Hindsight bias: response bias hyothesis

This model was originally conceived during eyewitness testimony research. This was to account for the fact that witnesses receiving misleading information about a previously observed event, show a poorer memory for it. Similar to the hindsight bias, the misleading information effect was originally attributed to memory impairment.

Subsequent interpretations offered the following explanation: Rather than altering existing memory traces, the new information may be used as a reference point by those who cannot remember the original information and therefore need to guess. Misleading information does not alter the original representation, but simply serves as an anchor to perceivers who are unable to retrieve it. Parallels between the hindsight bias and the misleading information paradigms are obvious. Both lines of research query whether information stored in the memory might be less accessible after being confronted with inconsistent new information. The experimental designs of both research traditions show strong similarities. Important differences: In the misleading information paradigm, the original information is presented by the experimenter. In hindsight bias studies however, the original estimates are generated by the subjects. Misleading information is presented unobtrusively without the subjects being aware of its misleading nature. Outcome information given in hindsight studies is explicitly labeled as the correct information.

Return to: biased reconstruction

Literature: McCloskey & Zaragoza (1985), Stahlberg & Maass (1998)

Entry by: Stefan Schwarz

June 11, 1999
Direct questions and comments to: Glossary master