Abstract
Seeking to support the assumption of self-perception theory that attitudes existing prior to manipulation are not salient to the phenomenology of Ss in an experiment on cognitive dissonance, Bern and McConnell report an experiment in which Ss could not accurately recall their original attitude ratings after engaging in behaviour counter to that attitude. Chris and Woodyard in a variation of the Bern and McConnell experiment found that Ss, for whom the original attitude was important, did not make more errors in recall than control Ss. The present study, using a different measure of the importance of the original attitude, produced results consistent with those obtained by Chris and Woodyard. These results challenge Bern's assumption that attitudes existing prior to manipulation are not salient in cognitive dissonance phenomenology.
Get full access to this article
View all access options for this article.
