یک تأیید خود و ذهن باز: خودتأییدی و حساسیت به قدرت استدلال

Abstract Self-affirmation seems to enable an individual to objectively evaluate information that would otherwise evoke a defensive reaction. If this objectivity reflects freedom from self-evaluative concerns, affirmation should sensitize people to central cues of a persuasive message, like argument strength. If affirmation simply induces agreeableness or trivializes the issue, affirmed participants should not particularly heed argument strength. Affirmed and non-affirmed participants rated the persuasiveness of pro- and counterattitudinal arguments that varied in strength. Among participants who rated their attitudes as personally important, self-affirmation decreased bias and increased sensitivity to argument strength, as predicted by self-affirmation theory.

Introduction Research has demonstrated that people often resist information that conflicts with their personal views. Though new information may improve understanding, people seem to be motivated to discount both the source and the content of a challenging message in an effort to protect their existing beliefs (Cohen, Aronson, & Steele, 2000; Reed & Aspinwall, 1998; Sherman, Nelson, & Steele, 2000; for a review, see Sherman & Cohen, 2002). This close-mindedness apparently stems, at least in part, from concern over self-regard. Cohen and his colleagues suggest that, because people derive a sense of worth from their beliefs, and from the notion that those beliefs are an accurate assessment of reality, contradictory information can endanger their sense of personal worth. To protect that sense of worth, they attempt to reject the conflicting information. This argument follows from self-affirmation theory (Steele, 1988; Steele, Spencer, & Lynch, 1993), which suggests that every person strives for positive self-regard and, to achieve it, draws on successes in important domains in her or his life. These domains constitute aspects of individual identity, including important social roles, abilities, and beliefs. Because the individual depends on a constellation of domains for feelings of adequacy, a threat to one of the domains can prompt a defensive reaction. A crucial tenet of self-affirmation theory, though, is that the ultimate goal of a defensive reaction is the security of the global sense of self-worth, not the security of the domain, per se. The individual should defend a given domain only to the degree that the more general sense of self-worth is compromised by its loss. Accordingly, if global self-worth is temporarily bolstered by success in a second, unrelated domain, the individual should be more willing to tolerate a threat to the domain of interest. Typically, self-affirmation research involves exactly this type of test. In research on persuasion, a self-affirmation in one domain leads people to acknowledge the merits of counterattitudinal arguments they would otherwise reject. Cohen et al. (2000), for example, asked half of their participants to write a paragraph about an important value (to affirm their sense of self-worth) before exposing them to arguments that challenged their views on capital punishment or abortion. Compared with control participants who wrote about less important values, those who wrote about a central value were more willing to recognize the strengths of the challenging argument. Mechanisms behind affirmation-based openness Though research has repeatedly demonstrated affirmation’s effect on open-mindedness, little has been done to document the ostensible underlying mechanism. Self-affirmation theory suggests that diminished concern for the self allows an affirmed individual to dispassionately evaluate persuasive communication. At least two plausible alternatives exist, however. Affirmation may reduce bias if it renders an individual generally more agreeable to any persuasive communication. This agreeableness may minimize bias by dramatically improving reactions to counterattitudinal statements, which would otherwise bear the brunt of an individual’s criticism. Affirmation may also reduce bias if it leads the individual to trivialize the issue ( Simon, Greenberg, & Brehm, 1995). If an issue is taken less seriously, the individual may neither favor a proattitudinal view so strongly nor so adamantly oppose a counterattitudinal position, and so appear less biased. Though these three mechanisms (objectivity, agreeableness, and trivialization) all predict that affirmation will promote openness to counterattitudinal arguments, they differ in important ways. Agreeableness suggests that an affirmed individual should respond more favorably, not just to counterattitudinal arguments, but to proattitudinal positions, as well (though the greater shift may be reserved for counterattitudinal views, formerly deprived of favorable attention). Objectivity and trivialization, on the other hand, represent a lack of passionate involvement, and this dispassion may be expected to foster a more critical interpretation of the proattitudinal case, even as it allows a more favorable evaluation of the counterattitudinal view. The majority of existing research has examined reactions to only counterattitudinal arguments ( Cohen et al., 2000, Studies 1 & 2; Sherman et al., 2000). To our knowledge, only two studies have included a proattitudinal position ( Cohen et al., 2000, Study 