Until recently the stereotype researchers have usually measured knowledge o
f stereotypes but not the endorsement of these stereotypes. Moreover, most
investigations have focused on either encoding or retrieval effects of ster
eotypical information. The aim of the present study was to provide evidence
for both effects in one and the same memory test in the context of sexist
prejudices. It was hypothesized that highly sexist persons are, on the one
hand, better at remembering inconsistent information (inconsistency effect)
, but on the other hand, because of their sexist beliefs, they also provide
stereotype-guided answers (consistency effect). Participants had either to
form an impression of, or to learn items about a target person who was eit
her male or female and was described by gender stereotype congruent and inc
ongruent attributes. Later, they were asked to recollect the information in
a recognition test and participants' modern sexism was assessed. The resul
ts confirm the hypotheses. Implications of these findings for the study of
person memory and social stereotypes are discussed.