameterisation of model entities [ 25 ]. The development of highly-realistic agent-based simulations, parameterised with high-resolution data, for the management of bovine tuberculosis in multi-host systems can contribute to answering important policy questions and how best to select management directions. In practice, this allows for the totality of data collected in complex multi-host systems to be incorporated into a single environment where they may be measured against one another in the simulation to deduce the possible effects of each predictor. Take for example the European badger as the primary wildlife host in Ireland as a case study. Badgers are prevalent in the agroecological mosaic of natural habitats and farms in Ireland. Agent-based simulations can utilise data from badger tracking studies [ 51 , 125 , 126 ], habitat suitability [ 127 ], culling and vaccination programmes [ 128 ] and disturbance regimes [ 52 , 129 ] to simulate badger movement and behaviour realistically. GIS data for farm location and characteristics [ 130 ], as well as ecological and environmental data streams, can then place the badger agent into a highly realistic environment to examine how these factors affect badger movement, behaviour, and other parameters, for instance, contact rates with domestic animals. Interactions between agents and the environment can be modulated by sub-models to further increase the strength of the model. For instance, weather sub-models (e.g., rainfall) may influence agricultural practice and thus contact rates, as well as the length of time Mycobacterium bovis persists in the environment. Alternatively, disease transmission could also be sub-modelled so that contact rates may/may not result in infection [ 25 ]. Finally, management decisions can be trialled within the simulation to see how likely decisions change the status of disease within the study system, allowing for “What if?” scenarios to play out without risk to animal or human welfare or livelihoods.

5 Conclusion

Our exploration of the recent literature on multi-host bTB episystems, as an example of zoonotic One Health challenges, has revealed a significant body of work utilising a diversity of methodologies at different spatio-temporal scales and subjects (individual vs. group) levels. There was a significant bias in the literature towards one particular episystem, the badger-cattle system that predominates in north-western Europe, reflecting large financial burdens (for both governments as well as the agricultural sector) and research funding investments. Alternatively, there were comparatively less publications from the global south, especially in complex muti-host episystems in southern Africa and India. In such episystems, the cost-effective and efficient collection, collation, and use of data are essential to drive greatest added value to inform on policy options.

Given the results from our scoping review, we reflect on several areas where progress could be made. This includes the need for high-quality data on wildlife hosts, even in episystems where significant research investments have already been made. Such careful collection and utilisation of empirical data could then help feed the development of social network analyses, Bayesian distribution models and eventually mathematical and simulation-based models. Mathematical simulations, such as ABMs trained on synthetic data and parameterised by real empirical data, can answer questions that would otherwise be too costly, unethical, or both. Such models can also be used to explore different scenarios in an increasingly human-dominated world, under different levels of land-use and climate change, or with the appearance of invasive species in already complex multi-host epidemiological systems. In addition, it can help build cross-disciplinary bridges with other areas, deriving significant insights into interspecific transmission like phylodynamic modelling.

We have used our Irish experience to inspire researchers from across the globe; Ireland invests considerably in surveying, culling, and vaccinating badgers [ 131 , 132 ]. However, the question remains - which applies to other countries and zoonotic episystems - should we be doing more or can we be smarter with the data we already have? We suggest the latter. Yes, there is a need to be smarter, arranging ad hoc data collections using the latest technological tools to estimate unknown or uncertain parameters. But we also have to focus our efforts on mathematical modelling (ABMs, INLA-Bayesian) to optimise our information gain from the large, high-quality datasets collected over the last few decades (and for sparser datasets, taking advantage of recently developed statistical tools for enhanced inferences, see [ 54 , 68 ]). We have (almost) all the data required to parameterise simulations with significant utility: this should be one main focus in future research. We believe that, ideally, the feedback of simulation and mathematical tools to inform data collection, and the “virtuous cycle” of feeding this new data to improve the next generation of model is a priority for decision making tools for policy makers and programme managers.

Availability of data and materials

The datasets generated and/or analysed during the current study are available in the “A scoping review on bovine tuberculosis highlights the need for novel data streams and analytical approaches to curb zoonotic diseases” repository [ 133 ].

