OBJECTIVES: We evaluated the effect of cognitive training among 1,534 participants in the Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) randomized controlled trial (RCT) on 5-year improvements in 3 cognitive-specific measures of locus of control-internal, chance, and powerful others. METHODS: ACTIVE was a multisite RCT (age > or = 65), with 4 groups (memory, reasoning, speed of processing, and no-contact control). Complete 5-year follow-up data were available for 1,534 (55%) of the 2,802 participants. A propensity score model was used to adjust for potential attrition bias. Clinically important improvements (and decrements) in the cognitive-specific locus of control scale scores were defined as greater than or equal to 0.5 SD (medium) and greater than or equal to 1.0 SD (large). Multinomial logistic regression was used to simultaneously contrast those who improved and those who declined with those whose locus of control scale score was unchanged. RESULTS: Statistically significant effects reflecting medium-sized (> or = 0.5 SD) improvements in internal locus of control between baseline and the 5-year follow-up were found for the reasoning and speed of processing intervention groups who were 76% (p < .01) and 68% (p < .05) more likely, respectively, to improve than the no-contact control group. No improvement effects were found on the chance or powerful others locus of control measures or for the memory intervention group. CONCLUSION: Cognitive training that targets reasoning and speed of processing can improve the cognitive-specific sense of personal control over one's life in older adults.