Jennifer Eberhardt

Introduction

Jennifer L. Eberhardt is an American social psychologist who studies how race is coded, categorised and perceived, and what that does to decisions in the criminal justice system. Her work established that racial imagery is not merely a matter of what people believe but of what they see: exposure to Black faces changes how people interpret ambiguous objects, how they look at crime-related images, and how severely they judge defendants. In a line of studies now taught in every social psychology textbook, she and her collaborators showed that the more stereotypically Black a defendant's face is rated, the more likely that defendant is to receive a death sentence — an effect that survives statistical control for the severity of the crime.

Eberhardt is a professor at Stanford University, where she holds appointments in psychology, organizational behavior and public policy, and directs SPARQ, a centre that pairs researchers with practitioners. She received a MacArthur Fellowship in 2014, was elected to the National Academy of Sciences in 2016, and in 2021 became the first African American president of the Association for Psychological Science, an organisation with more than 25,000 members.

Early Life and Education

Eberhardt grew up in Cleveland, Ohio, the daughter of Harlan Eberhardt, a postal mail worker, and Mary Eberhardt. She was the first person in her family to attend college and has described watching how sharply educational opportunity varied between the neighbourhoods she lived in as the experience that oriented her toward inequality research; her mother was inspired to enrol in college herself in midlife after her daughter completed a doctorate.

She earned a B.A. from the University of Cincinnati in 1987, followed by an A.M. in 1990 and a Ph.D. in 1993 from Harvard University. Her training placed her at the intersection of the two traditions her career would combine: the experimental study of automatic social cognition, and the empirical study of race and law.

Academic and Professional Career

From 1995 to 1998 Eberhardt held a joint appointment at Yale University in the Departments of Psychology and of African and African American Studies, where she was also a research fellow at the Center for Race, Inequality, and Politics. She joined the Stanford faculty in 1998 and has been there since.

At Stanford she is the William R. Kimball Professor and a professor of organizational behavior at the Graduate School of Business, a professor of psychology in the School of Humanities and Sciences, the Morris M. Doyle Centennial Professor of Public Policy, and a professor of law by courtesy. She is faculty director of SPARQ — Social Psychological Answers to Real-World Questions — a "do tank" that brings behavioural scientists together with police departments, courts, schools, technology companies and nonprofits to test interventions against bias and disparity.

Her honours include the 2014 MacArthur Fellowship, which carried a $625,000 stipend and which she described as giving her "the space to pursue my research with new energy"; designation as one of Foreign Policy's 100 Leading Global Thinkers; election to the American Academy of Arts and Sciences and the National Academy of Sciences in 2016; and election to the British Academy in 2022. Her 2019 book Biased won book awards from the American Psychological Association and the Association for Psychological Science, was runner-up for a non-fiction Literary Peace Prize, and made her the first social scientist to receive the Rockefeller University Lewis Thomas Prize, an award established to honour the scientist as poet.

Beyond the university she has advised and spoken at the White House, the U.S. Department of Justice, the California Department of Justice, the Supreme Court of California, and numerous corporations and foundations. Her TED talk "How racial bias works — and how to disrupt it" has been viewed more than 3.3 million times.

Major Contributions to Psychology

Visual processing and the race-crime association

Eberhardt's early experimental work asked whether the association between race and crime changes perception rather than only judgment. Participants exposed subliminally to Black faces were subsequently more likely to identify a degraded image of a hand tool as a gun, and to misclassify crime-relevant objects more generally. In related work, participants shown Black faces allocated more visual attention to crime-related images, and police officers were more likely to misidentify Black faces as criminal and to judge the most stereotypically Black faces as the most criminal. The finding mattered because it located the effect upstream of decision-making: what people saw, not only what they chose, was being shaped by race.

Facial stereotypicality and sentencing

Her best-known archival study examined capital cases and found that defendants whose faces were rated as more stereotypically Black were more likely to be sentenced to death, and that the effect was concentrated in cases where the victim was White. Because the analysis controlled for legally relevant case characteristics, the result offered the strongest available field evidence that phenotypic bias enters sentencing directly.

The fragility of childhood

Extending the work to juveniles, Eberhardt and colleagues showed that merely bringing to mind a single Black juvenile offender led participants — across the political spectrum and at all levels of self-reported prejudice — to perceive juveniles in general as more adult-like, more blameworthy and more deserving of life without parole. The result reframed protections for young defendants as fragile rather than fixed, and it informed the line of research on the dehumanization of Black children that her collaborators carried forward.

Disparity and public support for punishment

A second programme, largely with Rebecca Hetey, asked how the public responds to information about racial disparity. Exposure to the scale of racial disparity in incarceration increased support for punitive policies among White participants, while making the same prison population appear more racially balanced reduced it. The counter-intuitive result established a paradox at the centre of reform communication: documenting the problem can entrench it.