Destoumieux-Garzón D, Mavingui P, Boetsch G, Boissier J, Darriet F, Duboz P, Fritsch C, Giraudoux P, Le Roux F, Morand S, Paillard C, Pontier D, Sueur C, Voituron Y (2018) The One Health concept: 10 years old and a long road ahead. Front Vet Sci 5:14

Article   PubMed   PubMed Central   Google Scholar  

Jones KE, Patel NG, Levy MA, Storeygard A, Balk D, Gittleman JL, Daszak P (2008) Global trends in emerging infectious diseases. Nature 451:990–993

Article   CAS   PubMed   PubMed Central   Google Scholar  

Karesh WB, Dobson A, Lloyd-Smith JO, Lubroth J, Dixon MA, Bennett M, Aldrich S, Harrington T, Formenty P, Loh EH, Machalaba CC, Thomas MJ, Heymann DL (2012) Ecology of zoonoses: natural and unnatural histories. Lancet 380:1936–1945

EFSA and ECDC (European Food Safety Authority and European Centre for Disease Prevention and Control) (2021) The European Union One Health 2019 zoonoses report. EFSA J 19:6406

Google Scholar  

Beard R, Wentz E, Scotch M (2018) A systematic review of spatial decision support systems in public health informatics supporting the identification of high risk areas for zoonotic disease outbreaks. Int J Health Geogr 17:38

Jones BA, Grace D, Kock R, Alonso S, Rushton J, Said MY, Mckeever D, Mutua F, Young J, Mcdermott J, Pfeiffer DU (2013) Zoonosis emergence linked to agricultural intensification and environmental change. Proc Natl Acad Sci USA 110:8399–8404

Lowry H, Lill A, Wong BBM (2013) Behavioural responses of wildlife to urban environments. Biol Rev 88:537–549

Article   PubMed   Google Scholar  

Craft ME (2015) Infectious disease transmission and contact networks in wildlife and livestock. Phil Trans R Soc Lond B Biol Sci 370:20140107

Article   Google Scholar  

Johnson CK, Hitchens PL, Pandit PS, Rushmore J, Evans TS, Young CCW, Doyle MM (2020) Global shifts in mammalian population trends reveal key predictors of virus spillover risk. Proc Biol Sci 287:20192736

PubMed   PubMed Central   Google Scholar  

WHO, FAO, OIE (2017) Roadmap for zoonotic tuberculosis. WHO, Geneva

DAFM (2023) National bovine TB statistics. https://www.gov.ie/en/publication/5986c-national-bovine-tb-statistics-2020/ . Accessed 28 Mar 2024

DAFM (2019) Spending review 2019. Animal health: TB Eradication. DAFM, Dublin

DEFRA (2018) Bovine TB strategy review. DEFRA, UK

Corner LAL, O’Meara D, Costello E, Lesellier S, Gormley E (2012) The distribution of Mycobacterium bovis infection in naturally infected badgers. Vet J 194:166–172

Vicente J, Höfle U, Garrido JM, Fernández-De-Mera IG, Juste R, Barral M, Gortazar C (2006) Wild boar and red deer display high prevalences of tuberculosis-like lesions in Spain. Vet Res 37:107–119

Allen AR, Skuce RA, Byrne AW (2018) Bovine tuberculosis in Britain and Ireland—a perfect storm? The confluence of potential ecological and epidemiological impediments to controlling a chronic infectious disease. Front Vet Sci 5:109

Twelve Manhattan principles formulated. https://www.onehealthcommission.org/index.cfm/37526/75802/twelve_manhattan_principles_formulated . Accessed 26 Feb 2019

Zinsstag J, Schelling E, Waltner-Toews D, Tanner M (2011) From “one medicine” to “one health” and systemic approaches to health and well-being. Prev Vet Med 101:148–156

Estrada-Peña A, Ostfeld RS, Peterson AT, Poulin R, De La Fuente J (2014) Effects of environmental change on zoonotic disease risk: an ecological primer. Trends Parasitol 30:205–214

Castillo-Neyra R, Zegarra E, Monroy Y, Bernedo R, Cornejo-Rosello I, Paz-Soldan V, Levy M (2017) Spatial association of sanine rabies outbreak and ecological urban corridors, Arequipa, Peru. Trop Med Infect Dis 2:38

Fisher CR, Streicker DG, Schnell MJ (2018) The spread and evolution of rabies virus: conquering new frontiers. Nat Rev Microbiol 16:241–255

Siettos CI, Russo L (2013) Mathematical modeling of infectious disease dynamics. Virulence 4:295–306

Adam D (2020) Special report: the simulations driving the world’s response to COVID-19. Nature 580:316–318

Article   CAS   PubMed   Google Scholar  

Craft ME, Caillaud D (2011) Network models: an underutilized tool in wildlife epidemiology? Interdiscip Perspect Infect Dis 2011:676949

Murphy KJ, Ciuti S, Kane A (2020) An introduction to agent-based models as an accessible surrogate to field-based research and teaching. Ecol Evol 10:12482–12498

Vasylyeva TI, Friedman SR, Paraskevis D, Magiorkinis G (2016) Integrating molecular epidemiology and social network analysis to study infectious diseases: towards a socio-molecular era for public health. Infect Genet Evol 46:248–255

Schrödle B, Held L (2011) Spatio-temporal disease mapping using INLA. Environmetrics 22:725–734

