General Impressions
Political Symbols and Social Order: Confederate Monuments and Performative Violence in the Post-Reconstruction U.S. South
Lee-Or Ankori-Karlinsky
American Political Science Review, forthcoming
Abstract:
Violent conflicts are often accompanied by symbols commemorating past violence. I argue that political symbols exert a causal effect on future violence. Such symbols generate shared understandings of the prevailing social order. Symbols that affirm this order may act as substitutes for performative violence motivated by status concerns, while their removal may signal contestation, increasing violence. I test this theory by examining the effect of Confederate monument construction on lynchings and public executions in the postbellum U.S. South. Using a difference-in-differences design and original archival work, I find that Confederate monuments reduced violence, acting as a substitute for performative violence in constructing a white supremacist social order. Effects are concentrated in counties where racial threat is higher. I then test the effects of Confederate monument removals in the present-day US and find that removals increased the likelihood of anti-Black hate crimes.
Racial Cues from Unfamiliar Sources and Their Effects on Americans’ Policy Preferences
Viviana Rivera-Burgos
Journal of Experimental Political Science, forthcoming
Abstract:
Americans increasingly confront policy messages not from high-profile political figures but from everyday citizens. Much is known about the effects of racial source cues from well-known political figures with salient racial identities. Less is known about how subtle racial cues from non-recognizable sources affect Americans’ support for policies that are race-targeted and those that are not. In this paper, I conduct a randomized experiment that varies a cue of the source’s racial identity and the type of policy for which the source advocates. I uncover little evidence for the hypothesis that subtle racial source cues activate racial attitudes that lead Americans to racialize policies that are (at least explicitly) race-neutral. I find instead that subtle cues of a Black vs. White source decrease support only for race-targeted policies. I reason that two mechanisms possibly driving this effect are: (1) subtle racial source cues become salient for only race-targeted policies, thereby activating racial stereotypes for these policies but not others, and (2) Black sources are perceived as less objective policy messengers when the policy explicitly aims to rectify injustices against Black Americans. More generally, the paper’s overall findings suggest that subtle racial cues of who advocates for race-targeted policies matter for whether such policies can garner the public support they presumably need to come to fruition.
Assessing Customer Discrimination in National Television Viewership: Evidence From NCAA Basketball
Byungju Kang & Steven Salaga
American Behavioral Scientist, forthcoming
Abstract:
How do consumers respond to a product based on the racial composition of the individuals representing that product? This is a challenging question to empirically assess, given the difficulty in acquiring data capturing actual consumer behavior while simultaneously controlling for product quality and characteristics. We empirically test for potential customer discrimination in live national television viewership of National Collegiate Athletic Association men’s college basketball games. After holding contest quality and characteristics constant, we uncover statistically significant evidence of increased viewership when the number of minutes played by non-White players on the combined team rosters is larger. In models assessing individual race, we find a statistically significant increase in viewership when the number of minutes played by Black players is greater. The results oppose previous empirical work assessing customer discrimination in market-level television viewership in professional sport.
Segregation, Spillovers, and the Locus of Racial Change
Donald Davis, Matthew Easton & Stephan Thies
NBER Working Paper, September 2025
Abstract:
We use a discrete choice framework to provide the first nesting of Thomas C. Schelling’s canonical models of racial segregation amenable to empirical examination. Using U.S. Census data from 1970–2000, we demonstrate a central role for spatial racial spillovers in shaping racial clustering, patterns of racial shares and housing prices at the boundary of racial clusters, and the locus of racial change. Our results on the locus of racial change conflict strongly with prominent prior results on racial tipping. Our theory provides a foundation for spatially stratified regressions. The strongest spatial effects in the prior work are not tipping, but the distinct biased White suburbanization. Tipping effects in urban areas remote from Minority clusters are small or insignificant. In urban areas proximate to Minority clusters they average less than half those reported in prior pooled results. Policies promoting racial integration must thus attend to the heterogeneous fragility of neighborhoods.
Age and gender distortion in online media and large language models
Douglas Guilbeault, Solène Delecourt & Bhargav Srinivasa Desikan
Nature, forthcoming
Abstract:
Are widespread stereotypes accurate or socially distorted? This continuing debate is limited by the lack of large-scale multimodal data on stereotypical associations and the inability to compare these to ground truth indicators. Here we overcame these challenges in the analysis of age-related gender bias, for which age provides an objective anchor for evaluating stereotype accuracy. Despite there being no systematic age differences between women and men in the workforce according to the US Census, we found that women are represented as younger than men across occupations and social roles in nearly 1.4 million images and videos from Google, Wikipedia, IMDb, Flickr and YouTube, as well as in nine language models trained on billions of words from the internet. This age gap is the starkest for content depicting occupations with higher status and earnings. We demonstrate how mainstream algorithms amplify this bias. A nationally representative pre-registered experiment (n = 459) found that Googling images of occupations amplifies age-related gender bias in participants’ beliefs and hiring preferences. Furthermore, when generating and evaluating resumes, ChatGPT assumes that women are younger and less experienced, rating older male applicants as of higher quality. Our study shows how gender and age are jointly distorted throughout the internet and its mediating algorithms, thereby revealing critical challenges and opportunities in the fight against inequality.
