On the effectiveness of group brainstorming: Test of One Underlying Cognitive Mechanism

Recent studies suggest that group brainstorming, in its electronic form, can be an effective method of generating ideas, if the group is sufficiently large. We report here an experiment probing the mechanism underlying this good performance. Because larger groups are more likely than small ones to generate rare ideas, we hypothesized that rare ideas might be especially stimulating to further idea generation, and thus boost the performance of large, interacting groups. Experimental subjects working alone generated ideas while exposed to streams of either rare, common, or no stimulus ideas. No support for the hypothesized stimulating effect of rare ideas was found. We suggest several alternative mechanisms worth exploring.