A four-week online course of loving-kindness meditation did not outperform progressive muscle relaxation in reducing physical pain or boosting social well-being among healthy adults, a randomised trial has found. Participants who practised muscle relaxation reported a small rise in perceived social support, while those practising loving-kindness reported no such improvement; neither method showed a clear advantage on pain measures. Those whose sense of social support grew during the study tended to endure a painful task longer at follow-up, though the association was modest.
Loving-kindness meditation is designed to cultivate warmth and compassion towards oneself and others, making its social aim explicit. Unlike mindfulness, which centres on neutral awareness of present sensations, loving-kindness invites practitioners to extend wishes for safety, health, happiness and peace first to themselves and then to others.
Historical hints suggested that this approach might ease discomfort, with early trials in chronic back pain reporting reductions in daily pain and distress. Biological work has also proposed that social bonding and pain relief share brain pathways, with warm social interactions linked to natural pain-relieving neurochemicals.
Georgetown University researchers Olivia Young and Katherine O’Connell, alongside colleagues, sought to test whether loving-kindness could increase pain tolerance by strengthening social well-being. Their work built on a 2016 finding that longer tolerance in an uncomfortable posture correlated with larger social networks.
The two-arm trial ran between December 2020 and July 2021, during the early stages of the COVID-19 pandemic. The study was registered on ClinicalTrials.gov and preregistered on the Open Science Framework, with data and materials posted publicly. Seventy-five healthy U.S. adults aged 18 to 55, with no prior meditation experience, chronic pain, mental health diagnoses or regular pain medication use, were recruited; six participants dropped out, leaving 69 completing the study (mean age 31; 64 per cent women).
Testing occurred remotely via video calls and online surveys due to pandemic restrictions. In a baseline double-blind session, each participant performed a wall-sit to gauge acute muscular pain, indicating pain threshold and tolerance and rating pain on a 0 to 100 scale.
Participants completed baseline questionnaires on perceived social support, loneliness and trait anxiety. For five days before and after the training, diaries tracked daily helping behaviours and time spent socialising, with county-level COVID metrics and self-reported isolation and social distancing also monitored.
Subjects were randomised to four weeks of loving-kindness meditation (34 participants) or progressive muscle relaxation (35 participants), with six 15-minute online sessions weekly for a total of 24 sessions. The loving-kindness audio guided gradual expansion of wishes from oneself to others, while the PMR audios directed sequential relaxation of muscle groups, without social content. A follow-up session about five weeks later repeated the wall sit and surveys.
Attendance was high in both arms, with about 88 per cent of sessions completed on average. Contrary to predictions, loving-kindness did not outperform PMR on any measure of social well-being. Perceived social support rose slightly after PMR (28.4 to 29.2 on a 36-point scale) and dipped after loving-kindness (28.5 to 27.4), a between-group difference that reached statistical significance.
“Our ‘control’ meditation, PMR, beat LKM on improving social support,” O’Connell told PsyPost. “And participants loved the PMR recordings, which are freely available.”
Loneliness declined after muscle relaxation but did not change after loving-kindness training, a between-group difference that fell just short of statistical significance. Trait anxiety declined somewhat in both groups, more after PMR, but the difference between programmes was not statistically significant. Changes in emotional isolation, daily helping and time spent socialising were also not significant.
Neither intervention clearly outperformed the other on pain. Wall-sit times increased by around eight seconds on average after PMR and were largely unchanged after loving-kindness, with the between-group difference not statistically significant. Pain ratings during the wall sit did not change meaningfully over time.
At baseline, greater perceived social support correlated with longer wall-sit endurance, echoing prior work. An exploratory analysis suggested that participants whose social support grew over the four weeks tended to show larger gains in wall-sit time, though the association was modest and not statistically significant within either group, and in the PMR group it neared zero.
On the size of the observed effects, O’Connell described them as small but worth pursuing. “We found some evidence for a link between social well-being and pain; people whose sense of social support grew during the study also increased their pain tolerance, whichever meditation they did.”
The findings align with a separate 2026 randomised trial from the same Georgetown laboratory that examined loneliness and empathy and reached a similar conclusion—loving-kindness did not outperform progressive muscle relaxation on these measures.
Conversely, earlier work involving older adults and a longer, more intensive loving-kindness and compassion programme reported increased feelings of connection, creating tension with the current, four-week online study conducted in a pandemic-era general adult sample.
The researchers suggest that the mixed results across studies may reflect methodological weaknesses in earlier work, such as the absence of an active comparator and susceptibility to expectancy effects. A 2018 systematic review and meta-analysis similarly found that compassion gains were more likely when studies used passive waiting-list controls or when the meditation teacher was a co-author.
A key caveat is that this trial had no passive, no-treatment control group, making it difficult to determine whether observed changes were due to the training or other factors, such as evolving social conditions during vaccine rollouts in 2021.
O’Connell noted the unique circumstances of conducting the study during lockdowns and suggested that LKM might spark a desire for connection that participants could not act on. “We also had no ‘do nothing’ group, so it could be that both meditations actually helped to some extent compared to people not doing any meditation during this socially challenging time.”
The sample consisted of young, predominantly female, meditation-naive adults without chronic pain, and the wall-sit test was administered via video conference, an approach described as a practical innovation with a strong baseline link between pain threshold and tolerance. The researchers acknowledge that remote pain testing should be validated against laboratory methods such as heat or cold-water immersion and captures only one form of muscular discomfort.
Nevertheless, O’Connell emphasised the potential of remote pain testing to broaden inclusivity in future work. “Lots of next steps, but one to highlight is that we showed you can run a pain test over Zoom and replicate prior research,” she said. “I think it should help make pain research accessible to more people (e.g., in rural areas).”
The study, Assessing the effect of Loving-Kindness Meditation on pain and social well-being using a randomized controlled trial design, was led by Olivia Young, Katherine O’Connell, Naomi Nero, Melinda C. Somers, Paige A. Freeburg, Mary Ann Dutton, and Abigail A. Marsh.
