How choice modifies preference: Neural correlates of choice justification

Abstract

When making a difficult choice, people often justify the choice by increasing their liking for the chosen object and decreasing their liking for the rejected object. To uncover the neural signatures of choice justification, we used functional magnetic resonance imaging to monitor neural activity when subjects rated their preference for chosen and rejected musical CDs before and after they made their choices. We observed that the trial-by-trial attitude change (i.e., increase of preference for chosen items and decrease of preference for rejected items) was predicted by post-choice activity in the ventral medial prefrontal cortex (MPFC), right temporal–parietal junction, anterior insula, and bilateral cerebellum. Furthermore, individual difference in choice justification (i.e., increased preference for chosen items minus decreased preference for rejected items) was predicted by post-choice neural activity in the dorsal MPFC, left lateral prefrontal cortex, and right precentral cortex positively. In addition, interdependent self-construal was correlated with decreased activity in the ventral MPFC in the post-choice than pre-choice sessions. These findings suggest that both negative arousal/regulation and self-reflection are associated with choice justification. This provides evidence for the self-threat theory of choice justification.

Introduction

Cognitive dissonance has been investigated using a wide variety of methodologies (see Harmon-Jones and Harmon-Jones, 2007 for a review). One of the most commonly used experimental paradigms involves a choice between two equally attractive objects (Brehm, 1956). Numerous behavioral studies have shown that, after making a difficult choice, people justify this choice by increasing their liking for the chosen item and decreasing their liking for the rejected item. The choice justification is believed to occur because people are motivated to reduce their cognitive conflict or dissonance (Brehm, 1956, Festinger, 1957). Researchers have hypothesized that this choice-induced conflict, and the resulting dissonance reduction, may be most likely to occur when the conflict poses a threat to a person's private sense of the self as rational and competent (Steele, 1988), the sense of the self as publicly recognized as rational and decent (Kitayama et al., 2004, Tedeschi and Reiss, 1981), or both.

Recent functional magnetic resonance imaging (fMRI) studies have provided some insight into the neural correlates of dissonance. In one study, van Veen et al. (2009) found that neural activity in the dorsal anterior cingulate cortex (dorsal ACC) and in the anterior insula increased to statements that conflicted with subjective feelings. This suggests that detection of cognitive conflict (dorsal ACC) and aversive somatic arousal (anterior insula) constitute important elements of cognitive dissonance, as implied by Festinger (1957) in his original formulation. Indeed, as would be predicted by the dissonance theory, the dissonance as indexed by the activity in these brain regions during the choice predicted subsequent attitude change in the form of justifying the dissonance-producing behavior (van Veen et al., 2009).

In another study, Jarcho et al. (in press) found that choice justification is reliably predicted by increased activations in the right inferior frontal gyrus and medial frontoparietal regions during the choice. The finding suggests that choice justification is mediated by regulation of negative arousal through inhibition of both competing information (right inferior frontal gyrus) and conscious attention (frontoparietal regions) to the chosen and rejected items. This study, however, scanned the brain only during the choice. Therefore, it is not clear whether and how representations of the chosen and rejected items might change as a consequence of choice.

A more recent study (Sharot et al., 2009) addressed this issue by testing choices among hypothetical vacation destinations. It was found that the attitude change involved in dissonance was mirrored by caudate activations in relation to the chosen and rejected items after the choice. However, this finding might not be applicable to difficult decisions involving objects one may actually be able to actually possess. Given this, much has yet to be learned about the neural mechanisms underlying post-decisional choice justification or dissonance reduction.

To fill the gap of empirical knowledge on neural mechanisms underlying cognitive dissonance, we used a modified free-choice paradigm and scanned healthy young Chinese adults as they rated a set of CDs both before and after making a series of choices between these CDs. During the choice, the CDs were paired in such a way that the two CDs in each pair were equally attractive, as previous work shows that dissonance arises only when choices are difficult (Brehm, 1956, Sharot et al., 2009, Jarcho et al., 2010). Moreover, in order to increase choice justification during the post-choice sessions, subjects were reminded which choice they had made earlier. We had two primary aims.

First, we aimed to investigate the brain regions recruited when subjects justified their choices. Previous research has found that choice justification is eliminated when one's sense of the self is affirmed after making a difficult choice (Hoshino-Browne et al., 2005, Steele, 1988). This supports the proposal that individuals justify their choice in order to eliminate a threat to the self. On the basis of this literature, we predicted that self-related brain areas such as the ventral MPFC (Kelley et al., 2002) and the dorsal/ventral lateral prefrontal cortex (Lieberman, 2010) would be engaged in post-decisional choice justification. Furthermore, because the public sense of the self involves taking the perspectives of others (Imada and Kitayama, 2010, Kitayama et al., 2004), we anticipated that brain areas implicated in mind reading such as temporal–parietal junction (TPJ, e.g., Saxe and Kanwisher, 2003) and dorsal MPFC (e.g., Gallagher et al., 2000) might also be related to choice justification. In addition, since individuals justify their choices by inhibiting choice-inconsistent information while augmenting choice-consistent information (Jarcho et al., in press), we predicted that the brain areas implicated in regulation, such as the dorsal MPFC (Venkatraman et al., 2010), the dorsal LPFC (Ochsner and Gross, 2008), and the inferior frontal gyrus (Jarcho et al., in press), would also be involved.

Second, we aimed to examine whether, similar to the Sharot et al. (2009) study, choice justification might be tracked by neural activity that is related to subjectively experienced preferences. We expected that neural activities reflecting subjects' preferences, such as caudate (Sharot et al., 2009), ventral MPFC (McClure et al., 2004), and/or PCC (Kawabata and Zeki, 2008), would be altered by choice justification. In addition, given cultural differences in cognitive dissonance (Hoshino-Browne et al., 2005, Imada and Kitayama, 2010) and considerable variation within cultures in the extent to which they endorse their cultural norms, we assessed the relationship between change in the neural signatures related to subjects' preference and individual differences in independent self-construals (i.e., the view the self as an autonomous entity separate from others) and interdependent self-construals (i.e., the view of the self as interconnected with others as well as the social contexts; Markus and Kitayama, 1991).