عواقب عاطفی خوشبینی و بدبینی در صورت شکست: شواهدی از یک اعتدال اسنادی

Abstract The present experiment set out to investigate the affective consequences of dispositional optimism and attribution in performance settings. Optimistic and pessimistic participants (N = 42 each) experienced failure at solving two cognitive tasks in an alleged team setting. The failure could either be attributed to themselves (internal condition) or a teammate (external condition). We found disordinal interactions of optimism and attribution on the feelings of success and feelings of failure. While the affective state of optimists deteriorated significantly if they attributed the failure internally compared to externally, pessimists were emotionally unaffected by the locus of attribution. Moreover, optimists experienced affective benefits compared to pessimists when they attributed the outcome externally. The reverse was true if they had attributed internally. Affective consequences of optimism and pessimism after failure therefore seem to differ depending on attributions. Furthermore, pessimists seemed to be unresponsive to the affective effects of attribution in our study. This insensitiveness implies differences in the cognitive processing of outcomes, a trait × cognition interaction that should be investigated further.

Introduction Dispositional optimism is usually considered to be highly functional across a wide range of domains, including health, well-being, and social resources (e.g., relationships). Pessimism, in turn, is linked to unfavorable outcomes (Carver, Scheier, & Segerstrom, 2010). While research on optimism often focuses on the positive effects in these areas and their mediating mechanisms, such as coping strategies (Carver et al., 2010), experimental settings with a focus on performance and the pressure to succeed are largely neglected. However, pessimism might have positive consequences in achievement contexts. According to James (1890, p. 310f.), feelings of self-worth are determined by the ratio of actual success to pretensions. This means that low expectancies, a key feature of pessimism, might yield affective benefits: lower standards should lead to higher positive affect (e.g., pride) in the case of success and lower negative affect (e.g., disappointment) in the case of failure. The evidence related to this prediction is mixed so far. While studies on task expectancies find no support for benefits of a pessimistic outlook (Brown and Marshall, 2001 and Marshall and Brown, 2006), others show that optimism could indeed have affective costs (Sweeny & Shepperd, 2010). Moreover, people tend to prepare themselves for an (potentially unfavorable) outcome feedback by means of a “pessimistic shift” in expectations (Sweeny, Carroll, & Shepperd, 2006). The latter result suggests that pessimism may indeed improve affect in performance settings, although James’s assumption about outcomes was not tested directly. However, previous research found evidence that dispositional optimism can have a discernible effect on affective reactions in line with James’s predictions. In two experiments the authors ensured that participants succeeded at several tasks (Lau et al., 2014). They separated optimism from the outcome attribution (internal vs. external) and analyzed their effects on feelings of success (e.g., pride). Interestingly, James’s prediction about the benefits of pessimism was supported only when using an extreme groups sample (i.e., including participants from both ends of the optimism dimension): compared to optimists, pessimists experienced increased feelings of success, regardless of the attribution (Lau et al., 2014, Experiment 2). Attribution had no significant effect on affect, which implies that the effect of attribution depends on the trait characteristics of the sample under investigation. In summary, this study demonstrated that pessimism was associated with affective benefits, if the analyzed spectrum of optimism was wide enough. Furthermore, it was important to experimentally separate the effects of optimism and attributional style, which otherwise are probably confounded (see Marshall & Brown, 2006). Yet, there remained some open questions in Lau et al.’s study (2014). An emotional benefit of pessimism was found for the case of success and positive affect. However, James (1890) suggested that low expectations would also serve to protect from disappointments. Presumably, the effects of failure might be more important for well-being. Moreover, the role of attribution still seems ambiguous, as there was a main effect in Experiment 1 but not in Experiment 2. The authors suggested an interaction but could only observe a disordinal trend that did not reach significance. This may be due to a limitation of the study that saw unsuccessful participants dropping out of the experiment and hence the reduced statistical power (see Lau et al., 2014). The present study aims to build upon and extend this previous research, while addressing some of its limitations. The experimental design will be adapted and transferred to the case of performance failure as outcome, in order to test James’s (1890) hypothesis comprehensively. In line with James we predicted that pessimism would show its benefits by a decreased negative affective reaction to failure (i.e., less disappointment) and also a lesser decline in positive affect. Moreover, we sought to clarify the possibility of an interaction between optimism and attribution by preventing participants from dropping out and thereby increasing statistical power.

Conclusion The present research examined the differential impact of optimism and pessimism in the face of failure and found an interesting moderation by attribution. In terms of the relative affective impact, pessimists were better off than optimists, but only if the negative outcome was mainly ascribed to oneself. Thus, pessimism represents no general shielding against the woes of disappointment. Interestingly, optimists did benefit from a functional attributional style, that is, attributing a failure to external factors, whereas pessimists exhibited some kind of immunity towards the affective effects of attribution. In sum, we found James’s (1890) hypothesis to be correct but too simplistic. In comparison to optimists pessimists can experience affective benefits, however, the quality of which depends on the conditions of outcome and attribution. Our results imply that the extreme ends of optimism may be associated with significantly different processing styles in the domain of outcome evaluation. This is an interesting interaction of traits and cognition, and the identification of the mechanisms that are responsible for this has to be pursued in further studies.