The authors predicted that individuals high in dispositional anxiousness would perform poorly when brainstorming in groups but not during solitary brainstorming. Experiment 1 demonstrated this result in a comparison of groups of 4 that were all high or all low in interaction anxiousness. In groups with 2 low-anxious and 2 high-anxious individuals, the low-anxious individuals lowered their performance in the direction of the high-anxious individuals. These results suggest that part of the productivity loss observed in interactive brainstorming groups may be due to the inhibited performance of individuals who are uncomfortable with group interaction. Moreover, these individuals may influence others in the group to lower their performance in line with that inhibited performance level. Experiment 2 demonstrated that the poor performance of socially anxious groups in interactive brainstorming is not dependent on whether group members have individual microphones or share 1 common microphone.