The effect of the cognitive and social variables on devil's advocacy and brainstorming process

This experiment investigated the effects of devil's advocacy, social comparision and idea exemplars on the brainstorming performance. This experiment includes two parts. In the first experiment investigated the effects of devil's advocacy and idea exemplars on the brainstorming performance. The participants in the devil's advocacy condition were instructed to come up with the alternative or new ideas to their partner's ideas during brainstorming session. On the other hand, those in the no role condition were given the instruction that they would generate ideas by considering the ideas of his or her partners. The number of participants with devil's advocacy role within a group was manipulated by providing the group members either no role, one, or two written devil's advocacy roles. The participants were also given some exemplars of either unique or feasible ideas prior to the brainstorming session. Then all they were brainstormed on the university problem for 15 minute session in a brainwriting paradigm (Paulus and Yang, 2000). A 3 X 2 ANOVA analysis showed that the groups with 2 devil advocacy role or members generated more unique ideas and flexible ideas than those having the one and no devil advocacy member. The groups with one devil's advocate in a group also generated more ideas than those without any devil's advocate. However, providing unique or feasible ideas prior to the brainstorming session did not make a significant difference on the performance of groups. Second experiment, investigated the effects of devil's advocacy and social comparison targets on the brainstorming performance. The number of participants with devil's advocacy role within a group was manipulated by providing either all the group members devil's advocacy role or those no role. The participants were also instructed to be compared with the performance of either a computer, similar student, or either classmates. Those in the no comparison condition or target were given only information that their performance would be compared. Then all they were brainstormed on the extra thumb problem in a 15 minute session in a brainwriting paradigm (Paulus and Yang, 2000). A 2 X 4 ANOVA analysis showed that the groups with 3 devil advocacy role or members generated more ideas than those with no devil advocacy member. The participants compared with classmates also generated more ideas than those compared with a computer and those with no comparison target. No interaction effect was obtained. The results will be discussed in lights of the relevant literature.