Neuroscience Journals Lack Consistent Rules for Reporting Race and Ethnicity, Review Finds

A sweeping review of the world’s leading neuroscience journals has revealed a striking inconsistency in how researchers are asked to report one of the most ethically charged variables in human research: race and ethnicity. The analysis, published in Frontiers in Neuroimaging, examined author guidelines across 27 high-impact neuroscience journals and found that nearly a quarter offer no guidance at all on reporting race and ethnicity population descriptors, while the remainder draw on at least six different, often conflicting, sets of recommendations. The findings arrive at a pivotal moment for the field, as population neuroscience increasingly relies on massive datasets where demographic variables can shape every downstream conclusion.

The research team, led by Lauren Lang of Emory University and Julianna Gerold of the Massachusetts General Hospital Athinoula A. Martinos Center for Biomedical Imaging, together with colleagues at Harvard, Brown, and the University of Minnesota, set out to answer two questions. The first was empirical: what guidance do neuroscience journals currently provide, and how much does it vary? The second was normative: should that guidance be standardized, and if so, what form should the standards take? Their answer to the first question is unambiguous. Their answer to the second is more nuanced, and it challenges the field to think carefully about the difference between how variables are reported and what those variables actually mean.

To conduct the survey, the researchers examined the 25 highest impact factor journals in the Neurosciences category of the Observatory of International Research rankings as of May 2024. They excluded two journals, Molecular Neurodegeneration and Sleep Medicine Reviews, for insufficient human neuroimaging content or insufficient website information, and added four others where influential human neuroimaging work is frequently published: Human Brain Mapping, Science, Nature, and Oxford Open Neurosciences. Between June 28 and July 10, 2024, the team reviewed each journal’s website for author guidance on population descriptors and emailed senior editors to ask about their journals’ standards, policies, or informal norms for reporting demographic data. Nineteen of the 25 journals contacted responded.

The results paint a picture of a field without a common playbook. Twenty-one of the 27 journals, or 78 percent, provide authors with at least some guidance on reporting race and ethnicity, but four of those limit their guidelines exclusively to clinical trials rather than all human studies. That leaves 22 percent of journals offering no guidance whatsoever. Among the journals that do provide direction, six different guideline frameworks are in circulation, with the International Committee of Medical Journal Editors recommendations, Nature’s guidance, and the American Medical Association Manual of Style the most commonly adopted. Most guidelines simply require that reported categories be defined and that any decision to omit race or ethnicity be explained, without specifying how authors should define or operationalize the categories themselves. Only one journal, Human Brain Mapping, states an explicit preference, favoring ethnicity over race.

A smaller group of journals has adopted more comprehensive frameworks. Seven of the 27, or 26 percent, follow either Nature’s guidelines or the AMA Manual of Style. Nature’s guidance calls for reporting race and ethnicity or justifying any omission, using inclusive and respectful language, considering genetic ancestry, and warning that manuscripts may be revised or rejected for racist or harmful content. The AMA’s 2024 update goes further in one crucial respect: it states explicitly that race, ethnicity, and geographic origin are not valid proxies for genetic ancestry, and it requires authors to use inclusive and specific language, attend to privacy and ethics, apply nuanced categorization, and acknowledge limits on generalizability. Yet even these detailed frameworks stop short of telling researchers how to define the categories they report.

Perhaps the most revealing finding concerns the editors themselves. When the researchers corresponded with one senior editor whose journal publishes author guidelines on its website, the editor replied that the journal had, in the editor’s words, no specific policies on the description of race and ethnicity demographic variables. If the people running journals are not always aware of their own published policies, the authors reasoned, author compliance is likely to suffer. To test that suspicion, the team conducted a pilot analysis of Biological Psychiatry, one of only two journals in their sample that had adopted the AMA guidelines, which require reporting of race and ethnicity or an explanation for their absence. Reviewing all human subjects research articles published in the journal between February 1 and August 15, 2024, they found that only 30 of 61 articles, or 49 percent, reported participants’ race, ethnicity, or ancestry. Just a single article complied with all AMA requirements, and it did so largely because its homogenous sample rendered many of the reporting guidelines inapplicable.

