MDMA does not alter responses to the Trier Social Stress Test in humans

MDMA (3,4-methylenedioxymethamphetamine) is a stimulant-psychedelic drug noted for prosocial effects such as increased empathy and social closeness. Earlier laboratory studies report that MDMA can reduce behavioural and neural reactivity to negative social stimuli (for example, threatening faces or simulated social rejection), yet the drug also produces physiological changes that resemble a stress response (increased cortisol, heart rate, blood pressure) and is commonly used in socially intense, physiologically challenging settings. These mixed findings leave unresolved whether acute MDMA reduces reactivity to an acute psychosocial stressor or instead augments physiological stress responses. Bershad and colleagues designed the present study to test whether single low doses of MDMA would dampen subjective and physiological responses to a standard psychosocial stressor, the Trier Social Stress Test (TSST), which involves an evaluative public speaking and arithmetic task. They hypothesised that MDMA would reduce negative subjective responses to the TSST (for example, ratings of stress or insecurity) while possibly increasing physiological measures such as cortisol and cardiovascular activity. The study focused on healthy young adults with light to moderate past MDMA experience and compared placebo, 0.5 mg/kg, and 1.0 mg/kg MDMA under double-blind conditions.

The study used a mixed design: within-subjects for task (stress vs no-stress) and between-subjects for drug (placebo, 0.5 mg/kg MDMA, 1.0 mg/kg MDMA). Thirty-nine healthy adults were randomised to one of the three drug groups (N = 13 per group) and completed two 4-hour laboratory sessions one week apart, receiving the same capsule (drug or placebo) on both visits. Sessions were counterbalanced so each participant experienced one TSST session and one non-stress control session. The TSST comprised 10 minutes preparation, a 5-minute job-speech in front of two interviewers and a camera, and an unexpected 5-minute mental arithmetic task. The control session involved talking about a favourite book or movie and a 5-minute game of solitaire. Participants were healthy volunteers aged 18–35 years, with light to moderate prior MDMA use (4–40 lifetime uses), BMI 19–26, at least high-school education, fluent in English, no regular medications and no DSM-IV Axis I disorder in the past year. Women were tested during the follicular phase (days 2–14) to reduce hormonal variability. Screening included psychiatric and medical assessments, ECG and drug/alcohol testing; prior TSST participation was exclusionary. MDMA hydrochloride and placebo capsules were prepared by the investigational pharmacy. Doses were weight-adjusted (0.5 and 1.0 mg/kg) and chosen on safety grounds. Participants took the capsule at 12:40 and completed the task 90 minutes later. Subjective measures included the Drug Effects Questionnaire (DEQ) and visual analogue scales (VAS) for stress, tension and insecurity, administered at baseline and at multiple post-dose time points (60, 85, 115, 145, 175 minutes). Pre-task appraisal and post-task ratings of threat, challenge and performance satisfaction were also obtained. Physiological measures were continuous heart rate (ambulatory ECG), periodic blood pressure, and salivary cortisol sampled at baseline, 60 minutes post-dose, and 10, 20 and 60 minutes after the task. Analyses used mixed ANOVAs: drug (between-subjects), task and time (within-subjects). Cortisol values were natural-log transformed; two outliers with baseline cortisol >2 SD from the mean were excluded. Missing data were deleted listwise. Greenhouse-Geisser correction was applied when sphericity was violated. Post hoc comparisons used Dunnett's tests. Effect sizes are reported as partial eta squared (ɳp2). Statistical significance was set at p < 0.05.