Talisuna AO, Okiro EA, Yahaya AA, Stephen M, Bonkoungou B, Musa EO, Minkoulou EM, Okeibunor J, Impouma B, Djingarey HM, Yao NKM, Oka S, Yoti Z, Fall IS (2020) Spatial and temporal distribution of infectious disease epidemics, disasters and other potential public health emergencies in the World Health Organisation Africa region, 2016–2018. Global Health 16:9

Charron MV, Kluiters G, Langlais M, Seegers H, Baylis M, Ezanno P (2013) Seasonal and spatial heterogeneities in host and vector abundances impact the spatiotemporal spread of bluetongue. Vet Res 44:44

Brock PM, Fornace KM, Grigg MJ, Anstey NM, William T, Cox J, Drakeley CJ, Ferguson HM, Kao RR (2019) Predictive analysis across spatial scales links zoonotic malaria to deforestation. Proc Biol Sci 286:20182351

Cohen JM, Civitello DJ, Brace AJ, Feichtinger EM, Ortega CN, Richardson JC, Sauer EL, Liu X, Rohr JR (2016) Spatial scale modulates the strength of ecological processes driving disease distributions. Proc Natl Acad Sci USA 113:E3359–E3364

Diallo AB, Coiffard B, Leone M, Mezouar S, Mege J-L (2020) For whom the clock ticks: clinical chronobiology for infectious diseases. Front Immunol 11:1457

Kronfeld-Schor N, Bloch G, Schwartz WJ (2013) Animal clocks: when science meets nature. Proc Biol Sci 280:20131354

Lal A, Hales S, French N, Baker MG (2012) Seasonality in human zoonotic enteric diseases: a systematic review. PLoS One 7:e31883

Tricco AC, Lillie E, Zarin W, O’Brien KK, Colquhoun H, Levac D, Moher D, Peters MDJ, Horsley T, Weeks L, Hempel S, Akl EA, Chang C, McGowan J, Stewart L, Hartling L, Aldcroft A, Wilson MG, Garritty C, Lewin S, Godfrey CM, Macdonald MT, Langlois EV, Soares-Weiser K, Moriarty J, Clifford T, Tunçalp Ö, Straus SE (2018) PRISMA extension for scoping reviews (PRISMA-ScR): checklist and explanation. Ann Intern Med 169:467–473

Munn Z, Stern C, Aromataris E, Lockwood C, Jordan Z (2018) What kind of systematic review should I conduct? A proposed typology and guidance for systematic reviewers in the medical and health sciences. Med Res Methodol 18:5

Determine the key concepts of your topic. https://libguides.city.ac.uk/SHS-Litsearchguide/frameworks . Accessed 26 Feb 2019

R Core Team (2020) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria

Katale BZ, Fyumagwa RD, Mjingo EE, Sayalel K, Batamuzi EK, Matee MI, Keyyu JD, Muumba J, Mdaki M, Mbugi E, Rweyemamu MM, Mpanduji DG (2017) Screaning for bovine tuberculosis in african buffalo ( Syncerus caffer ) in Ngorongoro conservation area, Northern Tanzania: implications for public health. J Wildl Dis 53:711–717

Meunier NV, Sebulime P, White RG, Kock R (2017) Wildlife-livestock interactions and risk areas for cross-species spread of bovine tuberculosis. Onderstepoort J Vet Res 84:e1–e10

Wielgus E, Caron A, Bennitt E, De Garine-Wichatitsky M, Cain B, Fritz H, Miguel E, Cornélis D, Chamaillé-Jammes S (2021) Inter-group social behavior, contact patterns and risk for pathogen transmission in cape buffalo populations. J Wild Manag 85:1574–1590

Gormley AM, Anderson DP, Nugent G (2018) Cost-based optimization of the stopping threshold for local disease surveillance during progressive eradication of tuberculosis from New Zealand wildlife. Transbound Emerg Dis 65:186–196

Smith GC, Budgey R (2021) Simulating the next steps in badger control for bovine tuberculosis in England. PLoS One 16:e0248426

Birch CPD, Goddard A, Tearne O (2018) A new bovine tuberculosis model for England and Wales (BoTMEW) to simulate epidemiology, surveillance and control. Vet Res 14:273

Macdonald DW, Riordan P, Mathews F (2006) Biological hurdles to the control of TB in cattle: a test of two hypotheses concerning wildlife to explain the failure of control. Biol Conserv 131:268–286

Ellwanger JH, Fearnside PM, Ziliotto M, Valverde-Villegas JM, Veiga ABGD, Vieira GF, Bach E, Cardoso JC, Müller NFD, Lopes G, Caesar L, Kulmann-Leal B, Kaminski VL, Silveira ES, Spilki FR, Weber MN, Almeida SEM, Hora VPD, Chies JAB (2022) Synthesizing the connections between environmental disturbances and zoonotic spillover. An Acad Bras Ciênc 94(suppl 3):e20211530

