MDMA alters emotional processing and facilitates positive social interaction

MDMA (3,4-methylenedioxymethamphetamine, “ecstasy”) produces characteristic stimulant effects together with distinctive ‘‘prosocial’’ effects such as increased feelings of closeness, empathy and desire to socialise. Earlier research has shown that MDMA can reduce perception of negative facial expressions and reduce amygdala responses to anger, and some studies suggest it may enhance recognition or neural responses to positive expressions. However, previous work has focused mainly on perception of static stimuli, leaving uncertain whether MDMA also alters emotional responses after perception and whether these effects translate into measurable changes in real social behaviour. Wardle and colleagues set out to characterise how MDMA affects both perception of and psychophysiological responses to emotional facial expressions using dynamic, full‑colour video stimuli, and to test whether it changes behaviour and subjective perceptions during a controlled face‑to‑face social interaction. They also probed, on a preliminary basis, whether any prosocial effects related to self‑reported desire to take the drug again as an index of abuse liability.

The study used a within‑subject, double‑blind design in which 36 healthy occasional MDMA users attended three laboratory sessions and received placebo, 0.75 mg/kg MDMA and 1.5 mg/kg MDMA in counterbalanced order. Sessions were at least 7 days apart (mean 26 days). Participants were 18 males and 18 females aged 18–35, screened medically and psychiatrically; inclusion criteria required prior ecstasy use (4–40 occasions) and absence of recent DSM‑IV Axis I disorders or stimulant dependence. Women not on hormonal contraception were tested in the follicular phase. Compliance with abstinence instructions was verified by breath and urine tests. Sessions ran in a living‑room style laboratory from 09:00 to 14:00. Baseline measures were collected, then participants swallowed an opaque capsule containing MDMA powder or placebo. Primary assessments included mood (Profile of Mood States, POMS; Elation and Arousal subscales), cardiovascular measures (mean arterial pressure, MAP), and a visual analogue scale (VAS) for social emotions (13 adjectives, with Playful and Loving highlighted). A Dynamic Emotional Identification Task (DEIT) presented 40 video sequences (10 actors × 4 emotions) that progressed from neutral to full expression; participants pressed a key when they identified the expression. Perception was quantified as the percent intensity at identification, and accuracy was high (mean 93%, SD=4). Emotional responses were measured with surface electromyography (EMG) of corrugator (frown) and zygomatic (smile) muscles during the final second of each correctly identified display, baseline‑corrected. To assess real‑time social behaviour, participants completed a 5‑minute semi‑structured interaction with a trained research assistant about an important person in their life; speech was transcribed and analysed using Linguistic Inquiry and Word Count (LIWC) to quantify positive and negative emotion words. Immediately after the interaction participants rated the assistant on regard, empathy and congruence using a brief Barrett‑Lennard inventory. Desire to take the drug again was measured at session end by a single VAS (0–100). Statistical analysis used linear mixed‑effects models (LME) implemented in R, with dose as a fixed factor and subject (and actor where relevant) as random effects. For repeated time courses (POMS, MAP, VAS) an area under the curve (AUC) relative to baseline was computed; missing time points (<8 total) were imputed by adjacent means. Emotion contrasts in DEIT/EMG used Helmert coding (happy vs negative, anger vs other negative, fear vs sadness). EMG models included display intensity at identification as a covariate. Exploratory per‑participant linear regressions estimated individual drug effects on social and stimulant outcomes, and these estimates were tested as predictors of desire to take the drug again.

Mood and cardiovascular: Both doses of MDMA increased POMS Elation and Arousal relative to placebo. The linear drug effect on Elation was B=422.5 (SE=159.3), t(35)=2.65, p=0.01; for Arousal it was B=1,636.0 (SE=404.9), t(35)=4.04, p<0.001. The arousal effect was slightly stronger in women, though present in both sexes. The extracted text does not clearly report numerical results for MAP in the Results section. Emotional perception (DEIT): Four participants were excluded from DEIT analyses because they deferred identification until the end of sequences, leaving n=32. A session effect showed faster identification across repeated testing, so session was included as a covariate. At the 1.5 mg/kg dose MDMA increased the intensity required to identify angry expressions compared with other negative emotions (linear drug × anger vs other negatives: B=2.83, SE=1.16, t(31)=2.44, p=0.03). Anger was nonetheless identified at lower intensity than the other negative emotions overall (B=−5.03, SE=0.49, t(31)=10.36, p<0.001). No drug effects were observed on identification of fear, sadness or happiness. Psychophysiological responses (EMG): EMG data were lost or excluded for some participants (five lost for jaw clenching, one technical loss for corrugator, one zygomatic outlier), leaving n=30 for each EMG analysis. Corrugator activity decreased across sessions, so session was covaried. At 1.5 mg/kg MDMA corrugator activity was reduced to happy faces relative to negative faces, indicating more positive facial muscle responses to happiness on drug (linear drug × emotion interaction: B=−0.76, SE=0.35, t(29)=2.20, p=0.04). This corrugator effect was moderated by participant sex and was primarily evident in female participants (sex × linear drug × emotion: B=1.92, SE=0.69, t(29)=2.78, p=0.009). Zygomatic activity increased to happy versus negative expressions at 1.5 mg/kg (linear drug × emotion contrast: B=0.82, SE=0.32, t(29)=2.58, p=0.02); this effect did not vary by sex. Behaviour and perceptions in social interaction: One participant had incomplete speech data (n=35 for word counts). MDMA at both doses increased percentage use of positive emotion words without affecting negative words (drug × word type interaction: B=0.66, SE=0.32, t(34)=2.08, p=0.05). Ratings of perceived regard showed a marginal linear drug effect (B=0.18, SE=0.10, t(35)=1.82, p=0.08) and empathy increased slightly but significantly (B=0.18, SE=0.09, t(34)=2.06, p=0.05). Follow‑up paired tests indicated a significant difference between 1.5 mg/kg and placebo for regard (t[35]=2.04, p=0.05) and a marginal effect on empathy (t[35]=1.91, p not fully reported). There was no significant effect on congruence. Relationship to desire to take the drug again: MDMA robustly increased self‑reported desire to take the drug again (linear drug effect: B=28.88, SE=6.81, t(35)=4.24, p<0.001). In exploratory analyses, only VAS Playful and POMS Elation individually predicted desire to take the drug again; when both were entered together the overall model was significant (F(2,33)=3.80, p=0.03) but individual coefficients were not, suggesting overlapping explanatory power.

Wardle and colleagues interpret their findings as evidence that MDMA alters both the perception and the downstream emotional response to others' facial expressions in ways that could promote social interaction. Specifically, the drug slowed identification of angry expressions and enhanced positive psychophysiological responses to happy faces, the latter effect being stronger in women. In a real‑time social task MDMA increased use of positive emotional language and modestly increased perceptions that the interaction partner was empathic and caring. The authors place these results in the context of prior work showing reduced amygdala responsiveness to anger and mixed findings for recognition of positive expressions; they note that use of dynamic, full‑colour stimuli and EMG adds sensitivity compared with earlier studies using static images. Comparisons with other stimulant and serotonergic drugs suggest a distinct profile for MDMA: unlike amphetamine or methylphenidate, which can facilitate detection of emotions more generally, MDMA selectively slowed detection of negative emotion while amplifying positive responses. