The intuition behind committees: more perspectives, more information, better outcomes, but the research tells a more complicated story.

Groupthink: harmony at the cost of accuracy

Irving Janis developed the groupthink theory in 1972 after analysing some of the most consequential decision failures of the twentieth century — the Bay of Pigs invasion, Pearl Harbor, the Vietnam escalation. His central finding was that highly cohesive groups under pressure systematically trade decision quality for group harmony.

The mechanism is conformity. Members who hold doubts self-censor. Dissent feels like a violation of group loyalty rather than a contribution to group accuracy. The result is an illusion of unanimity — a false consensus that nobody individually endorses but everyone collectively produces.

The conditions Janis identified as precursors — high cohesion, strong directive leadership, high-stakes decisions under time pressure — describe most founding teams almost exactly. Early-stage companies are structurally primed for groupthink. The culture that makes them move fast is the same culture that makes critical challenge feel unwelcome.

The most evidence-based single intervention Janis recommended is assigning a formal devil’s advocate role — someone structurally tasked with making the best case against the group’s current direction. The role matters specifically because it makes dissent legitimate rather than an act of disloyalty.

Group polarisation: deliberation makes positions more extreme

The second mechanism contradicts the most common assumption about committees — that they produce cautious, balanced consensus. Moscovici and Zavalloni’s 1969 research with 140 secondary school students found the opposite: after group discussion, participants held more extreme views than they did before it. Group discussion amplifies the direction the group already leans rather than moderating it.

Two mechanisms drive this. Under persuasive arguments theory, members who lean in one direction are disproportionately exposed to arguments supporting that direction during discussion — creating an information environment biased toward amplifying the initial lean. Under social comparison theory, individuals shift toward more extreme positions to signal appropriate commitment to the group’s direction.

The diffusion of responsibility dimension compounds this in committees specifically. When a group collectively owns a decision, each individual bears only a fraction of the accountability. That fractured accountability makes each member willing to advocate for a more extreme position than they would endorse if the decision were entirely their own.

A committee is not a moderation mechanism, it is an amplification mechanism for whatever direction the group already leans.

Information pooling failure: groups discuss what everyone already knows

Stasser and Titus’s 1985 research at Miami University produced the most practically underappreciated finding in group decision-making. In a series of experiments, they found that groups systematically fail to discuss information held uniquely by individual members — instead spending most deliberation time on information already common to everyone.

This is the common knowledge effect: the probability of information being raised in a group meeting is proportional to how many people already hold it, not to how decision-relevant it is. The committee member with the uniquely critical perspective — the one data point that would change the group’s direction — is the least likely person in the room to have that information actively surface and processed.

The practical consequence is a committee that is less informed than it could be, despite collectively holding the knowledge required for a better decision. The group’s deliberation creates an information environment systematically biased toward confirming what members already believe.

The structural correction is simple: ask each member to submit their single most important piece of information or perspective before the meeting rather than during it. Equalising the salience of shared and uniquely-held information before discussion begins directly counteracts the mechanism.

The honest qualification

James Surowiecki’s research in The Wisdom of Crowds established that under specific conditions — genuine diversity, independence of judgement, and proper aggregation — groups do outperform individuals. The research this article describes is not a case against collective input. It is a case against the specific conditions that committees typically create: high conformity pressure, shared information dominance, and diffused accountability.

The question worth asking before any committee decision is which set of conditions is actually present in the room.

A book worth reading alongside this

Victims of Groupthink by Irving Janis is the primary source and worth reading directly. His case-study analysis of the Bay of Pigs, Pearl Harbor, and the Cuban Missile Crisis counterexample remains the most compelling application of group psychology to high-stakes decision-making — and it is short enough to read in a weekend.

This article discusses psychological patterns documented in research on group decision-making and organisational behaviour. It is not designed to identify, diagnose, or assess any psychological condition, and it is not a substitute for professional support. If group dynamics or decision-making difficulties are significantly affecting your work or wellbeing, speaking with an organisational psychologist can provide personalised guidance that an article cannot.

This article is for educational and informational purposes only. It is not a substitute for professional psychological advice, diagnosis, or treatment. If you are experiencing significant psychological distress, please consult a qualified mental health professional.

Sources: Janis, I.L. (1972/1982), Victims of Groupthink. Moscovici, S. & Zavalloni, M. (1969), Journal of Personality and Social Psychology, 12(2). Stasser, G. & Titus, W. (1985), Journal of Personality and Social Psychology, 48(6). Isenberg, D.J. (1986), Journal of Personality and Social Psychology, 50(6).