Gibb R, Redding DW, Chin KQ, Donnelly CA, Blackburn TM, Newbold T, Jones KE (2020) Zoonotic host diversity increases in human-dominated ecosystems. Nature 584:398–402

Keesing F, Ostfeld RS (2021) Impacts of biodiversity and biodiversity loss on zoonotic diseases. Proc Natl Acad Sci USA 118:e2023540118

Bastida A, Tellez MH, Bustamante Montes LP, Jaramillo Paniagua JN, Jaimes Beni Tes MA, Mendoza Barrera GN, Rami Rez-Dura NN (2017) Spatial analysis of bovine tuberculosis in the State of Mexico, Mexico. Vet Ital 53:39–46

PubMed   Google Scholar  

Gaughran A, Mullen E, Macwhite T, Maher P, Kelly DJ, Kelly R, Good M, Marples NM (2021) Badger territoriality maintained despite disturbance of major road construction. PLoS One 16:e0242586

O’Hagan MJH, McCormick CM, Collins SF, McBride KR, Menzies FD (2021) Are major roads effective barriers for badger ( Meles meles ) movements? Res Vet Sci 138:49–52

Murphy KJ, Morera-Pujol V, Ryan E, Byrne AW, Breslin P, Ciuti S (2022) Habitat availability alters the relative risk of a bovine tuberculosis breakdown in the aftermath of a commercial forest clearfell disturbance. J Appl Ecol 59:2333–2345

Porphyre T, Stevenson MA, McKenzie J (2008) Risk factors for bovine tuberculosis in New Zealand cattle farms and their relationship with possum control strategies. Prev Vet Med 86:93–106

Brunton LA, Alexander N, Wint W, Ashton A, Broughan JM (2017) Using geographically weighted regression to explore the spatially heterogeneous spread of bovine tuberculosis in England and Wales. Stoch Environ Res Risk Assess 31:339–352

Allen AR, Ford T, Skuce RA (2021) Does Mycobacterium tuberculosis var. bovis survival in the environment confound bovine tuberculosis control and eradication? A literature review. Vet Med Int 2021:8812898

Brooks-Pollock E, Roberts GO, Keeling MJ (2014) A dynamic model of bovine tuberculosis spread and control in Great Britain. Nature 511:228–231

Redpath SHA, Marks NJ, Menzies FD, O’Hagan MJH, Wilson RP, Smith S, Magowan EA, Mcclune DW, Collins SF, Mccormick CM, Scantlebury DM (2023) Impact of test, vaccinate and remove protocol on home ranges and nightly movements of badgers in a medium density population. Sci Rep 13:2592

Gaughran A, Macwhite T, Mullen E, Maher P, Kelly DJ, Good M, Marples NM (2019) Dispersal patterns in a medium-density Irish badger population: implications for understanding the dynamics of tuberculosis transmission. Ecol Evol 9:13142–13152

Palisson A, Courcoul A, Durand B (2016) Role of cattle movements in bovine tuberculosis spread in France between 2005 and 2014. PLoS One 11:e0152578

Gortazar C, Diez-Delgado I, Barasona JA, Vicente J, De La Fuente J, Boadella M (2015) The wild side of disease control at the wildlife-livestock-human interface: a review. Front Vet Sci 1:27

Delahay CRJ, Smith GC, Barlow AM, Walker N, Harris A, Clifton-Hadley RS, Cheeseman CL (2007) Bovine tuberculosis infection in wild mammals in the South-West region of England: a survey of prevalence and a semi-quantitative assessment of the relative risks to cattle. Vet J 173:287–301

Crispell J, Cassidy S, Kenny K, Mcgrath G, Warde S, Cameron H, Rossi G, Macwhite T, White PCL, Lycett S, Kao RR, Moriarty J, Gordon SV (2020) Mycobacterium bovis genomics reveals transmission of infection between cattle and deer in Ireland. Microb Genom 6:mgen000388

Réveillaud É, Desvaux S, Boschiroli M-L, Hars J, Faure É, Fediaevsky A, Cavalerie L, Chevalier F, Jabert P, Poliak S, Tourette I, Hendrikx P, Richomme C (2018) Infection of wildlife by Mycobacterium bovis in France assessment through a National Surveillance System, Sylvatub. Front Vet Sci 5:262

Castillo JL, Fernández-Llario P, Mateos C, Carranza J, Benítez-Medina JM, García-Jiménez W, Bermejo-Martín F, Hermoso de Mendoza J (2011) Management practices and their association with Mycobacterium tuberculosis complex prevalence in red deer populations in Southwestern Spain. Prev Vet Med 98:58–63

