Search

In previous studies, Jones and Harris found that persons who wrote a favorable or unfavorable essay about Fidel Castro were judged as holding attitudes that corresponded to the essay's position, even when they had been compelled to endorse the position. The present research tested the robustness of this correspondence bias. In Study 1, subjects read two essays about Castro that they believed were written 9 months apart by the same author under forced-choice conditions. The correspondence bias emerged in subjects' judgments of the author's attitudes toward Castro at the time each essay was composed. Essay writers who endorsed two opposite positions were judged to have undergone greater attitude change than writers who twice endorsed the same position. Study 2 demonstrated the effect with an interval as brief as 1 week between constrained behaviors. These data underscore the robustness of the correspondence bias.