About

Yarrow Dunham is a Professor of Psychology at Yale University, holding an Ed.D. from Harvard University obtained in 2007. His research focuses on intergroup social cognition, exploring how humans, as perhaps the most social species, tend to divide individuals into social groups based on a wide array of properties such as shared beliefs, origins, traits, and affiliations. Dunham's work investigates how these social groupings influence perception and behavior, including biases towards ingroups and the immediate preference for unfamiliar groups assigned randomly. His lab addresses fundamental questions about the origins of these psychological tendencies by studying how knowledge of social groups is acquired in both cognitively mature adults and children. Dunham's research examines how people decide which properties to use as the basis for social grouping, whether there are natural tendencies to partition the social world in certain ways, and the role of cultural input in shaping the intergroup mind. Employing experimental and cross-cultural methodologies, his work aims to understand the development and universality of social categorization processes.

Research topics

• Computer Science
• Psychology
• Social psychology
• Sociology
• Political Science
• Data Mining
• Mathematics
• Epistemology
• Developmental psychology
• Philosophy
• Cognitive science
• Econometrics
• Neuroscience
• Statistics

Selected publications

• A children’s case for reparations: U.S. Children’s and adults’ ideas about collective responsibility.

  Journal of Experimental Psychology General · 2025-11-20

  articleSenior author

  Are current members of a social group responsible for making amends for the harm caused by their predecessors? We explored U.S. children's and adults' views on whether responsibility is limited to the original perpetrators or extends to the group's present members. Across four studies (342 children [6-12 years; 50% girls, 49% boys, 1% gender unknown; 52% White, 21% Asian, 7% Hispanic, 5% Black, 9% other, 6% race unknown] and 665 adults [47% women, 52% men, 1% gender unknown; 74% White, 10% Black, 6% Asian, 5% Hispanic, 4% other, 1% race unknown]), participants evaluated vignettes where an advantaged group could compensate a disadvantaged group that they had previously harmed or that was disadvantaged for independent reasons. Younger children believed wealthy groups should always compensate poorer ones, while older children and adults judged compensation more necessary when harm was caused by the advantaged group (Experiment 1). Older children also believed compensation and apology should be offered even when the disadvantaged group had recovered economically (Experiment 2) or when the current members of the advantageous group no longer benefit from the harm (Experiment 3). Adults, while ambivalent about monetary compensation in these cases, agreed on the need for apology. Experiment 4 replicated Experiment 2 with more realistic stimuli, confirming that adults' overall high collective responsibility ascriptions were not caused by the child-friendly materials used in Experiments 1-3. Although limited by a U.S. convenience sample, these findings suggest that children, like adults, attribute collective responsibility for harms committed by a group's predecessors. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

  PublisherDOI

• The sensitivity of implicit attitude formation to logical operations: Evidence from the bipolar–unipolar distinction

  2025-02-21

  preprintOpen access

  Emerging single-process propositional perspectives in psychology and philosophy have introduced the key idea that, much like their explicit (controlled) counterparts, implicit (automatic) attitudes should reflect logical operations such as negation. In the present project, we subject this idea to a particularly stringent test by probing not only whether implicit attitude formation is sensitive to negation but also whether such sensitivity additionally reflects the distinction between easy-to-negate bipolar adjectives (those with clear antonyms, e.g., “strong”) and difficult-to-negate unipolar adjectives (those without clear antonyms, e.g., “unique”). Across four experiments (three preregistered; total N = 6,707), we found (a) no significant difference in implicit attitude formation in response to affirmed bipolar vs. affirmed unipolar adjectives (β = 0.07); (b) modulation of implicit attitude formation by whether adjectives were affirmed or negated (β = 0.67); and, critically, (c) stronger modulation of implicit attitude formation by the negation of bipolar relative to unipolar adjectives (β = 0.23). These results generalized across three sets of adjectives, two implicit attitude measures (the Implicit Association Test and the Affect Misattribution Procedure), and two novel targets (social and nonsocial). Together, these data provide remarkably strong evidence to suggest that implicit attitude formation reflects logical operations, including not only whether an adjective is affirmed or negated but even whether adjectives are relatively easy or difficult to negate. We discuss theoretical implications for single-process and dual-process accounts of attitude formation and higher-order human cognition, as well as translational implications for theoretically informed attempts at modifying preexisting implicit social attitudes.

