The present investigation was designed to test the predictions of a control theory based view of self-regulation on the complex effects of self-efficacy on persistence (Vancouver, 2008). In particular, self-efficacy was expected to positively affect performance in a context where performance feedback never indicated goal achievement. In such a context, goal choice (i.e., whether to abandon the goal or not) is relevant and high self-efficacy individuals were expected to hold out longer. In contrast, when no performance feedback is available, those with high self-efficacy are more likely to believe they achieved the goal more readily, and thus persist less, than those with low self-efficacy. Participants (87 females and 67 males) were manipulated to have either high or low self-efficacy for a task involving hand grip strength by using a confederate who claimed to have overworked their hand (high manipulated self-efficacy) or to be a tennis player (low manipulated self- efficacy). Feedback ambiguity was manipulated by the presence or absence of a physical barrier between the participant and the confederate as they performed. Performance was assessed by length of time an individual could grip at 50% of their maximum grip strength. The results supported the hypothesized interaction of self-efficacy and feedback condition. Moreover, self-efficacy was found to negatively relate to self-reported effort regardless of feedback condition.