Mindset effects on information search in self - evaluation UTE

Research on mindset theory (GollwitTer & Baye4 1999) observed that people in an implemental mindset show an orientation towards positive illusionary self-evaluations, whereas people in a deliberative mindset opt for accura.te self-evaluations. In the present study, we tested whether these self-evaluative orientations and the associated search for certain types of self-relevant information (feedback) are moderated by low versus high self-views. With high self-view participants we observed the hypothesized mindset effects on information search, but we obtained the reverse pattern for low self-view participants. The latter finding points to self-defensiveness in low self-view individuals. Implications are discussed in terms of the consequences of accurate versus positive illusionary selfevaluations for the successful control of goal pursuits, and individual dffirences in mindset effects. Copyright @ 2005 John Wiley & Sons, Ltd. Self-evaluation is guided by different motives or purposes (Pomerantz, Saxon, & Kenney, 2001; Sedikides & Skowronski, 2000; Sedikides & Strube, 1997). First, a concern for self-assessment motivates people to reduce uncertainty about their abilities and personal attributes. This is achieved by performing high diagnostic tasks and searching for diagnostic information (Trope, 1936). Second, people's self-evaiuations also serve self-enhancement concerns. The self is protected from negative information by selectively processing positive information. The valence of feedback and the personal importance of the attribute in question are of primary importance. People guidedby self-enhancement concerns will thus find information diagnostic of success, high ability, or any other positive personal attribute to be more attractive than information diagnostic of failure, low ability, or any other negative personal attribute (Brown & Dutton, 1995; Kunda,1990; Taylor & Brown, 1988). Thfud, people's self-evaluations are also guided by self-verification concerns which aim at endorsing preexisting self-conceptions. Self-verification applies to both positive and negative aspects of the self. People seek verification of their certain self-concepts to a larger degree than $reir uncertain self-concepts. What matters is the consistency between self-concept and feedback rather than selfconcept valence or feedback valence (Swann, 1990,1997). Finally, people are motivated to improve their traits, abilities, skills, health status, or well-being. This motive is conceptually different from the *Correspondence to: Ute C. Bayeq Social Psychology and Motivation, University of Konstanz, Box D39, D-78457 Konstanz, Germany. E-mail: bayer@ soz.psychologie.uni-konstanz.de Received I3 November 2003 Accepted 31 August 2004 Copyright O 2005 John Wiley & Sons, Ltd. 314 Ute C. Bayer and Peter M. Gollwitzer other three motives (Taylor, Neter, & Wayment, 1995). Self-improvement focuses on genuine improvement, which does not necessarily include self-concept positivity. Attempts at self-improvement will result in a sense of progress, hope, or growth (Collins, 1996; Wheeier, 1966). Operating on the basis of one or the other of the four described concerns of self-evaluation has differential consequences for the type of self-knowledge that accrues, each of which comes with its typical advantages and usefulness (Sedikides & Skowronski, 2000). Serving self-assessment concerns leads to accurate self-knowledge which makes it easier for people to select tasks that will not overtax them. Serving self-enhancement concerns, on the other hand, leads to favourable self-concepts which sffengthen people's resilience in overcoming barriers to their goal pursuits and in dealing with everyday hardships. Serving self-verification concerns leads to stable self-views that allow for easy predictability of one's future behaviours (and the reactions of others towards us). Finally, selfimprovement focuses on self-betterment regardless of self-concept accuracy. Even though the four self-evaluation concerns listed are rather pervasive and accompany almost all ofpeople's daily pursuits, researchers have started to search for those variables that favour one ofthese concerns over the other. For instance, it was observed that feeling uncertain about one's abiiities (Trope & Bassok, t982), feeling good about oneself in general (Trope & Neter, L994), and placing a good deal of importance to be competent in a particular area (Trope & Pomerantz, 1998) strengthen the self-assessment orientation. Self-verification concerns, on the other hand, are the more prevalent the more certain people are of their self-conceptions, as peopie are motivated to defend their existing self-concepts against threats (Swann & Ely, 1984). With respect to self-enhancement it was observed that this concern is particularly strong in people with high self-esteem (Wayment & Tayior, 1995), and when the personal atffibute implied is perceived as fixed and nonmodifiable (Dunning, 1995). Finally, feelings of threat or inadequacy are more likely to instigate the self-improvement motive. It was observed that upwards comparisons for the purpose of self-improvement occur frequently when people have to cope with novel situations (e.g. in coping with cancer, Taylor, Aspinwall, Guilano, Dakof, & Readon, 1993; perfonnance on novel tasks, Buunk,1995). MINDSETS AS MODERATORS OF SELF.ENHANCEMENT AND SELF.ASSESSMENT CONCERNS Taylor and Brown (1988) proposed that a mentally healthy person is characterized not by accurate assessments of her or his qualities but typically by holding mildly self-aggrandizing perceptions of the