In one study participants were divided into two groups using the minimal group paradigm. Participants then filled out an attitude questionnaire twice, once to record their own attitudes and once to record how they thought another ingroup or outgroup member might respond. The results showed that participants tended to
infer ingroup members’ beliefs more accurately than outgroup members’ beliefs.
assume that their beliefs were more similar to those of fellow ingroup members than to those of outgroup members.
overestimate variability in outgroup attitudes but underestimate variability in ingroup attitudes.
assume that ingroup members held more socially desirable beliefs than outgroup members.