The authors speculate that low adherence reflects both the relative newness of these policies and inconsistent enforcement. Many working investigators were trained at a time when instruction on reporting race and ethnicity was not a priority in scientific writing. But the deeper issue the review raises is conceptual. Race is a sociopolitical construct that does not map onto genetic ancestry, yet it is frequently used as a proxy for genetic ancestry or for exposure to social determinants of health. When race or ethnicity stands in for social determinants, researchers cannot precisely identify the mechanisms underlying observed effects, and attributing differential health outcomes to race places the burden of disease on individuals while obscuring systemic causes. The stakes of getting this wrong are not abstract. In the nineteenth and early twentieth centuries, practices such as cranial measurement were deployed to rank intelligence across racialized groups, reinforcing biologically deterministic hierarchies. Although those claims have been discredited, their methodological legacy persists whenever socially constructed categories are treated as proxies for biological difference in contemporary neuroimaging analyses.

At the same time, the authors are careful not to argue that race and ethnicity should be abandoned as research variables. The 2022 National Academies report on improving representation in clinical trials and research documents how recognizing the distinct medical needs of various groups has contributed to reductions in health disparities. Mindful use of population descriptors can promote representativeness, increase generalizability, and keep findings interpretable for the populations to which they apply. The challenge, then, is to design standards that neither reify biological race nor erase the ethical relevance of these variables. The authors draw a key distinction between procedural standards, which require researchers to report their decisions about whether and how to include race, and substantive standards, which would require every study to define race the same way. Given the recognized variation in how individuals define race and ethnicity, they conclude that a universal substantive standard is not advisable. Instead, they endorse the view articulated by Cardenas-Iniguez and Gonzalez in 2024 that no categorical definition of race and ethnicity is exhaustive and without limitations, and that researchers should be intentional, equitable, and responsible in their chosen operationalization.

Models for such an approach already exist in adjacent fields and large-scale consortia. The Adolescent Brain Cognitive Development Study cautions against misusing population descriptors in genetic analyses and emphasizes responsible interpretation as a condition of data access, noting that self-reported race and ethnicity primarily reflect social and environmental experiences rather than genetic variation. The UK Biobank employs standardized ethnic background categories with defined subcategories while publishing clear statements about selection biases, including healthy volunteer bias and socioeconomic skew, giving researchers context for interpretation rather than implying global representativeness. In genetics and epidemiology, professional societies such as the International Genetic Epidemiology Society and programs like the National Heart, Lung, and Blood Institute’s Trans-Omics for Precision Medicine initiative have issued recommendations emphasizing careful terminology, explicit justification for population groupings, avoidance of race or ethnicity as proxies for genetic ancestry, and consideration of potential harms alongside analytic benefits.

The authors close with three concrete recommendations. First, journals should require transparency about whether and how race, ethnicity, and ancestry are used in every human neuroscience study, giving researchers flexibility while demanding justification. Second, journal editors should collaborate to develop field-wide procedural standards, drawing on existing frameworks from Nature, the AMA, and ICMJE, and the team is pursuing this goal through a new project called Improving Recruitment, Engagement, and Access for Community Health Equity for BRAIN Next-Generation Human Neuroimaging Research, or REACH for BRAIN, which is convening an expert working group with community input to fill the guidance gap. Third, the field should learn from the Sex and Gender Equity in Research guidelines, which took three years to develop, were published in 2016, and required years of sustained follow-up, including adoption by the World Health Organization, before achieving widespread implementation. The lesson is that writing guidelines is only the first step. At present, most human neuroscience studies are silent about why race or ethnicity were or were not included and about how those variables were defined. Changing that, the authors argue, would improve interpretability, enable cross-cohort harmonization, and force researchers to confront a scientific history in which purported racial differences in brain structure were used to perpetuate racism, ensuring that modern analytic frameworks do not quietly reintroduce outdated assumptions.

Subject of Research: Ethical and methodological challenges in reporting race and ethnicity population descriptors in human neuroscience research

Article Title: Addressing ethical challenges in the reporting of race and ethnicity population descriptors in human neuroscience research

Article References: Lang, L., Gerold, J., Little, E., McFarland, C., Tawe, B., Jain, R. K., & Shen, F. X. (2026). Addressing ethical challenges in the reporting of race and ethnicity population descriptors in human neuroscience research. Frontiers in Neuroimaging, 5, Article 1744569. https://doi.org/10.3389/fnimg.2026.1744569

Image Credits: AI Generated

DOI: 10.3389/fnimg.2026.1744569

Keywords: neuroscience, neuroimaging, race and ethnicity, population descriptors, journal guidelines, research ethics, health equity, biobanks, reproducibility, genetic ancestry, scientific publishing, population neuroscience