self. They argued that instead of being maladaptive these positively distorted self-perceptions actually foster positive self-regard, the ability to care for and about other people, the capacity for creative and productive work, and the ability to effectively manage stress. Despite the obvious positive consequences of moderate self-enhancement, hese flndings raise the disturbing question: How do people with inflated self-concepts effectively identify and make use of negative feedback they may encounter in the world? One potential resolution is the possibiiity that there may be times when people are more honest with themselves, during which they recognize and incorporate negative feedback. Mindset theory (Goliwitzer, 1990, 2003) suggests a set of circumstances"when such a window to realism opens up. It is argued that successful goal pursuit involves solving four consecutive tasks: choosing between potential goals, planning the implementation of a chosen goal, acting on the chosen goal, and evaluating what has been achieved. When people get involved in these tasks, different cognitive procedures are said to become activated (i.e. different mindsets) which make it easier to live up to the respective task demands. More specifically, it is argued that deliberation in the predecisional phase invoives the careful appraisal of whether one could pursue (or not) a potential goal. Therefore, Copyright @ 2005 John Wiiey & Sons, Ltd. Eur. J. Soc. Psychol. 35,313-327 (2005) Mindset and self-evaluation 315 the pros and cons in regards to the potential goals have to be weighted and the feasibility of these goals has to be assessed. Consequentiy, the respective cognitive procedure (i.e. the deliberative mindset) should foster relatively even-handedand accurate appraisals of evidence. With respect to selfassessment concerns, mindset theory predicts that the deliberative mindset should foster the accurate analysis of feasibility-related information on potential goals. Planning the implementation of a chosen goal, however, poses a problem associated with different task demands. Deliberation of the goal in question needs to be ended and people need to look toward implementation. Good opportunities need to be discovered and linked to appropriate goal-directed behaviours (i.e. plans need to be made with respect to when, where, and how one wants to act). Consequentiy, the respective cognitive procedure (i.e. the implemental mindset) should foster a biased analysis of feasibility and desirability-related information so that deliberation of the goal will not start anew. The person starts to conceive of the feasibility of chosen goals in an overly optimistic way and views the desirability of the chosen goal in a partial manner (i.e. sees more pros than cons). Moreover, people tune in to implementation-related information as it is needed to get started on one's goals. A research progranme aimed at testing the proposed different cognitive features of the deliberate and implemental mindset reported by Gollwitzer and Bayer (1999; Gollwitzer, 2003).In this research, the deliberative mindset was induced by asking participants to either extensively deliberate an unresolved personal problem by listing the short-term and long-term pros and cons of both making and not making a goal decision. For the implemental mindset, participants were asked to list the five most important steps of implementing a chosen goal, and then to specify, when, where, and how they intent to execute each step. Thereafter, both the deliberating and planning pafücipants were asked to perform presumably unrelated intellectual tasks (usually performed by a different experimenter in a different situational context). Indeed, research participants placed into an implemental mindset (by asking them to plan the implementation of an important life decision they had already made) reported strong illusions of control over frequent, but uncontrollable outcomes in a classic contingency learning task (Alloy & Abramson, 1979), whereas deliberative mindset participants (who had to contemplate the pros and cons of making a major life decision) indicated reduced illusions (Gollwitzer & Kinney, 1989). Taylor and Gollwitzer (1995) explored the influence of deliberative and implemental mindsets on several other self-relevant judgments. It was again observed that participants in the deliberative mindset were in a position to open a window to realism and thus reported a less positive illusionary standing on various personal attributes (e.g. cheerfulness, academic ability), as well as less positive illusionary judgments of their invulnerability to controllable (e.g. divorce, having a drinking problem) and uncontrollable risks (e.g. developing a heart disease, loosing a limb). More recently, Armor and Taylor (2003) reported that implemental mindsets do not only produce enhanced self-efficacy, optimistic outcome expectations, and perceptions of the task at hand as easy, but also could show that these positive illusions helped people to succeed at the task at hand. INDTVIDUAL DIFFERENCES IN MINDSET EFFECTS: THE IMPACT OF LOWHIGH SELF.VIEWS Meanwhile, research on mindsets has addressed individual differences in the activation of deliberative and implemental mindsets and their effects on cognition and behaviour. For instance, mindset effects have been found to be dependent on a person's achievement motivation (Pucca & Schmalt, 2001), social anxiety (Hiemisch, Ehlers, & Westermann, 2002), and goal commitment (Gagn6 & Lydon, 2001). In the present study, we investigate the impact of a person's low/trigh self-view as a potential Copyright @ 2005 John Wiley & Sons, Ltd. Eur