On the Robustness and Generality of the Correspondence Bias

People routinely assume correspondence between acts and dispositions, a systematic error prior research has labeled the "correspondence bias." Four laboratory studies investigate the robustness and generality of this tendency, and suggest that it may be even more fundamental than prior theories have supposed. Most of the research documenting the correspondence bias uses paradigms in which the behavior is more salient and easier to assess than the situation. Our studies employ a new paradigm in which people have perfect information about both the situation and the behavior. Using this paradigm across different settings, we find that the correspondence bias generalizes to inanimate objects. Furthermore, the bias persists in situations where avoiding it amounts to a simple subtraction problem, and even in the face of a wide array of debiasing manipulations designed to eradicate it. The results provide new insights into the ultimate causes of the correspondence bias.