Designing health messages that successfully elicit message-consistent behavior continues to be a challenge, in large part because people are often poor predictors of their future actions. Past work that aimed to improve our predictive abilities has suggested that activation in a ventral subregion of medial prefrontal cortex (MPFC) during receipt of a persuasive message can reliably predict downstream behavior (Falk et al., 2010; Falk et al., 2011; Falk, Berkman, & Lieberman, 2012; Falk et al., 2015; Cooper, Tompson, O’Donnell, & Falk, in press), that gain-framed messages are more effective in promoting prevention behaviors than loss-framed messages (Detweiler, Bedel, Salovey, Pronin, & Rothman, 1999; Rothman et al., 1993; Christophersen & Gyulay, 1981; Robberson & Rogers, 1988; Treiber, 1986), and that persuasive messages that contain action plans are more effective than those that do not (Gollwitzer, 1993; Gollwitzer & Brandstätter, 1997; Gollwitzer & Sheeran, 2006; Hagger & Luszczynska, 2014; Sniehotta, Scholz, & Schwarzer, 2006); however, the psychological mechanisms that support these effects are not fully understood. In the current experiment, we argue that persuasion may occur via self-integration—the incorporation of persuasive messages into one’s self-concept and identity—by connecting these bodies of literature in our study design. Pulling from the topics of framing effects and action planning in health psychology, along with neuroscience work in persuasion and action understanding, we exposed participants to four types of messages about sunscreen use in the fMRI scanner while also tracking their sunscreen use behaviors and intentions: 1) Fact – facts about sunscreen (control), 2) How – how to wear sunscreen, 3) Whygain – why one should wear sunscreen to garner a benefit, and 4) Whyloss – why one should wear sunscreen to avoid a loss. We replicated past findings on the relationship between MPFC activity during message exposure and future behavior controlling for intentions, along with past action understanding work on the role of rostral inferior parietal lobule (rIPL) and posterior inferior frontal gyrus (pIFG) in response to How>Why and conversely MPFC in response to Why>How; however, we also found preliminary support for our theory of persuasion as a self-integration process by focusing on different message types individually. We found greater MPFC activity during gain-framed messages relative to loss-framed messages, raising the possibility that gain frames tend to be more effective than loss frames for prevention behaviors because they lead individuals to consider the personal positive value of the behavior, which may support the integration of the behavior into one’s self-concept. We also found stronger correlations between MPFC activity and future behavior for participants who were not preexisting sunscreen users than for those that were, potentially suggesting that non-users may have more room for self-integration to facilitate behavioral choices (whereas users already consider sunscreen use part of their self-concept). Finally, the fact that activity in both MPFC and rIPL was related to message-consistent behavior suggests that both personal valuation and cognitive rehearsal may contribute to self-integration during message encoding and may support downstream behavioral choices. Both theoretical and practical implications of these findings are discussed along with future directions.