The most replicated finding in creativity research that organisations keep ignoring

Alex Osborn, the advertising executive who popularised brainstorming in 1953, claimed that the average person can think up twice as many ideas when working with a group than when working alone. A Yale study conducted just five years later, in 1958, tested this claim directly. Twelve groups of four men and 48 individuals worked on the same three problems using Osborn’s four brainstorming rules. The groups produced fewer ideas than the individuals, and the individuals’ ideas were judged more unique.

The finding was subsequently ignored for decades by practitioners while researchers kept replicating it. Mullen, Johnson and Salas’s 1991 meta-analysis of 38 controlled experiments found that individuals outperformed brainstorming groups by 83% on average. Diehl and Stroebe’s analysis of 23 experiments found the same result consistently. The performance deficit of face-to-face brainstorming groups relative to the same number of people working alone is one of the most robustly replicated findings in creativity research — not a tentative observation awaiting confirmation, but a settled empirical fact that most organisations invest significant time and money ignoring.

Understanding why it happens is the first step toward designing something better.

Production blocking: the mechanical constraint no rule can fix

Diehl and Stroebe’s 1987 foundational four-experiment study identified three independent causes of brainstorming group productivity loss. The primary cause is production blocking — the simple mechanical fact that in a group, only one person can speak at a time. While others listen, they are not generating ideas. And the ideas they had queued to contribute are either forgotten or discarded as less relevant by the time they can speak.

The arithmetic is stark. In a six-person brainstorming group, each participant spends up to 80% of the time listening rather than generating. The delay between idea formation and vocalisation causes ideas to be suppressed or lost. This happens regardless of the group’s quality, the facilitator’s skill, or how closely the rules are followed — because the rules do not change the physics of sequential speech. Only one voice at a time means only one idea-generation process at a time, while five people wait.

Evaluation apprehension: why the most original ideas are the first to be suppressed

The second mechanism is evaluation apprehension — the fear of negative evaluation that prevents people from voicing their most unusual, risky, or genuinely novel ideas. In a face-to-face brainstorming group, every idea is immediately and publicly attributed to its generator, and every other group member forms an instant judgment about both the idea and the person who offered it.

Original ideas carry the highest social risk in this format precisely because they are most likely to be unusual, most likely to invite critical response, and most likely to make the generator appear strange or poorly calibrated to the group’s norms. The evaluation apprehension mechanism ensures that the contributions most likely to be suppressed are the ones brainstorming was specifically designed to elicit.

The effect is strongly amplified by hierarchy. When a senior leader is in the room, employees voice fewer and more conservative ideas. When people believe other group members are more knowledgeable than they are, their own contributions decrease. The most valuable brainstorming contributions — counterintuitive ideas that diverge from current group consensus — are precisely the contributions that evaluation apprehension most reliably prevents.

If the social inhibition described here resonates as more than a meeting problem — if the fear of judgment is significantly affecting your sense of safety in your working environment or beyond — support is available. UK: Samaritans (116 123, free, 24/7). Mind (0300 123 3393). international: internationaltherapistdirectory.com

Social matching: how groups anchor performance downward

The third mechanism is social matching — the process by which individuals calibrate their effort to what they perceive others doing. Paulus and Dzindolet’s 1993 research found that participants in brainstorming groups reduced their generation rate when they perceived that other group members were not working hard. The social comparison mechanism anchors performance to the observed group norm — and because the group norm is depressed by production blocking, the anchoring effect compounds the original deficit.

The spiral is self-reinforcing. As each person reduces their generation rate to match the perceived norm, the norm itself declines, prompting further reduction. The characteristically flat performance curve of real brainstorming groups after the first few minutes is this spiral operating in real time.

What Diehl and Stroebe’s camera experiment revealed

In one of the most elegantly designed experiments in the brainstorming literature, Diehl and Stroebe filmed participants individually to make each person’s contributions identifiable, specifically to test the social loafing component. The filming eliminated the social loafing deficit — but production blocking and evaluation apprehension continued to produce performance deficits relative to individuals working alone. The three mechanisms are independent. Addressing any one of them in isolation does not recover the full performance advantage of solo working.

What actually works

The Nominal Group Technique — developed in 1975 specifically in response to the brainstorming evidence — separates the generation phase from the social phase. Participants first generate ideas individually and silently, then share and discuss them in a structured round-robin format. By removing social dynamics from the generation stage, NGT eliminates both production blocking and evaluation apprehension at the point where cognitive demand is highest, while preserving the elaboration and combination benefits of group discussion afterward.

Brainwriting 6-3-5 takes a similar approach mechanically: six participants simultaneously write three ideas on a sheet, then pass it and build on others’ ideas for five rounds, generating up to 108 ideas in 30 minutes. Because everyone generates simultaneously in writing, production blocking is eliminated entirely. The initial anonymity of written contributions also substantially reduces evaluation apprehension.

Electronic brainstorming eliminates production blocking through simultaneous digital generation and reduces evaluation apprehension when anonymity is provided. Research shows it consistently outperforms face-to-face brainstorming, with the advantage increasing with group size as production blocking’s arithmetic cost grows.

IDEO’s process architecture, arrived at through practitioner testing rather than academic citation, independently instantiates the same principles: individual generation time before group sessions, structured protocols preventing early idea convergence, and facilitation specifically trained to counteract dominance and social matching. The congruence between the research recommendations and what a commercially successful innovation firm discovered through practice is the strongest available real-world validation of the findings.

A book worth reading alongside this

Group Genius: The Creative Power of Collaboration by Keith Sawyer is the most practically grounded treatment of what actually makes group creative processes work — and what does not. Sawyer synthesises the brainstorming research literature and identifies the specific conditions under which group interaction adds genuine value to individual creativity as opposed to subtracting from it. His argument is not that groups are always worse than individuals but that the specific format of Osborn’s face-to-face brainstorming is systematically designed to activate the mechanisms that produce the deficit, and that redesigning the process architecture along the lines the research prescribes recovers most of the lost value. For any entrepreneur who runs team ideation sessions and wants to understand both why the current format is underperforming and what specifically to replace it with, this is the most direct available starting point.

Have questions about this article?

If any part of this article raised questions you want to explore further, courbot.co is built for exactly that. It is courben.co’s AI assistant, designed around the psychology of entrepreneurship. Ask it anything from this article.

If the dynamics described in this article — social inhibition, fear of judgment, suppression of genuine thinking in group settings — are significantly affecting your wellbeing, speaking with a psychologist is the right next step. UK: Samaritans (116 123, free, 24/7). Mind (0300 123 3393). BACP therapist finder: bacp.co.uk/search/Therapists. International: iasp.info/resources/Crisis_Centres. Crisis Text Line — text HOME to 741741.

This article is for educational and informational purposes only. Sources: Diehl, M. & Stroebe, W. (1987), Journal of Personality and Social Psychology, 53(3), 497–509. Mullen, B., Johnson, C. & Salas, E. (1991), Basic and Applied Social Psychology, 12(1), 3–23. Paulus, P.B. & Dzindolet, M.T. (1993), Journal of Personality and Social Psychology, 64(4), 575–586. Taylor, D.W., Berry, P.C. & Block, C.H. (1958), Administrative Science Quarterly.