The psychology of marathon running was studied by employing the cognitive strategies of association and dissociation (Morgan, 1978; Morgan & Pollock, 1977). Two shortcomings in the current literature were cited. These included the failure to study marathon runners in an actual race and the absence of an acceptable theory to explain the use of these strategies. In the present research, runners participating in a marathon were utilized and measures of dissociation, association, performance time, injury, and reasons for running a marathon were taken. The results indicated that motivations may have accounted for the use of cognitive strategies and that injury was not related to dissociation, as previously hypothesized. Additionally, runners overwhelmingly preferred to associate. A new theory regarding the use of these strategies was offered.