In testing possible cultural effects of the use of the self as an habitual
reference point to which others are compared, we expected that: (a) individ
ualistic participants (i.e., those who give priority to personal goals) wou
ld rate self-other similarity higher when asked "How similar is X to you?''
than when asked "How similar are you to X?", whereas nondirectional simila
rity judgements ("How similar are these two people?'') would resemble the f
ormer directional comparison; (b) collectivistic participants (i.e., those
who give priority to in-group goals) would show a weaker or, possibly, reve
rsed pattern, especially using in-group comparison others. Neither hypothes
is was upheld. However, the individualists perceived the in-group to be rel
atively more similar to themselves as compared to the collectivists. This d
ifference cannot be explained by response bias, status asymmetry, or role d
ifferentiation. We propose an explanation in terms of the differential rela
tionship between self and other representations for people from collectivis
t versus individualist cultures.