Testing the Role of Narrative Modeling on Self-Compassion

Abstract: The present research examines the effects of modeling self-compassion through a narrative. Three experimental studies ( N 1 = 168, N 2 = 219, N 3 = 281) varied narrative self-compassion (high, low). Narrative self-compassion was manipulated through the reactions of the main character to mistakes and misfortunes. In the high self-compassion condition, the character’s response was forgiving and self-reassuring, emphasizing the main aspects of self-compassionate behaviors; in the low-self-compassion condition, the character’s response was self-critical. Contrary to expectations, there was no significant effect of story condition on self-compassionate outcomes in Studies 1 and 2. In Study 3, exposure to the low self-compassion story condition resulted in a significant decrease in self-compassion levels reported by participants. Transportation into the narrative and identification with the main character were also found to be higher in the low-self compassion story condition (Study 1), and to have a significant negative effect on self-compassion outcomes (Study 2). These findings suggest that using narrative modeling to evoke self-compassionate outcomes is ineffective, at least in a cross-sectional design. It is possible that participants did not have enough time to manage the negative emotions that may have been sparked by the narrative and grapple with long-held beliefs immediately postexposure; future research should examine longitudinal effects of self-compassion narratives.