Brain training is increasingly popular, and many believe in the efficacy of such programs without empirical evidence. We examined whether instructions promising memory improvement would influence subjective and objective cognition. Participants (n = 145; age: M = 50.64) were randomly assigned to a memory improvement or memory task condition. Participants completed demographic and perceived control over cognition measures, in addition to cognitive tasks and subjective cognition items for 7 days. Participants in the improvement condition reported significantly greater memory increases than those in the memory task condition. This effect was moderated by perceived control over cognition; participants in the improvement condition with high control beliefs were significantly more likely than those with low control beliefs or those in the task condition to report better memory. Individuals with higher control beliefs may be more susceptible to the claims of brain training programs, which is significant given that such programs are increasingly popular and commercially available.