OBJECTIVE: To examine how positive affect influences ability to benefit from heart rate variability (HRV) biofeedback treatment for individuals with severe brain injury. METHOD: Secondary data analysis of a nonrandomized experimental study that assessed the efficacy of biofeedback treatment for executive dysfunction in 13 individuals with chronic severe brain injury. RESULTS: Bivariate correlations between the predictors (levels of HRV and positive affect) and the outcome (change in Category Test errors) showed large effect sizes for higher levels of HRV coherence (r = -.495, p = .085) but not for positive affect (r = .069, p = .824). Although positive affect had a negligible effect on Category Test improvements by itself, positive affect played a moderating role that complemented the effect of HRV coherence. HRV coherence had a stronger effect on Category Test performance among those participants who demonstrated higher positive affect. A regression model was fit that included main effects for HRV coherence and positive affect, as well as their interaction. The interaction term was significant in a 1-tailed test (b = -3.902, SE = 1.914, p = .072). CONCLUSIONS: Participants who had the most positive emotions made the most gains in the HRV biofeedback training and performed better posttreatment on a test designed to measure problem-solving ability. Results indicate that positive affect can improve cognition, specifically mental flexibility and abstract thinking. Addressing factors that shape negative affect such as irrational beliefs and self-doubt is an important target for therapeutic intervention even in those with severe, chronic deficits. (PsycINFO Database Record