We Knew It All Along: Hindsight Bias in Groups

In Exp 1, with 379 high school students, small groups vs individuals made hypothetical predictions with or without outcome information. Both individuals and groups distorted their predictions in the direction of the alleged outcome. Exp 2 employed a memory design in which individuals vs groups made a series of predictions for which they subsequently received outcome information that was either above or below their prediction or for which they received no outcome information. 72 university students had to recall their initial prediction. Results indicate that (1) hindsight bias was slightly attenuated in groups, (2) groups were more likely to recall their original predictions correctly, (3) this recall advantage of groups disappeared when time taken to make the initial prediction was held constant, and (4) outcome information affected hindsight bias but not hit rates. Results were interpreted as supporting the response bias perspective.