Undergraduate students, 27 white and 27 black, provided saliva samples for
testosterone assay, completed measures of prejudice, and viewed pictures of
black and white target persons. Pupil dilation was measured continuously w
hile subjects viewed the pictures and thought about meeting the persons who
m they were seeing. Testosterone interacted with prejudice, such that preju
diced individuals who were low in testosterone showed large pupil dilation
and prejudiced individuals who were high in testosterone showed small pupil
dilation. The interaction of testosterone with prejudice in affecting dila
tion was not related to the race either of the subject, or of the target be
ing viewed.