TY - JOUR
T1 - Investigation of gender bias in the mental imagery of faces
AU - Kunst, Jonas R.
AU - Juettemeier, Marilena
AU - Bailey, April H.
AU - Anjum, Gulnaz
AU - English, Alexander S.
AU - Obaidi, Milan
AU - Sam, David L.
AU - Yaşın-Tekizoğlu, Fatma
AU - Agyemang, Collins B.
N1 - MJ, JRK, and AHB designed Study 1. All authors collected the cross-cultural data in Study 1. JRK collected the machine-learning results and human ratings, and analyzed the data in Study 1. JRK and AHB designed Study 2. JRK collected the data and conducted the analyses in Study 2. JRK, AHB, and MJ drafted the manuscript. JRK and AHB revised the manuscript. All authors provided critical revisions.
PY - 2024/9
Y1 - 2024/9
N2 - People tend to think of the prototypical person as a man more than as a woman, but this bias has primarily been observed in language-based tasks. Here, we investigated whether this bias is also present in the mental imagery of faces. A preregistered cross-cultural reverse-correlation study including participants from six WEIRD and non-WEIRD countries varying in gender equality (i.e., China, Ghana, Norway, Pakistan, Turkey, and the US; N = 645) unexpectedly suggested that people imagine the face of a generic “person” more as a woman than as a man. Replicating this unexpected result, a second preregistered study (N = 115) showed that U.S. participants imagine the face of a typical person as being more similar to their imagined face of a woman than of a man. We discuss explanations for these unexpected findings, including the possibility that the prototypical person is male-biased—consistent with previous work—but the default face may be female-biased.
AB - People tend to think of the prototypical person as a man more than as a woman, but this bias has primarily been observed in language-based tasks. Here, we investigated whether this bias is also present in the mental imagery of faces. A preregistered cross-cultural reverse-correlation study including participants from six WEIRD and non-WEIRD countries varying in gender equality (i.e., China, Ghana, Norway, Pakistan, Turkey, and the US; N = 645) unexpectedly suggested that people imagine the face of a generic “person” more as a woman than as a man. Replicating this unexpected result, a second preregistered study (N = 115) showed that U.S. participants imagine the face of a typical person as being more similar to their imagined face of a woman than of a man. We discuss explanations for these unexpected findings, including the possibility that the prototypical person is male-biased—consistent with previous work—but the default face may be female-biased.
KW - androcentrism
KW - cross-cultural
KW - faces
KW - gender bias
KW - reverse-correlation
UR - https://osf.io/s6nbq/?view_only=ee6d93e17cd04456a825647a3b4ead3a
UR - https://osf.io/s3ge2/?view_only=30023a1c96994a94b1f47390982f024b
UR - https://osf.io/kx8yz/?view_only=b454b180400f4b3d85d5a2fde4cb37f7
U2 - 10.1177/13684302231200168
DO - 10.1177/13684302231200168
M3 - Article
AN - SCOPUS:85175313226
SN - 1368-4302
VL - 27
SP - 1376
EP - 1402
JO - Group Processes and Intergroup Relations
JF - Group Processes and Intergroup Relations
IS - 6
ER -