Sex Differences in Children's Motivation and Action Patterns for Climbing as Behavioral Relicts of Ancestral Sexual-Size Dimorphism
Richard Coss, Victor Geisler & Michael Newmann
Evolutionary Psychology, July 2025
Abstract:
Four studies investigated sex differences in children's motivation and action patterns for climbing playground structures and a gymnasium rock wall to assess any influence of ancestral sexual-size dimorphism limiting tree-climbing agility. Study 1 examined yearly incidences of children aged 3 to 13 falling from monkey bars and jungle gyms in a 1985–1989 National Electronic Injury Surveillance System dataset. Injury incidences of 3- to 6-year-old girls were lower than those of same-aged boys with the inverse occurring between ages 7 through 10 (p < 0.001). Study 2 determined that, during two recess periods in 13 elementary schools, 3.14% of enrolled girls were climbing playground structures compared with 1.45% of enrolled boys (p = 0.021) who were less inclined to climb as they aged. Study 3 showed that 6 to 8 year-old girls climbing alone perched longer (p = 0.0004) on 3 jungle gyms in a regional park longer than same-aged boys. Extended perching by girls might reflect their greater desire for surveillance useful historically for assessing danger. For Study 4, video recordings were made of the climbing actions of 28 children 7- to 12- years of age enrolled in an indoor rock-wall climbing class for beginners. Girls exhibited marked climbing differences (p = 0.005), with discriminant function analysis classifying 84.6% of girls correctly and 86.7% of boys correctly. While tree climbing was not studied directly, the sex differences shown in these studies indicates that girls are motivated to climb playground structures more than boys and climb rock walls using different action patterns.
Nice to meet you.(!) Gendered norms in punctuation usage
Yidan Yin, Gil Appel & Cheryl Jan Wakslak
Journal of Experimental Social Psychology, November 2025
Abstract:
People face a myriad of daily decisions about how to communicate, especially in today's digital world. We consider the decision to use exclamation points as a window into how men and women navigate the mundane choices that guide so much of their day to day communication. Across five studies, our findings suggest that exclamation point usage is associated more with women than with men, that these normative expectations are impactful, and that women -- who are more sensitive to potential downstream impression formation implications of using exclamation points -- think about this issue more than men and are more uncertain of their exclamation point usage. We further find that the decision to use exclamations does indeed shape social perception, leading to more positive impressions overall but also some negative concerns; however, we do not find evidence that these effects are moderated by communicator gender. Our findings provide insight into how men and women engage in everyday communication in the face of normative expectations related to gender and shed light on the unexpected burdens that this can create.
When and Why Do Sex Differences in Handgrip Strength Emerge? Age-Varying Effects of Testosterone From Childhood to Older Adulthood
Jun Seob Song, Heontae Kim & Myungjin Jung
American Journal of Human Biology, October 2025
Methods: Time-varying effect modeling (TVEM) was used to examine age-specific trajectories of handgrip strength and testosterone, and to assess whether these trajectories differed by sex. A moderated mediation analysis was conducted to test whether the sex difference in handgrip strength was mediated by testosterone level, and whether this effect varied across age. Data were drawn from the 2011–2012 and 2013–2014 cycles (N = 11,035) of the National Health and Nutrition Examination Survey.
Results: TVEM indicated that the sex difference in handgrip strength emerged at age 8, whereas the sex difference in testosterone level became evident at age 10. A moderated mediation analysis revealed that testosterone mediated the association between sex and handgrip strength, and this effect decreased with age (IMM = −0.18, 95% CI: −0.20, −0.16).
When a Man Says He Is Pregnant: Event-related Potential Evidence for a Rational Account of Speaker-contextualized Language Comprehension
Hanlin Wu & Zhenguang Cai
Journal of Cognitive Neuroscience, forthcoming
Abstract:
Spoken language is often, if not always, understood in a context formed by the identity of the speaker. For example, we can easily make sense of an utterance such as “I’m going to have a manicure this weekend” or “The first time I got pregnant I had a hard time” when spoken by a woman, but it would be harder to understand when it is spoken by a man. Previous ERP studies have shown mixed results regarding the neurophysiological responses to such speaker–content mismatches, with some reporting an N400 effect and others a P600 effect. In an EEG experiment involving 64 participants, we used social and biological mismatches as test cases to demonstrate how these distinct ERP patterns reflect different aspects of rational inference. We showed that when the mismatch involves social stereotypes (e.g., men getting a manicure), listeners can arrive at a “literal” interpretation by integrating the content with their social knowledge, though this integration requires additional effort due to stereotype violations -- resulting in an N400 effect. In contrast, when the mismatch involves biological knowledge (e.g., men getting pregnant), a “literal” interpretation becomes highly implausible or impossible, leading listeners to treat the input as potentially containing errors and engage in correction processes -- resulting in a P600 effect. Supporting this rational inference framework, we found that the social N400 effect decreased as a function of the listener’s personality trait of openness (as more open-minded individuals maintain more flexible social expectations), while the biological P600 effect remained robust (as biological constraints are recognized regardless of individual personalities). Our findings help to reconcile empirical inconsistencies and reveal how rational inference shapes speaker-contextualized language comprehension. We demonstrate how listeners flexibly adapt its processing strategy based on contextual information, which may be part of the general information-processing principles of the human brain.