The effects of repeated expressions on attitude polarization during group discussions.

Classic explanations of the "group polarization phenomenon" emphasize interpersonal processes such as informational influence and social comparison (Myers & Lamm, 1976). Based on earlier research, we hypothesized that at least part of the polarization observed during group discussion might be due to repeated attitude expression. Two studies provide support for this hypothesis. In Study 1, we manipulated how often each group member talked about an issue and how often he or she heard other group members talk about the issue. We found that repeated expression produced a reliable shift in extremity. A detailed coding of the groups' discussions showed that the effect of repeated expression on attitude polarization was enhanced in groups where the group members repeated each other's arguments and used them in their own line of reasoning. Study 2 tested for this effect experimentally. The results showed that the effect of repeated expression was augmented in groups where subjects were instructed to use each others' arguments compared to groups where instructions were given to avoid such repetitions. Taken together, these studies show that repeated expression accounts for at least part of the attitude polarization observed in the typical studies on group polarization and that this effect is augmented by social interaction, i.e., it occurs particularly in an environment where group members repeat and validate each other's ideas.