Brainstorming, when people gather to generate ideas together, is great in theory: many perspectives mesh to generate diverse outputs. In practice, evidence shows that brainstorming groups often perform more poorly than an equivalent number of soloists (often called a 'nominal' group). Some reasons are social, such as a pressure not to offer wild ideas in public; these can be mitigated by changing norms or tweaking process, e.g. sharing ideas anonymously using computers. A recent article focuses on the other side of the equation: the mental or cognitive narrowing that happens when you hear others' ideas.
Nicholas Kohn and Steven Smith ran a series of studies with undergraduate students, who spent twenty minutes on a computer responding to the challenge "List ways in which to improve Texas A&M University." Half the participants were in brainstorming groups, accessing the ideas of three other group members in a chat window, whereas the others worked independently with their outputs combined after the fact to make nominal groups. The first experiment affirmed that nominal groups did better – they accessed more categories of idea, and had more ideas overall.
Kohn and Smith suspected something called cognitive fixation, where being exposed to another's idea makes it more salient in your mind and blocks ideas of other types. They examined this in experiment two, where each participant was grouped with a single partner who was actually a confederate of the experimenters. This allowed them to systematically manipulate the number of ideas a participant saw in their chat window, presenting between one and twenty typical ideas from the most common categories generated in experiment one, such as Transportation or Food.
As expected, a high number of cues led to less novel ideas within fewer categories, which were rarely the uncued, uncommon ones. However, the overall number of ideas was not significantly affected, meaning candidates went more deeply into those fewer categories that they did consider. This suggests fixation: inspired by – but stuck on – the concepts presented to them.
A final experiment suggested that fixation can be shaken by taking a break. Participants who had been fed typical cues during just the first half of the study generated 86% more ideas and explored 57% more categories in the second half if they were put to work on an unrelated five-minute task in between. The break had no effect when participants were not exposed to fixation cues in the first half.
Although brainstorming didn't outperform a nominal group, the study suggests instances where it might be preferred: "if the goal is to explore a few categories in depth, then interaction among the members should be encouraged", preferably with a break and time to work more independently. Conversely, when you are after variety and uniqueness of ideas, cognitive fixation on obvious topics may be a risk. One solutions is to elicit opportunities for solo free thinking, and have these outputs brought to the table instead; another might be to use techniques to guide thinking towards the fringes rather than gravitating back to our common concerns.
Kohn, N., & Smith, S. (2011). Collaborative fixation: Effects of others' ideas on brainstorming Applied Cognitive Psychology, 25 (3), 359-371 DOI: 10.1002/acp.1699