A theatrically based intervention was given to 122 older adults who took lessons twice a week for 4 weeks. The training consisted of multi-modal activities (cognitive-affective-physiological) typically employed in college acting classes. Comparison groups consisted of no-treatment controls and participants instructed in a different performing art, singing. Assessment of effectiveness was performed using a battery of 11 cognitive/affective test measures that included word recall, prose comprehension/recall, word generation, digit-span ability, and problem-solving. It was found that the acting group improved significantly from pretest to posttest over both other groups. Digit-span was the only measure that failed to improve. The gains were achieved despite the fact that no aspects of the intervention supplied specific training or practice on the test measures. Previous versions of the intervention with community-dwelling adults had produced similar findings but the current participants were older, less well-educated, and lived in subsidized, primarily low-income, retirement homes.