Therapists often assume a special association between mental imagery and emotion, though empirical evidence has been lacking. Using an interpretation training paradigm, we previously found that imagery had a greater impact on anxiety than did verbal processing of the same material (Holmes & Mathews, 2005). Although the finding of a differential impact of imagery versus verbal processing of negative material was replicated, findings did not extend to benign material. Results therefore left open the question of whether there may be a special association between imagery and positive emotion. The current experiment examined positive interpretation training. Numerous scenarios were presented with initial ambiguity as to positive outcome or not, with final information then yielding consistently positive resolutions. Participants were asked to either imagine these positive events or to listen to the same descriptions while thinking about their verbal meaning. Those participants in the imagery condition reported greater increases in positive affect and rated new descriptions as being more positive than did those in the verbal condition. Results suggest that positive training can be enhanced through imagery as opposed to verbal processing. This study also provides the first test of a standardized intervention using an "interpretive bias training" paradigm to improve positive mood.