The present study investigates the effect of priming on the use of soc
ial categorizations. Using sex and race as stimulus categorizations, p
revious studies failed to confirm the hypothesis that priming one of t
he two available categorizations would enhance the use of the primed c
ategorization relation to the not-primed categorization. As suggested
by Stangor, Lynch, Duan and Glass (1992), a momentary increase in acce
ssibility may be insufficient to further enhance the use of highly acc
essible categorizations like race and sex, but it may be expected that
priming will increase the use of categorizations which are less habit
ually used in daily life. The results of the present experiment suppor
t the hypothesis that, when the stimulus categorizations are weakly ac
cessible (university major, university town), the relative use of the
previously primed categorizations does indeed increase, compared to th
e alternative, not-primed, categorization.