Physiological response to self-compassion versus relaxation in a clinical population

Compassion-focused imagery (CFI) can be an effective emotion-regulation technique but can create threat-focused responses in some individuals. However, these findings have been based on tasks involving receiving compassion from others. This study sought to compare responses CFI involving self-compassion to relaxation and a control task, and to see whether any threat-responses to self-compassion and relaxation decrease with practice. 25 participants with depression/anxiety symptoms and high self-criticism and/or low self-compassion engaged in three tasks (control task, relaxation imagery, and CFI) at three or four separate testing sessions, every three days. Heart-rate variability (HRV) was used to explore group-level differences between tasks. Additionally, we identified how many individuals showed a clinically significant change in HRV in response to compassion (compared to baseline) and how many showed such a change during relaxation (compared to baseline). During session 1, more individuals had a clinically significant increase in HRV in response to CFI (56%) than in response to relaxation (44%), and fewer had a clinically significant decrease in HRV during CFI (16%) than during relaxation (28%). Comparing the group as a whole, no significant differences between tasks were seen. Repeated sessions led to fewer positive responses to CFI, perhaps reflecting habituation/boredom. These preliminary findings suggest that in high self-critics (those most likely to find self-compassion difficult), self-compassionate imagery is no more challenging than standard relaxation tasks. For both compassion and relaxation, some individuals respond positively and others negatively. For those who are not benefiting, practice alone is not sufficient to improve response. Effects may differ for other compassion tasks. Trial number: NCT04647318.