  PublisherDOI

• Children’s understanding of abstract and real-world social groups

  2025-05-14

  preprintOpen accessSenior author

  The present work explores how two intuitive theories inform young children’s inferences about social groups. In three studies (N = 821), we tested whether 3- to 7-year-old children view novel (Studies 1 and 3), gender (Studies 2 and 3), and racial (Studies 2 and 3) groups as (a) marking individuals who are fundamentally similar to one another, and (b) marking patterns of social relationships and interactions. We found evidence for both of these sets of beliefs. Children predicted that ingroup members would be more similar to one another than outgroup members for all of the groups tested. Children also predicted that novel group members would be friends with one another, would be nice to one another, and would avoid harming one another, and predictions regarding gender and racial groups increasingly followed patterns similar to predictions about novel groups across development. We also found preliminary evidence that individual differences among children inform the ways in which children develop their expectations of intergroup interaction. These findings suggest that children combine their abstract knowledge with their understanding of groups in the real world to navigate the complex social world that they inhabit.

  PublisherDOI

• Taking advantage: Predictions and moral judgments of leveraging outside options in ultimatum games

  Journal of Experimental Social Psychology · 2025-11-17

  articleSenior author
  PublisherDOI

• Logical Operations Shape the Formation of Implicit Attitudes: Evidence From the Negation of Bipolar and Unipolar Adjectives

  Open Mind · 2025-01-01

  articleOpen access

  Abstract Emerging single-process propositional perspectives in psychology and philosophy have introduced the key idea that, much like their explicit (deliberately retrieved) counterparts, implicit (automatically retrieved) attitudes should be sensitive to logical operations such as negation. In the present project, we subject this idea to a particularly stringent test by probing not only whether the formation of implicit attitudes is sensitive to negation but also whether such sensitivity additionally reflects the distinction between easy-to-negate bipolar adjectives (those with clear antonyms, e.g., strong) and difficult-to-negate unipolar adjectives (those without clear antonyms, e.g., unique). Across four experiments (three preregistered; total N = 6,707), we found (a) no significant difference in the formation of implicit attitudes in response to affirmed bipolar versus affirmed unipolar adjectives (β = 0.07); (b) modulation of the formation of implicit attitudes by whether adjectives were affirmed or negated (β = 0.67); and, critically, (c) stronger modulation of the formation of implicit attitudes by the negation of bipolar relative to unipolar adjectives (β = 0.23). These results generalized across three sets of adjectives, two implicit attitude measures (the Implicit Association Test and the Affect Misattribution Procedure), and two novel targets (social and nonsocial). Together, these data provide strong evidence to suggest that the formation of implicit attitudes is sensitive to logical operations, including not only whether an adjective is affirmed or negated but even whether adjectives are relatively easy or difficult to negate. We discuss theoretical implications for single-process and dual-process accounts of attitude formation and higher-order human cognition.

  PublisherOA PDFDOI

• The Development of Morality and Conventionality Across Cultures: Implementing a Two‐Stage Model for Cross‐Cultural Research

  Developmental Science · 2025-12-15

  article

  Establishing a shared sense of right and wrong is an essential milestone for human cooperation, raising the question of whether a universal set of moral intuitions exists. However, tests of universality in the domain of human morality are hindered by the overrepresentation of participants from Western, educated, industrialized, rich, and democratic (WEIRD) societies and issues of validity that arise from the use of WEIRD measures (i.e., measures originating in WEIRD societies and primarily normed on WEIRD samples) to make cross-cultural comparisons. Here we address the tension between cross-cultural generalizability and validity by deploying a two-stage approach to investigate the moral beliefs of 5- to 10-year-olds from four diverse societies (N = 331). Specifically, we test a classic case study in which strong universality claims have previously been made: the "moral/conventional" distinction. In Study 1, we test for the distinction cross-culturally using standardized measures widely used in moral cognition research and find robust evidence of the distinction in Canadian children, but a more variable pattern among Korean, Indian, and Iranian children, with Iranian children showing the weakest evidence for a distinction. In Study 2, we focus specifically on Iran and tailor experimental stimuli to reflect culture-specific norms in that country. We find that Iranian children residing under a theocracy also exhibit the moral/conventional distinction-so long as moral and conventional codes do not intersect with religious or legal concerns. These findings support the use of a two-stage model in which cultural comparisons are made using both shared and culturally specific measures. RESEARCH HIGHLIGHTS: Two studies examined whether children in four diverse societies make the distinction between violations of moral codes and social conventions. When standardized methods were used, evidence for the moral/conventional distinction was robust in Western children, variable in non-Western children, and weakest in Iranian children. Strong evidence for the moral/conventional distinction was found in Iranian children when culturally tailored measures were used. Our findings support the use of a two-stage model that combines the strengths of standardized and tailored measures for conducting cross-cultural research.

  PublisherDOI

• Children’s understanding of abstract and real-world social groups.

  Developmental Psychology · 2025-07-10 · 2 citations

  articleSenior author

  = 821), we tested whether 3- to 7-year-old children view novel (Studies 1 and 3), gender (Studies 2 and 3), and racial (Studies 2 and 3) groups as (a) marking individuals who are fundamentally similar to one another and (b) marking patterns of social relationships and interactions. We found evidence for both of these sets of beliefs. Children predicted that ingroup members would be more similar to one another than outgroup members for all of the groups tested. Children also predicted that novel group members would be friends with one another, would be nice to one another, and would avoid harming one another, and predictions regarding gender and racial groups increasingly followed patterns similar to predictions about novel groups across development. We also found preliminary evidence that individual differences among children inform the ways in which children develop their expectations of intergroup interaction. These findings suggest that children combine their abstract knowledge with their understanding of groups in the real-world to navigate the complex social world that they inhabit. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

  PublisherDOI

• Children’s understanding of abstract and real-world social groups

  2025-05-13

  preprintOpen accessSenior author

  The present work explores how two intuitive theories inform young children’s inferences about social groups. In three studies (N = 821), we tested whether 3- to 7-year-old children view novel (Studies 1 and 3), gender (Studies 2 and 3), and racial (Studies 2 and 3) groups as (a) marking individuals who are fundamentally similar to one another, and (b) marking patterns of social relationships and interactions. We found evidence for both of these sets of beliefs. Children predicted that ingroup members would be more similar to one another than outgroup members for all of the groups tested. Children also predicted that novel group members would be friends with one another, would be nice to one another, and would avoid harming one another, and predictions regarding gender and racial groups increasingly followed patterns similar to predictions about novel groups across development. We also found preliminary evidence that individual differences among children inform the ways in which children develop their expectations of intergroup interaction. These findings suggest that children combine their abstract knowledge with their understanding of groups in the real world to navigate the complex social world that they inhabit.

  PublisherOA PDFDOI

• Who Deserves to Be Rich? Children’s and Adults’ Judgments of the Wealthy

  Collabra Psychology · 2025-01-01 · 1 citations

  articleOpen accessSenior author

  While past work has suggested that children have highly positive views of the wealthy, little research has explored whether these beliefs are affected by the source of the wealth. Here we explore US-based 120 5–12-year-olds’ (49% female, 51% male; 57% White, 16% Asian, 8% Black, 5% Multiracial, 4% Latine, 11% did not report race) and 154 adults’ (58% female; 39% male, 3% other; 79% White, 7% Latine, 7% Multiracial, 3% Black, 3% East Asian, 2% other) views of individuals who acquired their wealth via merit, theft, luck, and inheritance. We find substantial variation in views of deservedness, warmth, and competence of these individuals, as well as differences in how older children and younger children judge wealthy targets. These age differences are particularly pronounced with respect to luck and inheritance. These results offer insights into evolving beliefs about wealth deservedness, especially regarding less-explored aspects like inheritance, which is important given its relevance to the intergenerational transmission of inequality.

  PublisherOA PDFDOI

• Roles guide rapid inferences about agent knowledge and behavior

  2025-03-26

  preprintOpen access

  The ability to predict and understand other people’s actions is critical for real-world social behavior. Here we hypothesized that representations of social roles (e.g., cashier, mechanic, doctor) enable people to build rapid expectations about what others know and how they might act. Using a self-paced read- ing paradigm, we show that role representations support real time expectations about how other people might act (Study 1) and the knowledge they might possess (Study 2). Moreover, people reported more surprisal when the events deviated from role expectations, and they were more likely to misremember what happened. Our results suggest that roles are a powerful route for social understanding that has been previously under- studied in social cognition.

  PublisherOA PDFDOI

Recent grants

• RR: Collaborative: Origins of Intergroup Perceptions and Attitudes Across Diverse Contexts

  NSF · $92k · 2017–2020

Frequent coauthors

• Benedek Kurdi

  29 shared

• Mahzarin R. Banaji

  Harvard University

  24 shared

• Mahesh Srinivasan

  University of California, Berkeley

  23 shared

• David Barner

  21 shared

• Nadia Chernyak

  20 shared

• Andrew Scott Baron

  Ocean Networks Canada Society

  19 shared

• David G. Rand

  Massachusetts Institute of Technology

  19 shared

• Alexander Noyes

  19 shared