J. Soc. Psychol.35,313-327 (2005) 316 Ute C. Bayer and Peter M. Gollwitzer moderator of mindset effects. More specifically, we investigate the search for self-relevant information by deliberative versus impiemental participants holding either a high or low self-view. As Trope and Neter (19,94) have pointed out, the processing of feedback about one's abiiities has two types of consequences: At an informational level, people learn about their standing on the respective ability. At an emotional level, positive feedback induces positive feelings (i.e. pride) and negative feedback leads to negative feelings (i.e. self-threat, disappointment, and self-doubts). According to Trope and Neter, people consider such potential costs and benefits of self-relevant information and their processing of self-relevant information guided by these anticipated positive and negative consequences. Following a cost-benefit perspective, it depends on a person's low versus high self-concept whether self-assessment or self-enhancement concerns are served in deliberative and implemental mindsets. With individuals holding a high seif-concept, self-assessment concerns triggered by a deliberative mindset are emotionally not very risky as she or he rightfully anticipates positive feedback. And serving self-enhancing concerns in the implemental mindset is informationally not very risky as positive illusions help the individual to implement the chosen goal (Armor & Taylor, 2003). Things are quite different with individuals holding low self-views. The accuracy orientation triggered by the deliberative mindset should create a state of self-threat in these individuals, as chances are high that they wili have to face weaknesses. Accordingly, they should refuse to process information accurately and instead serve self-improvement/self-enhancement concerns. Moreover, for participants holding a low self-view the positive illusions triggered by the implemental mindset are inconsistent with their self-view and thus should create a fear of invalidity. Accordingly, they should not serve selfenhancement concerns but modestly stick to an assessment orientation. These predictions are in line with recent research on deliberative and implemental mindsets that looks at individual differences as potential moderators of mindset effects. For example, Pucca and Schmalt (2001) anaiysed the effects of a person's achievement motive (fear of failure vs. hope of success) on thought content in deliberative and implemental mindsets. Participants either had to deliberate which of two different complex reaction time tasks they wanted to perform (deliberative mindset) or to choose one of these tasks and then plan task performance of the chosen task (implemental mindset). Interestingly, the classic finding of an optimistic bias in thought content in implemental as compared to deliberative individuals (Heckhausen & Gollwitzer, 1987: Taylor & Gollwitzer, 1995, Study 3) was moderated by the achievement motive. Whereas hope of success individuals showed the classic pattern, the reverse was true for fear of failure individuals: The latter reported more thoughts about strengths than about weaknesses in performing the difficult vigilance task in a deliberative mindset, and reported more thoughts about weaknesses rather than strengths in an implemental mindset. Apparently, chronic fear of failure individuals dealt with the self-threat triggered by a deliberative mindset by boosting their self-perception of competence. And chronic fear of failure individuals responded to the optimistic orientation (hopes) triggered by the implemental mindset with intensifying their fear of failure (i.e. thinking of having chosen the wrong task). Hiemisch et al. (2002) placed people with high versus low social anxiety in deliberative and implemental mindsets and measured their processing of deliberative (i.e. pros and cons of choosing a goal) versus implemental information (i.e. when, where, and how of goal pursuit) in solving a critical interpersonal problem. Low socially anxious people showed better procÖssing of implemental as compared to deliberative information in the implemental mindset and equal processing in the deliberative mindset. High socially anxious people, on the other hand, processed deliberative information better than implemental information in the implemental mindset, whereas the reserve was true for the deliberative mindset. The authors suggest that socially anxious people coped with the threat of the deliberative mindset (having to look not only on sffengths but also on weaknesses) by committing themselves to a certain way of solving the interpersonal problem at hand, thus entering an implemental Copyright O 2005 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol.35,313-327 (2005) Mindset and self-evaluation 3I7 mindset that protected them from having to look at their weaknesses. Socially anxious people who were in an implemental mindset to begin with, on the other hand, quickly discovered problems of solving the social problem at hand.