Morera-Pujol V, Mostert PS, Murphy KJ, Burkitt T, Coad B, Mcmahon BJ, Nieuwenhuis M, Morelle K, Ward AI, Ciuti S (2023) Bayesian species distribution models integrate presence-only and presence–absence data to predict deer distribution and relative abundance. Ecography 2023:e06451

Gaidet N, Cappelle J, Takekawa JY, Prosser DJ, Iverson SA, Douglas DC, Perry WM, Mundkur T, Newman SH (2010) Potential spread of highly pathogenic avian influenza H5N1 by wildfowl: dispersal ranges and rates determined from large-scale satellite telemetry. J App Ecol 47:1147–1157

Raizman EA, Rasmussen HB, King LE, Ihwagi FW, Douglas-Hamilton I (2013) Feasibility study on the spatial and temporal movement of Samburu’s cattle and wildlife in Kenya using GPS radio-tracking, remote sensing and GIS. Prev Vet Med 111:76–80

Wearn O, Glover-Kapfer P (2017) Camera-trapping for conservation: a guide to best-practices. WWF, Woking

Vazquez C, Rowcliffe JM, Spoelstra K, Jansen PA (2019) Comparing diel activity patterns of wildlife across latitudes and seasons: time transformations using day length. Methods Ecol Evol 10:2057–2066

Wevers J, Beenaerts N, Casaer J, Zimmermann F, Artois T, Fattebert J (2021) Modelling species distribution from camera trap by-catch using a scale-optimized occupancy approach. Remote Sens Ecol Conserv 7:534–549

Henrich M, Hartig F, Dormann CF, Kühl HS, Peters W, Franke F, Peterka T, Šustr P, Heurich M (2022) Deer behavior affects density estimates with camera traps, but is outweighed by spatial variability. Front Ecol Evol 10:881502

Kukielka VE, Barasona JA, Cowie CE, Drewe JA, Gortazar C, Cotarelo I, Vicente J (2013) Spatial and temporal interactions between livestock and wildlife in South Central Spain assessed by camera traps. Prev Vet Med 112:213–221

Gaudiano L, Pucciarelli L, Mori E (2021) Livestock grazing affects movements and activity pattern of Italian roe deer in Southern Italy. Eur J Wildl Res 67:1–8

Triguero-Ocaña R, Vicente J, Palencia P, Laguna E, Acevedo P (2020) Quantifying wildlife-livestock interactions and their spatio-temporal patterns: is regular grid camera trapping a suitable approach? Ecol Indic 117:106565

Hinojo A, Christe P, Moreno I, Hofmeister RJ, Dandliker G, Zimmermann F (2022) Estimating roe deer density using motion-sensitive cameras in Switzerland. J Wildl Manag 86:e22307

Schilling AK, Mazzamuto MV, Romeo C (2022) A review of non-invasive sampling in wildlife disease and health research: what’s new? Animals 12:1719

Childress J, Faust CL, Deiner K (2023) Introduction to special issue: advancing disease ecology through eDNA monitoring of infectious agents. Environ DNA 6:e502

Sieber N, Hartikainen H, Vorburger C (2020) Validation of an eDNA-based method for the detection of wildlife pathogens in water. Dis Aquat Organ 141:171–184

Sengupta ME, Lynggaard C, Mukaratirwa S, Vennervald BJ, Stensgaard AS (2022) Environmental DNA in human and veterinary parasitology—current applications and future prospects for monitoring and control. Food Waterborne Parasitol 29:e00183

Cheng YC, Hannaoui S, John TR, Dudas S, Czub S, Gilch S (2016) Early and non-invasive detection of chronic wasting disease prions in elk feces by real-time quaking induced conversion. PLoS One 11:e0166187

King HC, Murphy A, James P, Travis E, Porter D, Sawyer J, Cork J, Delahay RJ, Gaze W, Courtenay O, Wellington EM (2015) Performance of a noninvasive test for detecting Mycobacterium bovis shedding in European Badger ( Meles meles ) populations. J Clin Microbiol 53:2316–2323

Pronyk PM, de Alwis R, Rockett R, Basile K, Boucher YF, Pang V, Sessions O, Getchell M, Golubchik T, Lam C, Lin R, Mak T-M, Marais B, Ong RT-H, Clapham HE, Wang L, Cahyorini Y, Polotan FGM, Rukminiati Y, Sim E, Suster C, Smith GJD, Sintchenko V (2023) Advancing pathogen genomics in resource-limited settings. Cell Genom 3:100443

Reis AC, Cunha MV (2021) Genome-wide estimation of recombination, mutation and positive selection enlightens diversification drivers of Mycobacterium bovis . Sci Rep 11:18789

Wilmers CC, Nickel B, Bryce CM, Smith JA, Wheat RE, Yovovich V (2015) The golden age of bio-logging: how animal-borne sensors are advancing the frontiers of ecology. Ecology 96:1741–1753

