The primary goal of this study was to determine whether exercise-associated improvements of the immune response to influenza vaccination were mediated by improvements in psychosocial factors in older adults. At baseline, prior to the exercise intervention, older adult participants were immunized with influenza vaccine. Blood samples collected pre-immunization, 1, 4, and 12 weeks post-immunization were analyzed for anti-influenza antibody, whereas influenza-specific cytokine (IFNgamma) was evaluated at 1 week post-immunization. Depression and sense of coherence were measured pre-immunization. Four weeks post-immunization, participants were randomly assigned to either an aerobic exercise group (n=14) or a control group (n=14). After a 10-month exercise intervention, the immunization, blood collections, and psychosocial measures were repeated. At the post-intervention evaluation, exercise participants had improved scores on depression and sense of coherence. Also post-intervention, exercise participants had a greater increase in antibody and IFNgamma production. After controlling for the effect of both psychosocial measures, the exercise treatment remained significant with respect to antibody titer suggesting that the increases in antibody were not mediated by improvement in the psychosocial factors. In contrast, the enhancement of IFNgamma appeared to be mediated at least in part by the psychosocial factors. After controlling for psychosocial factors, exercise treatment was no longer significantly related to the change in IFNgamma. Taken together, our findings may suggest that the mechanism(s) of exercise-induced improvement in immunocompetence involve both physiological and psychological pathways.