Socially warm experiences, when one feels connected to others, have been linked with physical warmth. Opioids, hypothesized to support social bonding with close others and, separately, physical warmth, may underlie both experiences. In order to test this hypothesis, 80 participants were randomly assigned to the opioid antagonist, naltrexone or placebo before neural and emotional responses to social and physical warmth were collected. Social and physical warmth led to similar increases in ventral striatum (VS) and middle-insula (MI) activity. Further, feelings of social connection were positively related to neural activity to social warmth. However, naltrexone ( vs placebo) disrupted these effects by (i) reducing VS and MI activity to social and physical warmth, (ii) erasing the subjective experience–brain association to social warmth and (iii) disrupting the neural overlap between social and physical warmth. Results provide additional support for the theory that social and physical warmth share neurobiological, opioid receptor-dependent mechanisms and suggest multiple routes by which social connections may be maintained.

Close social bonds are critical to immediate and long-term well-being. However, the neurochemical mechanisms by which we remain connected to our closest loved ones are not well understood. Opioids have long been theorized to contribute to social bonding via their actions on the brain. But feelings of social connection toward one’s own close others and direct comparisons of ventral striatum (VS) activity in response to close others and strangers, a neural correlate of social bonding, have not been explored. Therefore, the current clinical trial examined whether opioids causally affect neural and experiential signatures of social bonding. Eighty participants were administered naltrexone (n = 40), an opioid antagonist that blocks natural opioid processing, or placebo (n = 40) before completing a functional MRI scan where they viewed images of their close others and individuals they had not seen before (i.e., strangers). Feelings of social connection to the close others and physical symptoms commonly experienced when taking naltrexone were also collected. In support of hypotheses, naltrexone (vs. placebo) reduced feelings of social connection toward the close others (e.g., family, friends, romantic partners). Furthermore, naltrexone (vs. placebo) reduced left VS activity in response to images of the same close others, but did not alter left VS activity to strangers. Finally, the positive correlation between feelings of connection and VS activity to close others present in the placebo condition was erased by naltrexone. Effects remained after adjusting for physical symptoms. Together, results lend support to theories suggesting that opioids contribute to social bonding, especially with our closest loved ones.

Consoling touch is a powerful form of social support that has been repeatedly demonstrated to reduce the experience of physical pain. However, it remains unknown whether touch reduces emotional pain in the same way that it reduces physical pain. The present research sought to understand how handholding with a romantic partner shapes experiences of emotional pain and comfort during emotional recollection, as well as how it shapes lasting emotional pain associated with emotional experiences. Participants recalled emotionally painful memories or neutral memories with their partners, while holding their partner’s hand or holding a squeeze-ball. They additionally completed a follow-up survey to report how much emotional pain they associated with the emotional experiences after recalling them in the lab with their partners. Although consoling touch did not reduce emotional pain during the task, consoling touch increased feelings of comfort. Moreover, participants later recalled emotional memories that were paired with touch as being less emotionally painful than those that were not paired with touch. These findings suggest that touch does not decrease the immediate experience of emotional pain and may instead support adaptive processing of emotional experiences over time.

No abstract