Williams HJ, Taylor LA, Benhamou S, Bijleveld AI, Clay TA, Grissac S, Demšar U, English HM, Franconi N, Gómez-Laich A, Griffiths RC, Kay WP, Morales JM, Potts JR, Rogerson KF, Rutz C, Spelt A, Trevail AM, Wilson RP, Börger L (2020) Optimizing the use of biologgers for movement ecology research. J Anim Ecol 89:186–206

Drewe JA, O’Connor HM, Weber N, McDonald RA, Delahay RJ (2013) Patterns of direct and indirect contact between cattle and badgers naturally infected with tuberculosis. Epidemiol Infect 141:1467–1475

Woodroffe R, Donnelly CA, Ham C, Jackson SYB, Moyes K, Chapman K, Stratton NG, Cartwright SJ (2016) Badgers prefer cattle pasture but avoid cattle: implications for bovine tuberculosis control. Ecol Lett 19:1201–1208

Dougherty ER, Seidel DP, Carlson CJ, Spiegel O, Getz WM (2018) Going through the motions: incorporating movement analyses into disease research. Ecol Lett 21:588–604

Wilson RP, Grundy E, Massy R, Soltis J, Tysse B, Holton M, Cai Y, Parrott A, Downey LA, Qasem L, Butt T (2014) Wild state secrets: ultra-sensitive measurement of micro-movement can reveal internal processes in animals. Front Ecol Environ 12:582–587

Vázquez Diosdado JA, Barker ZE, Hodges HR, Amory JR, Croft DP, Bell NJ, Codling EA (2015) Classification of behaviour in housed dairy cows using an accelerometer-based activity monitoring system. Anim Biotelemetry 3:15

Barasona JA, Latham MC, Acevedo P, Armenteros JA, Latham ADM, Gortazar C, Carro F, Soriguer RC, Vicente J (2014) Spatiotemporal interactions between wild boar and cattle: implications for cross-species disease transmission. Vet Res 45:122

Woodroffe R, Donnelly CA, Chapman K, Ham C, Moyes K, Stratton NG, Cartwright SJ (2021) Successive use of shared space by badgers and cattle: implications for Mycobacterium bovis transmission. J Zool 314:132–142

Bartolín-Arnau LM, Todoli-Ferrandis D, Sempere-Payá V, Silvestre-Blanes J, Santonja-Climent S (2023) LoRaWAN networks for smart applications in rural settings. IETE Tech Rev 40:440–452

O’Kennedy M, Niesler T, Wolhuter R, Mitton N (2020) Practical evaluation of carrier sensing for a LoRa wildlife monitoring network. In: 2020 IFIP Networking Conference (Networking), Paris, France, June 2020, pp. 614–618

Shaheen MNF (2022) The concept of one health applied to the problem of zoonotic diseases. Rev Med Virol 32:e2326

Soulsbury CD, White PCL (2015) Human-wildlife interactions in urban areas: a review of conflicts, benefits and opportunities. Wildl Res 42:541

Griffin LL, Haigh A, Amin B, Faull J, Norman A, Ciuti S (2022) Artificial selection in human-wildlife feeding interactions. J Anim Ecol 91:1892–1905

Carrete M, Tella JL (2017) Behavioral correlations associated with fear of humans differ between rural and urban burrowing owls. Front Ecol Evol 5:54

Spiegel O, Leu ST, Bull CM, Sih A (2017) What’s your move? Movement as a link between personality and spatial dynamics in animal populations. Ecol Lett 20:3–18

Gartland LA, Firth JA, Laskowski KL, Jeanson R, Ioannou CC (2022) Sociability as a personality trait in animals: methods, causes and consequences. Biol Rev 97:802–816

Oster M, Scheel M, Muráni E, Ponsuksili S, Zebunke M, Puppe B, Wimmers K (2015) The fight-or-flight response is associated with PBMC expression profiles related to immune defence and recovery in swine. PLoS One 10:e0120153

Lloyd-Smith JO, Schreiber SJ, Kopp PE, Getz WM (2005) Superspreading and the effect of individual variation on disease emergence. Nature 438:355–359

Barber I, Dingemanse NJ (2010) Parasitism and the evolutionary ecology of animal personality. Philos Trans R Soc Lond B Biol Sci 365:4077–4088

Santicchia F, Romeo C, Ferrari N, Matthysen E, Vanlauwe L, Wauters LA, Martinoli A (2019) The price of being bold? Relationship between personality and endoparasitic infection in a tree squirrel. Mamm Biol 97:1–8

Poulin R (2013) Parasite manipulation of host personality and behavioural syndromes. J Exp Biol 216:18–26

Ward AI, Judge J, Delahay RJ (2010) Farm husbandry and badger behaviour: Opportunities to manage badger to cattle transmission of Mycobacterium bovis ? Prev Vet Med 93:2–10

FAO, OIE, UN, WHO (2021) OHHLEP annual report

OHHLEP (2022) OHHLEP-One-Health theory of chance

Triguero-Ocaña R, Martínez-López B, Vicente J, Barasona JA, Martínez-Guijosa J, Acevedo P (2020) Dynamic network of interactions in the wildlife-livestock interface in mediterranean Spain: an epidemiological point of view. Pathogens 9:120

Wilber MQ, Pepin KM, Campa H, Hygnstrom SE, Lavelle MJ, Xifara T, Vercauteren KC, Webb CT (2019) Modelling multi-species and multi-mode contact networks: implications for persistence of bovine tuberculosis at the wildlife–livestock interface. J App Ecol 56:1471–1481

Fielding HR, Silk MJ, McKinley TJ, Delahay RJ, Wilson-Aggarwal JK, Gauvin L, Ozella L, Cattuto C, McDonald RA (2021) Spatial and temporal variation in proximity networks of commercial dairy cattle in Great Britain. Prev Vet Med 194:105443

Tratalos JA, Madden JM, Mcgrath G, Graham DA, Collins ÁB, More SJ (2020) Spatial and network characteristics of Irish cattle movements. Prev Vet Med 183:105095

Silk MJ, Croft DP, Delahay RJ, Hodgson DJ, Boots M, Weber N, Mcdonald RA (2017) Using social network measures in wildlife disease ecology, epidemiology, and management. Biosci 67:245–257

Farine DR, Whitehead H (2015) Constructing, conducting and interpreting animal social network analysis. J Anim Ecol 84:1144–1163

Kaur P, Ciuti S, Ossi F, Cagnacci F, Morellet N, Loison A, Atmeh K, McLoughlin P, Reinking AK, Beck JL, Ortega AC, Kauffman M, Boyce MC, Salter-Townshend M (2023) Assessing bias and robustness of social network metrics using GPS based radio-telemetry data. bioRxiv. 2023.2003.2030.534779

Sosa S, Sueur C, Puga-Gonzalez I (2021) Network measures in animal social network analysis: their strengths, limits, interpretations and uses. Methods Ecol Evol 12:10–21

Allen T, Murray KA, Zambrana-Torrelio C, Morse SS, Rondinini C, Di Marco M, Breit N, Olival KJ, Daszak P (2017) Global hotspots and correlates of emerging zoonotic diseases. Nat Commun 8:1124

Newton-Cross G, White PCL, Harris S (2007) Modelling the distribution of badgers Meles meles : comparing predictions from field-based and remotely derived habitat data. Mamm Rev 37:54–70

Elith J, Leathwick JR (2009) Species distribution models: ecological explanation and prediction across space and time. Annu Rev Ecol Evol Syst 40:677–697

Lindgren F, Rue H (2015) Bayesian spatial modelling with R-INLA. J Stat Softw 63:1–25

Elith J (2019) Machine learning, random forests, and boosted regression trees. In: Brennan L, Tri A, Marcot B (eds) Quantitative analyses in wildlife science. Johns Hopkins University Press, Baltimore, pp 281–297

Isaac NJB, Jarzyna MA, Keil P, Dambly LI, Boersch-Supan PH, Browning E, Freeman SN, Golding N, Guillera-Arroita G, Henrys PA, Jarvis S, Lahoz-Monfort J, Pagel J, Pescott OL, Schmucki R, Simmonds EG, O’Hara RB (2020) Data integration for large-scale models of species distributions. Trends Ecol Evol 35:56–67

Bakka H, Rue H, Fuglstad GA, Riebler A, Bolin D, Illian J, Krainski E, Simpson D, Lindgren F (2018) Spatial modeling with R-INLA: a review. WIREs Comp Stats 10:e1443

Tracey JA, Bevins SN, Vandewoude S, Crooks KR (2014) An agent-based movement model to assess the impact of landscape fragmentation on disease transmission. Ecosphere 5:art119

Gaughran A, Kelly DJ, Macwhite T, Mullen E, Maher P, Good M, Marples NM (2018) Super-ranging: a new ranging strategy in European badgers. PLoS One 13:e0191818

O’Hagan MJH, Gordon AW, Mccormick CM, Collins SF, Trimble NA, Mcgeown CF, Mchugh GE, Mcbride KR, Menzies FD (2021) Effect of selective removal of badgers ( Meles meles ) on ranging behaviour during a ‘Test and Vaccinate or Remove’ intervention in Northern Ireland. Epidemiol Infect 149:1–31

Byrne AW, Acevedo P, Green S, O’Keeffe J (2014) Estimating badger social-group abundance in the Republic of Ireland using cross-validated species distribution modelling. Ecol Indic 43:94–102

Abdou M, Frankena K, O’Keeffe J, Byrne AW (2016) Effect of culling and vaccination on bovine tuberculosis infection in a European badger ( Meles meles ) population by spatial simulation modelling. Prev Vet Med 125:19–30

Barroso P, Breslin P, McGrath G, Jamie MM, Tratalos JA, More SJ, Ryan E, Byrne AW, Barrett D (2022) Is there an association between road building and bovine tuberculosis herd risk? A three time-point study in Ireland, 2011–2019. Prev Vet Med 198:105542

Jesko Z, Reamonn MF, Kevin L, Eva MM, Phillip OB, Ian P, Gavin S, Stuart G (2016) The Irish Land-Parcels Identification System (LPIS)—experiences in ongoing and recent environmental research and land cover mapping. Biol Environ 116B:53–62

Martin SW, O’Keeffe J, Byrne AW, Rosen LE, White PW, McGrath G (2020) Is moving from targeted culling to BCG-vaccination of badgers ( Meles meles ) associated with an unacceptable increased incidence of cattle herd tuberculosis in the Republic of Ireland? A practical non-inferiority wildlife intervention study in the Republic of Ireland (2011–2017). Prev Vet Med 179:105004

Ryan E, Breslin P, O’Keeffe J, Byrne AW, Wrigley K, Barrett D (2023) The Irish bTB eradication programme: combining stakeholder engagement and research-driven policy to tackle bovine tuberculosis. Ir Vet J 76:32

Bovine tuberculosis scoping review coded variables dataset. https://doi.org/10.5281/zenodo.10695407

Phylogenetic icons database. Accessed 23 Oct 2023. https://www.phylopic.org/

Creative commons attribution licenses. Accessed 23 Oct 2023. https://creativecommons.org/share-your-work/cclicenses/

Download references

Acknowledgements

We would like to acknowledge Sylvia Power for helping in reviewing the final version of the manuscript.

This publication has emanated from research conducted with the financial support of Science Foundation Ireland under Grant number 18/CRT/6049. For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. In addition, HME is funded by an Irish Research Council Government of Ireland postgraduate scholarship. KJM and VMP is funded by the Department of Agriculture, Food and the Marine (DAFM) in Ireland through the research grant 2019-R-417.

Author information

Authors and affiliations.

Laboratory of Wildlife Ecology and Behaviour, School of Biology and Environmental Science, University College Dublin, Dublin, Ireland

Kimberly Conteddu, Holly M. English, Bawan Amin, Laura L. Griffin, Virginia Morera-Pujol, Kilian J. Murphy & Simone Ciuti

Department of Agriculture, Food and the Marine, One Health Scientific Support Unit, Dublin, Ireland

Andrew W. Byrne

School of Mathematics and Statistics, University College Dublin, Dublin, Ireland

Prabhleen Kaur & Michael Salter-Townshend

Department of Wildlife Ecology and Management, Faculty of Environment and Natural Resources, University of Freiburg, Freiburg, Germany

Adam F. Smith

The Frankfurt Zoological Society, Frankfurt, Germany

Department of National Park Monitoring and Animal Management, Bavarian Forest National Park, Grafenau, Germany

You can also search for this author in PubMed   Google Scholar

Contributions

KC: conceived and designed the review, acquired, analysed and interpreted the data, wrote the first draft of the manuscript. HME, LLG: edited and revised the manuscript and contributed to the interpretation of the data. AWB: edited and revised the manuscript and contributed to the interpretation of the data, drafted the policy and research implications. BA, PK, VMP, KJM, AFS, MST: edited specific sections of the discussion and revised the whole manuscript. SC: conceived and designed the review, edited the manuscript and supervised the process. All Authors read and approved the final manuscript.

Corresponding author

Correspondence to Kimberly Conteddu .

Ethics declarations

Competing interests.

The authors declare that they have no competing interests.

Additional information

Handling editor: Pauline Ezanno.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1..

Search terms and list of articles . The additional file contains the search terms used in Web of Science to identify potentially relevant articles as well as the list of all the relevant articles found after the duplicate screening process.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Conteddu, K., English, H.M., Byrne, A.W. et al. A scoping review on bovine tuberculosis highlights the need for novel data streams and analytical approaches to curb zoonotic diseases. Vet Res 55 , 64 (2024). https://doi.org/10.1186/s13567-024-01314-w

Download citation

Received : 09 May 2023

Accepted : 20 March 2024

Published : 21 May 2024

DOI : https://doi.org/10.1186/s13567-024-01314-w

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• bovine tuberculosis
• Mycobacterium bovis
• infectious disease management
• mathematical modelling
• multi-host disease
• wildlife host

Veterinary Research

ISSN: 1297-9716

• Submission enquiries: [email protected]
• General enquiries: [email protected]