Effects of compassion training on brain responses to suffering others.

Compassion meditation (CM) is a promising intervention for enhancing compassion, although its active ingredients and neurobiological mechanisms are not well-understood. To investigate these, we conducted a three-armed placebo-controlled randomized trial (N = 57) with longitudinal functional magnetic resonance imaging (fMRI). We compared a 4-week CM program delivered by smartphone application with (i) a placebo condition, presented to participants as the compassion-enhancing hormone oxytocin, and (ii) a condition designed to control for increased familiarity with suffering others, an element of CM which may promote compassion. At pre- and post-intervention, participants listened to compassion-eliciting narratives describing suffering others during fMRI. CM increased brain responses to suffering others in the medial orbitofrontal cortex (mOFC) relative to the familiarity condition, p < 0.05 family-wise error rate corrected. Among CM participants, individual differences in increased mOFC responses positively correlated with increased compassion-related feelings and attributions, r = 0.50, p = 0.04. Relative to placebo, the CM group exhibited a similar increase in mOFC activity at an uncorrected threshold of P < 0.001 and 10 contiguous voxels. We conclude that the mOFC, a region closely related to affiliative affect and motivation, is an important brain mechanism of CM. Effects of CM on mOFC function are not explained by familiarity effects and are partly explained by placebo effects.

Effects of compassion meditation on a psychological model of charitable donation.

Compassion is critical for societal wellbeing. Yet, it remains unclear how specific thoughts and feelings motivate compassionate behavior, and we lack a scientific understanding of how to effectively cultivate compassion. Here, we conducted 2 studies designed to a) develop a psychological model predicting compassionate behavior, and b) test this model as a mediator of a Compassion Meditation (CM) intervention and identify the "active ingredients" of CM. In Study 1, we developed a model predicting compassionate behavior, operationalized as real-money charitable donation, from a linear combination of self-reported tenderness, personal distress, perceived blamelessness, and perceived instrumental value of helping with high cross-validated accuracy, r = .67, p < .0001. Perceived similarity to suffering others did not predict charitable donation when controlling for other feelings and attributions. In Study 2, a randomized controlled trial, we tested the Study 1 model as a mediator of CM and investigated active ingredients. We compared a smartphone-based CM program to 2 conditions-placebo oxytocin and a Familiarity intervention-to control for expectancy effects, demand characteristics, and familiarity effects. Relative to control conditions, CM increased charitable donations, and changes in the Study 1 model of feelings and attributions mediated this effect (pab = .002). The Familiarity intervention led to decreases in primary outcomes, while placebo oxytocin had no significant effects on primary outcomes. Overall, this work contributes a quantitative model of compassionate behavior, and informs our understanding of the change processes and intervention components of CM. (PsycINFO Database Record