From diagnosis to intervention

Eberhardt has increasingly moved from documenting bias to changing institutions. Working with the Oakland Police Department and others, she has analysed stop data, examined how organisational practices shape discretionary encounters, and tested whether narrative film interventions can increase empathic understanding of formerly incarcerated people and support for criminal justice reform. Her more recent published work with computer scientists analyses the language of police-civilian traffic-stop encounters and the racial and ethnic representation of literature taught in U.S. high schools.

Major Publications

  • Eberhardt, J. L., Goff, P. A., Purdie, V. J., & Davies, P. G. (2004). Seeing Black: Race, crime, and visual processing. Journal of Personality and Social Psychology, 87, 876–893.
  • Eberhardt, J. L., Davies, P. G., Purdie-Vaughns, V. J., & Johnson, S. L. (2006). Looking deathworthy: Perceived stereotypicality of Black defendants predicts capital-sentencing outcomes. Psychological Science, 17, 383–386.
  • Rattan, A., Levine, C. S., Dweck, C. S., & Eberhardt, J. L. (2012). Race and the fragility of the legal distinction between juveniles and adults. Psychological Science, 23, 724–731.
  • Hetey, R. C., & Eberhardt, J. L. (2014). Racial disparities in incarceration increase acceptance of punitive policies. Proceedings of the National Academy of Sciences, 111, 7803–7808.
  • Hetey, R. C., & Eberhardt, J. L. (2018). The numbers don't speak for themselves: Racial disparities and the persistence of racial inequality. Proceedings of the National Academy of Sciences, 115, 7934–7939.
  • Eberhardt, J. L. (2019). Biased: Uncovering the Hidden Prejudice That Shapes What We See, Think, and Do. Viking.

Impact and Legacy

Eberhardt moved the study of implicit bias out of the laboratory and into the institutions where it does damage. Police departments, prosecutors' offices, courts, school districts and large employers now run bias programmes, collect stop data and revise discretionary practices largely because her research made the cost of automatic associations measurable in real cases. Her election as the first Black president of the Association for Psychological Science, and her membership in three national academies, mark a shift in who the discipline recognises as central to it.

Her work also sits directly inside the most consequential methodological dispute in modern social psychology, and she has been criticised from within that dispute.

Reliability of implicit measures. The Implicit Association Test and related measures have test-retest correlations around r = .5, well below the standard for individual assessment, and implicit measures of race and gender are markedly less stable over time than implicit measures of political attitudes. A person classified as moderately biased in one session may be classified as neutral in the next. Eberhardt's most cited experimental findings rest on measures of this kind, and while she typically reports group-level differences — where the measures are reliable — the instability constrains what can be said about any individual officer or juror.

Predictive validity. Whether implicit measures predict behaviour is contested. Greenwald and colleagues' 2009 meta-analysis reported an average correlation of about r = .24 for intergroup behaviour, with implicit measures outperforming explicit ones in socially sensitive domains. Oswald and colleagues' 2013 reanalysis, using different inclusion criteria and updated techniques, put the figure at r = .14 and concluded that the IAT "provides little insight into who will discriminate against whom, and provides no more insight than explicit measures of bias." Carlsson and Agerström, restricting the set to genuinely discriminatory behaviours, reported r = .15. The dispute is unresolved and the estimates are small either way.

Does changing implicit bias change anything? Forscher and colleagues' 2019 meta-analysis of 492 studies covering 87,418 participants found that procedures can shift implicit scores, but that effects on behaviour were trivial by conventional standards and that there was no evidence that changes in implicit measures mediate changes in behaviour. Reviews of implicit bias training programmes, which many police departments adopted on the strength of this literature, have concluded they are largely ineffective, with effects that decay within days, that concentrate among the least biased participants, and that can provoke defensive responding or signal to majority-group members that they are being policed. Anthony Greenwald, one of the IAT's creators, has said publicly that implicit bias training has no demonstrated effect on behaviour. A 2023 review by Elizabeth Paluck and colleagues concluded that evidence for implicit prejudice-reduction techniques remains too thin to justify widespread adoption.

Archival inference. The sentencing findings come from observational data in a single jurisdiction's capital cases. Statistical control for legally relevant case factors is not randomisation, and the possibility that facial stereotypicality correlates with unmeasured case characteristics cannot be excluded. The result is strong for what it is — a controlled archival association — and weaker than an experiment would be.

Her own position responds to much of this. SPARQ's agenda has shifted deliberately from changing individuals' automatic associations to changing the structures within which discretion is exercised: data collection, encounter protocols, decision checklists and organisational practice. This partly concedes the critique — structural remedies do not depend on the reliability of an individual-level implicit score — and it is the reason her most recent work analyses organisational practices and language rather than reaction times. The tension the field is left with is that the diagnostic evidence for implicit bias is weaker at the individual level than the popular version of the idea assumes, while the disparities that motivated the research are not in doubt.

Learn More

References

  • Eberhardt, J. L., Goff, P. A., Purdie, V. J., & Davies, P. G. (2004). Seeing black: Race, crime, and visual processing. Journal of Personality and Social Psychology, 87(6), 876-893.

Last updated: 2026-10-20

Menu
Loading