People often set low goals in order to avoid future disappointment. This dissertation questions the assumption that future affect can be managed in this manner. This strategy can work only if performance is compared to the initially set goals. We argue that performance potential is instead spontaneously evoked at the time of performance feedback and used as the benchmark instead of goals. Even when goals are met, this comparison results in lower levels of satisfaction and greater disappointment when goals are set low vs. high. When performance falls short of initial goal, the default comparison appears to be to the initial goal, such that the resulting contrast exerts a downward pressure on satisfaction for both high and low goal setters. Contrary to people's assumption, however, satisfaction with achieving one's low goal was not any higher than falling short of one's high expectation, even when objectively, the outcome was superior in the former.

Five experiments mimicking real decisions made in the financial decision making domain provide support for this hypothesis. We manipulate levels of goal set by leading subjects to expect to make repeated decisions (Experiment 1), priming approach vs. avoidance motives (Experiments 2 and 4), or priming affect management concerns (Experiment 3). Participants set expected rates of return for their stock portfolio (i.e., goals) and perform a stock picking exercise. They then receive (false) feedback that the rate of return yielded by their portfolio has met their goal (Experiments l, 2 and 3) or has fallen short (Experiment 4). Finally in Experiment 5, we examine the possibility of whether this counterproductive tendency can be "corrected" by making people aware of this adverse consequence.

As expected, satisfaction is lower when goals are low vs. high even controlling for the absolute level of performance. Further, this effect does not hold when the goal is made salient at the time of performance feedback (Experiment 1). Experiment 3 demonstrates that counterfactual thoughts about effort that may be instigated by comparison of performance to potential (Experiment 2) are not necessary for the effect of goal level on satisfaction. In Experiment 4, we examine whether confirmed low goal is indeed higher in satisfaction than negatively disconfirmed high goal. Contrary to people's belief, results show that confirmed low goal leads to lower satisfaction compared to an outcome that falls short of one's expectation, and inferior in objective outcome.

This dissertation holds four main theoretical contributions. First and foremost, it shows that people's intuition about managing their future satisfaction by strategically lowering their expectation and goals at present is misguided, if not counterproductive. Second, it shows that setting low expectation exerts a negative influence on one's satisfaction with the outcome, even when the outcome meets the expectation. Third, satisfaction with confirmed low expectation exerts a negative influence on satisfaction even when compared to disconfirmed high expectation. Lastly, the proposed comparison mechanism, whereby expectation is not recruited as a benchmark to evaluate the outcome but rather a higher comparison standard is recruited instead, contributes to the satisfaction literature's conceptualization of multiple comparison standards.