From Acute MDMA Experiences to Lasting Change: Self-Compassion’s Role in MDMA-Assisted Psychotherapy for Social Anxiety Disorder
Christina Chwyl, Mary K. Lear, Sunjeev K. Kamboj, Joanna J Arch, Adrianne R Wilson-Poe, Jason B. Luoma
October 25, 2026 preprint DOI: 10.31234/osf.io/3ard8_v1 (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Secondary analysis |
|---|---|
| Sample size | 20 |
| Population | Adults with social anxiety disorder |
| Interventions | 3 4-Methylenedioxymethamphetamine (MDMA)-assisted therapy |
| Duration | 16 weeks |
| Topics | Anxiety MDMA Psychedelic-assisted therapy |
| Registration | NCT05138068 |
| Key points | MDMA-assisted therapy was associated with acute state self-compassion during dosing and increases in trait self-compassion across 16 weeks, with state self-compassion explaining additional variance in reduced depressive symptoms and internalized shame. Trait increases after the second dose corresponded to less internalized shame and thwarted belonging. The authors propose self-compassion as a candidate mechanism of change. |
Abstract
Objectives: Self-compassion may benefit individuals with social anxiety disorder and serve as a mechanism of change within 3,4-Methylenedioxymethamphetamine (MDMA)-assisted therapy (MDMA-AT).
Methods: In this secondary analysis of 20 adults with social anxiety disorder who underwent MDMA-AT for social anxiety disorder (NCT05138068), we examined whether both trait self-compassion changes across treatment (16-weeks) and state self-compassion experienced during MDMA (midomafetamine) dosing sessions contributed to post-treatment clinical outcomes—social anxiety, depression, internalized shame, functional impairment and (thwarted) belonging—beyond baseline levels of each respective variable.Results: Participants experienced high self-compassion levels during the two dosing sessions. Trait self-compassion increased most substantially after the first MDMA dosing session (ΔM=0.74, p<.001), though continued to increase after dose 2 (ΔM=0.43, p< .001). State self-compassion was not associated with trait self-compassion changes, though explained additional variance in reduced depressive symptoms (ΔR²=.19, p<.001) and internalized shame (ΔR²=.28, p=.02), beyond baseline symptom severity. Trait self-compassion increases following the second dosing session corresponded to reduced internalized shame (β=-.84, p<.001) and thwarted belonging (β=-.68, p=.01), beyond baseline self-compassion levels. Greater baseline self-compassion corresponded to reduced social anxiety, functional impairment, internalized shame, and thwarted belonging (βs<-0.41, ps<.05).Conclusions: MDMA-AT for social anxiety disorder appears capable of fostering both acute self-compassion experiences and enduring trait-level changes, each of which were associated with certain beneficial outcomes, especially reduced internalized shame. These constructs represent promising candidate mechanisms of change within MDMA-AT for social anxiety disorder, while baseline self-compassion may moderate treatment response,warranting investigation of self-compassion’s role in future larger psychedelic-assisted therapy trials.
Comparable studies
Other observational and cohort studies on MDMA for anxiety, most cited first.