-(because of their social anxieties) which in turn triggered deliberation of what kind of alternative goals they had better pursued. Thus, they entered a deliberative mindset which facilitated the processing of deliberative information. Finally, Gagn6 and Lydon (2001) suggest hat deliberation of goal decisions that have already been made can initiate defensive processing of information rather than accurate processing. In one study, they asked participants highly committed to a romantic relationship to deliberate this relationship decision or a nonrelationship goal decision. They found that when asked to rate how their partner compared with the average, those individuals asked to deliberate over the relationship goal decision gave much higher ratings than those who were asked to deliberate over a nonrelationship goal decision. Of interest, these ratings were also higher than those of implementation participants who had been planning the implementation of the relationship goal. Gagn6 and Lydon suggest that the deliberation of a relationship goal decision may have been perceived as threatening, resulting in an enhancement of the partner's positive attributes. Testing this idea, they assessed the degree of commitment participants felt to their relationship in a second study and hypothesized that high commitment participants should feel more threatened by the deliberative mindset than low commitment participants. Indeed, they found that high commitment but not low commitment participants defended against a threat of a deliberative mindset by increasing their positive views of their partner. SEEKING INFORMATION ON INTELLECTUAL CAPABILITY IN DELIBERATWE VERSUS IMPLEMENTAL MINDSETS In the present study, we assess whether high versus low self-view individuals differentially serve selfassessment, self-enhancement/improvement and self-verification concerns when placed in a deliberative versus an implemental mindset. The strengths of these concerns can be assessed via the strategies people use to gather self-relevant information (Kunda, 1990; Trope & Liberman, 1996). For exampie, in the service of self-enhancement, people spend more time reading favourable than unfavourable information about themselves (Baumeister & Cairns, 1992), and they search for ability-relevant information more intensively when they have reason to believe that they will gain positive rather than negative information about themselves (Brown, 1990). In the service of selfverification, people seek that kind of feedback (positive or negative) that they believe will confirm their self-concepts. For instance, people prefer interaction partners from whom they expect to receive confirmatory feedback with respect o their positive and negative attributes (Swann, Stein-Seroussi, & Giesler, 1992). A diagnostic sffategy of information search, on the other hand, leads people to ask high over low diagnostic questions (Trope & Bassok, 1982) or to choose high over low diagnostic tasks (Brown, 1990; Strube et al., 1986). Moreover, a diagnostic orientation is indicated by the extent to which favourable and unfavourable information is sought in an even manner (Trope & Neter, 1994). A self-improvement strategy is commonly investigated in terms of upward social compärisons as such comparisons provide a standard to strive for and offer an inspiration to meet the goal (Collins, 1996). It is seldom the case that the strength of the different self-evaluation concerns is measured in one and the same study (Sedikides, I993),In the present study, we constructed a feedback questionnaire that allowed determining the extent to which the participants are guided by self-enhancement, selfverification, or self-assessment concerns. This questionnaire offered favourable and unfavourable feedback, and at the same time varied the diagnosticity of this feedback (Devine, Hirt, & Gehrke, Copyright O 2005 John Wiley & Sons, Ltd. Eur J. Soc. Psychol.35,313-327 (2005) 318 Ute C. Bayer and Peter M. Gollwitzer 1990). The personal attribute we focused on was intellectual abilities, and thus we analysed a attributespecific self-view. We consider this attribute to be of high personal importance. Accordingly, selfdefensiveness with respect,to having to admit that one is low on this attribute should be rather high. All participants were informed that they would take part in two different experiments. In the presumed first experiment, they worked on an ostensible personality and ability test which involved questions about five personal attributes including the critical aspect of intellectual skills. In the presumed second experiment, participants worked on a task designed to induce either the deliberative or implemental mindset. Thereafter, we offered the participants feedback on their inteilectual abilities. Participants were asked to indicate their preference with respect to eight aspects of possessing or lacking intellectual abilities (four aspects related to strengths, and four to weaknesses; half of the strengths and weaknesses related questions were diagnostic, the other half nondiagnostic). We interpret a general preference for information on strengths over weaknesses as serving selfenhancement concerns. Serving self-assessment concerns in contrast is indicated if participants choose high diagnostic information over low diagnostic information, and if they choose information about strengths and weaknesses in an even manner. Finally, serving self-verification concerns is indicated when participants with low self-views prefer information on weaknesses over strengths and participants with high self-views prefer information on sffengths over information on weaknesses. We expected that participants holding a high self-view should be guided in their information search by self-assessment concerns in a deliberative mindset and by self-enhancement concerns in an implemental mindset. For participants holding a low self-view we expected an inverse pattern: Their preferences for self-relevant information should be guided by self-defensiveness. Feelings of selfthreat triggered in the deliberative mindset should lead to self-enhancement/improvement concerns in the deliberative mindset and fear of invalidity triggered in the planning mindset should lead to assessment concerns in the implemental mindset.