Sample characteristics were similar across groups: most participants were in their 20s (mean 24.1 ± 3.4 years), 44% identified as Caucasian, and the three drug groups did not differ on demographics or drug-use history. Subjective drug effects: the 1.0 mg/kg MDMA dose produced the expected acute subjective stimulant effects. Compared with placebo, 1.0 mg/kg significantly increased ratings of feeling high (drug × time, F(2,150) = 3.8, p < 0.001, ɳp2 = 0.20; peak at 85 min), liking the drug (drug × time, F(2,150) = 2.5, p < 0.01, ɳp2 = 0.14) and feeling the drug (drug × time, F(2,150) = 3.0, p < 0.01, ɳp2 = 0.17) across several post-dose time points. The 0.5 mg/kg dose did not produce these significant effects. Effects of the TSST: across all groups the TSST successfully induced psychosocial stress. Compared with the control task, the TSST increased VAS ratings of stress (task, F(1,35) = 30.5, p < 0.001, ɳp2 = 0.47), tension (F(1,35) = 17.4, p < 0.001, ɳp2 = 0.33) and insecurity (F(1,35) = 18.0, p < 0.001, ɳp2 = 0.34). On appraisal measures participants rated the TSST as more threatening and challenging, and reported lower self-efficacy and less satisfaction with their performance post-task. Physiologically, the TSST increased heart rate (task, F(1,34) = 20.4, p < 0.001, ɳp2 = 0.38), systolic blood pressure (task, F(1,35) = 6.5, p < 0.05, ɳp2 = 0.16), and cortisol (task, F(1,33) = 5.0, p < 0.01, ɳp2 = 0.13), particularly at 10 and 20 minutes post-task versus baseline. Effects of MDMA: contrary to the hypothesis that MDMA would dampen stress responses, MDMA increased several subjective and physiological measures independent of task. The 1.0 mg/kg dose raised ratings of stress (drug, F(2,35) = 3.6, p < 0.05, ɳp2 = 0.17), tension (drug, F(2,35) = 6.0, p < 0.01, ɳp2 = 0.25), and insecurity (drug, F(2,35) = 3.4, p < 0.05, ɳp2 = 0.16) during both the stress and no-stress sessions. Both MDMA doses increased how challenging participants found the tasks (drug, F(2,35) = 4.4, p < 0.05, ɳp2 = 0.20). Physiologically, the higher dose increased blood pressure overall (drug, F(2,35) = 7.4, p < 0.01, ɳp2 = 0.30), and both doses elevated salivary cortisol relative to placebo (drug, F(2,33) = 33.2, p < 0.001, ɳp2 = 0.67; post hoc p < 0.001 for 1.0 mg/kg, p < 0.05 for 0.5 mg/kg across 60–135 min). Importantly, there were no significant task × drug interactions for subjective effects or cardiovascular measures, indicating that MDMA did not moderate the additional response elicited by the TSST. In sum, the TSST produced the expected subjective and physiological stress responses, MDMA produced its expected subjective stimulant effects and increased cortisol and blood pressure, but it did not reduce reactivity to the psychosocial stressor.

Bershad and colleagues interpret their findings as failing to support the idea that single low doses of MDMA dampen acute psychosocial stress in healthy adults. Although prior work shows that MDMA can reduce behavioural and neural responses to negative social stimuli, in this controlled laboratory test MDMA did not attenuate either subjective or objective responses to the TSST. Instead the drug produced stress-like effects: increased ratings of stress, tension and insecurity, and increases in cortisol and blood pressure across both stress and non-stress sessions. The authors discuss several possible reasons for the discrepancy between these results and prior reports of MDMA’s prosocial or stress-dampening effects. Dose may be critical: MDMA increases plasma oxytocin (a hormone linked to prosocial effects) only at higher doses in some studies, and therapeutic MDMA doses used in PTSD trials are higher than those tested here. Context and task characteristics could also matter; the TSST is a particular form of evaluative social stress that may not tap the same processes as tasks showing MDMA-related reductions in negative social perception or as a psychotherapeutic setting in which MDMA has shown promise. Individual differences and sample size are additional concerns: the present sample was modest and not powered to examine sex differences or other moderators. The authors acknowledge limitations including the relatively low doses tested, the small sample size that limits subgroup analyses, and the possibility that the TSST may not be the optimal paradigm to reveal MDMA’s unique social effects. They also note that laboratory settings differ from naturalistic or psychotherapeutic environments where MDMA might have different effects. Finally, Bershad and colleagues highlight that MDMA’s effects are complex and may dissociate anxiety or arousal from changes in social behaviour or memory processing—for example, MDMA can increase anxiety while simultaneously making it easier to discuss emotional memories. They conclude that the conditions under which MDMA reduces responses to negative social stimuli—such as dose, context, participant characteristics, or outcome measures—remain to be determined.