3; Reed & Aspinwall, 1998). Contrary to an agreeableness account, both report that affirmed participants were more critical of proattitudinal positions than unaffirmed participants. One goal of the present study is to replicate these findings, and bolster the case against agreeableness as a potential mechanism. A second goal is to go beyond previous research by distinguishing between objectivity and trivialization as rival accounts for affirmation-induced open-mindedness. To the extent that affirmation promotes objectivity, it should help participants pay close attention to the arguments, recognizing their merits and demerits, with the result that participants will be more impressed by strong arguments than weak arguments. Trivialization makes the opposite prediction: if affirmation leads the individual to treat the issue less seriously “in the grand scheme of things,” it should reduce careful consideration of the argument and prompt less thorough, more heuristic processing (Petty & Cacioppo, 1984; Petty & Wegener, 1998). A trivialization account, then, suggests that affirmation should reduce attention to central cues like argument strength with the result that participants will show little difference in how impressive they find strong and weak arguments. By examining reactions to both pro- and counterattitudinal messages, the present research tests the viability of an agreeableness account. By examining reactions to arguments that vary in strength, it attempts to disentangle objectivity and trivialization. Of these various alternatives, only objectivity predicts that affirmation will both foster more positive reactions to counterattitudinal arguments but less positive reactions to proattitudinal arguments (H1), and increase sensitivity to argument strength (H2). The role of attitude importance It is crucial to note that the predictions of self-affirmation theory depend on the personal importance of the attitude in question. Beliefs that are central to an individual’s sense of self should evoke greater defensive reactions in the face of a threat because of their direct relation with the global self-concept (Boninger, Krosnick, & Berent, 1995; Steele, 1988; Zuwerink & Devine, 1996). Less important beliefs should have only minor implications for the self-concept and the individual should, accordingly, make only minor attempts to defend them. Past research has focused on issues such as capital punishment, abortion, and the risk of cancer—issues related to fundamental concerns such as human rights, religion, and personal survival. These studies may be seen as focusing largely on high-importance beliefs. The present study used a less central issue, namely the desirability of a tuition increase, on which participants’ importance ratings were expected to vary. We anticipated that this variability would moderate the effects of the self-affirmation manipulation. Given high importance, we predicted that defensiveness should also be high, and the ability of self-affirmation to reduce that defensiveness and promote objectivity should be most apparent. Given low importance, defensiveness should be minimal, and floor effects should attenuate any impact of self-affirmation. Overview Participants opposed to a tuition increase watched a videotaped debate on the issue (a subject being discussed by the school administration at the time of the study). In the debate, an advocate for raising tuition and an advocate opposed to the tuition increase took turns presenting a variety of arguments (some weak, some moderate, and some strong) for their respective positions. While watching the video, participants rated the persuasiveness of the debaters’ arguments. Prior to watching the videotape, participants completed a brief scale, which constituted the manipulation of affirmation.

Results We were principally interested in the possibility that self-affirmation promotes objectivity, reducing opposition to counterattitudinal views, fostering criticism of proattitudinal views, and increasing sensitivity to argument strength. In all analyses, we expected that any effect of affirmation should depend on the subjective importance of the tuition debate issue. Only when a belief is important or self-relevant should participants react defensively to the prospect of its loss (Steele, 1988). The analysis of persuasiveness ratings involved two between-participants factors, Affirmation Condition (affirmation vs. control) and Attitude Importance (measured as a continuous variable), and two within-participants factors, Advocate Position (proattitudinal vs. counterattitudinal) and Argument Strength (weak vs. moderate vs. strong). For this analysis, we collapsed each participant’s continuous perception analyzer ratings separately for each of the 20 key arguments, then collapsed those arguments, within advocate, by level of argument strength. For each participant, then, we obtained six mean persuasiveness ratings (following the 2 × 3 within-participant element of the design): for both the proattitudinal and the counterattitudinal advocate, we had ratings of weak, moderate and strong arguments. These six mean ratings served as our dependent variables. Except where noted otherwise, the importance measure was mean centered to facilitate interpretation of the Affirmation Condition effect, in the context of the affirmation by importance interaction.