Posterior medial frontal cortex regulates sympathy: A TMS study

ABSTRACT

Harm to some elicits greater sympathy than harm to others. Here, we examine the role of posterior medial frontal cortex (PMFC) in regulating sympathy, and explore the potential role of PMFC in the related phenomena of mentalizing and representing others as connected with oneself. We down-regulated either PMFC or a control region (middle temporal visual area), then assessed feelings of sympathy for and self-other overlap with two characters described as having suffered physical harm, and who were framed as adversarial or affiliative, respectively. We also measured mentalizing performance with regard to inferring the cognitive and affective states of the adversarial character. As hypothesized, down-regulating PMFC increased sympathy for both characters. Whereas we had predicted that down-regulating PMFC would decrease mentalizing ability given the postulated role of PMFC in the mentalizing network, participants in the PMFC down-regulation condition evinced greater second-order cognitive inference ability relative to controls. We observed no effect of the TMS manipulation on self-other overlap, although sympathy and self-other overlap were positively correlated. These findings are discussed as they may inform understanding of the functional role(s) of PMFC in regulating responses broadly linked with empathy.

KEYWORDS:

• Empathy
• sympathy
• theory of mind
• self-other overlap
• transcranial magnetic stimulation
• pre-supplementary motor area
• rostral cingulate zone
• posterior medial frontal cortex

Acknowledgments

We thank our research assistants Rachel Bryan, Priscilla Chan, Alison Crosby, Akeiylah DeWitt, Regino Fronda, Allan Garcia, Carlos Garcia-Gomez, Ryan Graser, Xianzhi Li, Elizabeth Marquez, and Antoinette Verrier.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed here

Notes

¹ In addition to the present research on sympathy, mentalizing, and self-other overlap, the study session also included measures of religiosity and group bias submitted for separate publication, as they involve distinct theoretical considerations (Holbrook et al., 2020). Those facets of the study session included a between-subjects manipulation involving writing about a threatening versus neutral topic. Follow-up analyses confirm that this task had no significant effects on any of the outcome measures reported here, and controlling for the writing manipulation does not alter the pattern or significance of any of the present findings. Accordingly, the writing task is not discussed further. (The full materials are available at https://osf.io/ycjt8/)

² A computer error was discovered during data collection; 16 participants had inadvertently skipped one or more blocks. We corrected the issue and ran 16 further participants, producing a final sample just short of our pre-registered target of 100. However, the window for recruitment had closed and the total of 96 (before dropping one participant for overly fast response latencies) was deemed sufficient.

³ 6.3% of the sample reported no feeling of personal identification, 75.8% reported moderate identification, and 17.9% reported extreme identification with the university. Exploratory follow-up analyses confirm that the observed pattern of effects of the TMS manipulation on sympathy for both authors holds, and in fact grows more significant (ps ≤ .01, η_p²s > .07), when excluding those participants who reported no personal identification with the university.