Stimuli
1. Full-Body Photo Database
This database contains 726 full-body target images of Asian-, Black-, and White-presenting individuals (54 Asian female, 63 Asian Male, 115 Black female, 154 Black male, 140 White female, 200 White male). Photos were gathered from volunteer members of the public and online between 2016 and 2019. All volunteers were asked to stand facing the camera with a neutral expression, and online images were selected to meet the same criteria. Adherence to these criteria, and photo resolution, varies between target images.
Please note: 272 of the targets were created in Adobe Photoshop by swapping different heads onto different bodies (and adjusting skin tone, etc, to make the resulting images as credible as possible). All targets with double names have been created this way. E.g., WM77-BM51.png is the head of White Male 77 on the body of Black Male 51 (the head is always the first ID listed).
In addition to the images, this database contains 490,359 explicit ratings of the 726 targets made by 3,311 US adults (1,875 women, 1031 men, 24 non-binary, 381 missing gender data, Mage = 23.8, SDage = 8.6, 1,116 Asian, 1,089 White, 414 Latino, 117 Black, 575 other race or unreported) on 24 different personality and demographic traits. Traits measured were: warmth, competence, honest/moral , dominant, submissive, hard-working, extraverted/enthusiastic, reserved/quiet, sympathetic/warm, critical/quarrelsome, dependable/self-disciplined, disorganized/careless, calm/emotionally stable, anxious/easily upset, open to new experiences/complex, conventional/uncreative, attractive, income, education, occupational prestige, subjective socioeconomic status, age, political orientation, and race (measured via a multiple choice categorical response.
You can download these images and ratings data here. I don’t have a publication specifically about this stimulus set, but if you use it, please cite the following paper, which was the first publication to use these images:
Connor, P., Varney, J., Keltner, D., & Chen, S. (2020). Social class competence stereotypes are amplified by socially-signalled economic inequality. Personality and Social Psychology Bulletin, 47(1): 89-105. https://doi.org/10.1177/0146167220916640
2. Face-Body Swapped Photo Database
This database contains 144 target images of Asian-, Black-, and White-presenting individuals (24 of each race/gender subgroup) created by swapping 24 faces obtained from the Chicago Face Database onto 24 different bodies obtained from the full-body photo database described above.
This stimulus set was deliberately created to simultaneously manipulate target race, gender, age, and social class as close to orthogonally as possible. Faces were matched on perceived attractiveness and racial prototypicality between race, age, and gender groups, on female/male categorization between race and age groups , on racial categorization between gender and age groups, and on perceived age between race and gender groups. Bodies were matched on perceived attractiveness between race, age, and gender groups, on perceived age between gender and SES groups, and on perceived SES and income between gender and age groups.
I used Adobe Photoshop software to attach each of the faces to 6 different bodies. This resulted in 144 total stimuli, which were then assembled into six target groups, in which no face or body appeared more than once. Each group contained 8 Asian, 8 Black, and 8 White targets, 12 female and 12 male targets, 12 young and 12 old targets, and 12 high-SES and 12 low-SES targets. See this publication for more details. Targets are also available in full-body or upper-body format, as shown below:
In addition to the images, this database contains 465,584 explicit ratings of the 144 targets made by 3,110 US adults (1,871 women, 1156 men, 38 non-binary, 45 missing gender data, Mage = 30.7, SDage = 13.9, 1,560 White, 905 Asian, 264 Latino, 188 Black, 193 other race or unreported) on 10 different traits. Traits measured were: gender, race (Asian, Black, or White), age, political liberalism, warmth, competence, subjective SES, attractiveness, extroversion, and photo blurriness (all traits were measured via 0-100 sliders).
You can download these images and ratings data here. I don’t have a publication specifically about this stimulus set, but if you use it, please cite the following paper, which was the first publication to use these images:
Connor, P., Weeks, M., Glaser, J., Chen, S., & Keltner, D. (2022). Intersectional implicit bias: Evidence for asymmetrically compounding bias and the predominance of target gender. Journal of personality and social psychology, 10.1037/pspa0000314. Advance online publication. https://doi.org/10.1037/pspa0000314