How to Study Choice-Induced Attitude Change: Strategies for Fixing the Free-Choice Paradigm

The theory of cognitive dissonance has been among the most influential theories in social psychology for the last 50 years. Support for the theory has come primarily from three experimental paradigms: free-choice, induced compliance, and effort justification. Recently, Chen and Risen (2010, Journal of Personality and Social Psychology, 99, 573‐594) have argued that although the free-choice paradigm reliably finds a ‘spreading of alternatives’ (i.e., after making a choice, participants’ evaluations of the chosen item improve and evaluations of the rejected item decline), these results cannot be interpreted as evidence for dissonance reduction or attitude change. Unlike the other dissonance paradigms, participants ‘self-select’ how they are treated in the free-choice paradigm, making it impossible to know whether spreading is because of the choice process or the information that is revealed about participants’ existing preferences. The current paper has two goals. First, we will describe the criticism developed by Chen and Risen (2010) and situate the criticism within the broader study of dissonance. Second, we will offer four suggestions for how researchers can isolate the effect of the choice process and properly test for choice-induced attitude change in the free-choice paradigm. In the mid-1950s ‐ during the heyday of consistency theories ‐ Leon Festinger developed the theory of cognitive dissonance. Since its formalization in 1957, dissonance theory has become one of the most influential theories and most researched topics in social psychology, tested primarily within three paradigms: free-choice, induced compliance, and effort justification. Recent research suggests, however, that the very first paradigm used to test dissonance ‐ the free-choice paradigm (Brehm, 1956) ‐ is fundamentally flawed. Chen and Risen (2010) have argued that although the spreading of alternatives has been found reliably in hundreds of free-choice experiments, it arises as an artifact of a flawed methodology and cannot be taken as evidence of dissonance reduction or attitude change. The current paper is written with two primary goals. First, we will describe the criticism developed by Chen and Risen (2010), situating it within the broader study of dissonance. We will explain how ‘self-selection’ occurs in the free-choice paradigm (FCP) and why the nonrandom treatment of participants is so problematic. Second, because the criticism is meant to improve the study of dissonance, we will offer four suggestions for properly studying choice-induced attitude change in the FCP. Researchers can isolate the effect of the choice process on subsequent preferences by (i) ensuring that all participants make the same choice, (ii) controlling for the information revealed by choice, (iii) removing the information revealed by choice, or (iv) manipulating the choices that people make.