A hiring panel is considering three candidates. Every member has already seen several strengths belonging to Candidate A, but only one person received a serious warning sign. During the meeting, the shared strengths can be repeated by several people. The unique concern gets only one opportunity to be voiced and may be brushed aside after an early preference has formed. The group collectively possesses the evidence needed for a better judgement, yet may never combine it.
Stasser and Titus studied this problem systematically with “hidden profile” experiments. Information is distributed among members so that the best option can be identified only by pooling unique clues, but discussion tends to favour material everyone already knows. This is called shared-information bias. Shared facts have more potential speakers; each repetition also feels like confirmation that the group agrees and that the information matters. Unique evidence is less likely to be repeated or socially verified, and it may conflict with an emerging preference.
A meta-analysis covering twenty-five years of hidden-profile research confirmed that these tasks remain difficult for groups, but the outcome is not fixed. Discussion time, task type, awareness of who knows what, and motivation to be accurate rather than merely reach agreement can all affect information exchange. A failed meeting therefore cannot be reduced to one quiet member or a defective team personality. The safer central judgement is that bringing people together creates an opportunity to pool knowledge; it does not perform the pooling automatically. When a decision genuinely depends on distributed evidence, asking members to report unique facts before stating preferences may expose hidden clues more effectively than beginning with a vote. That is a process design, not a guarantee for every meeting.
https://doi.org/10.1037/0022-3514.48.6.1467
https://doi.org/10.1177/1088868311417243
Discover more from Geoffrey Chen
Subscribe to get the latest posts sent to your email.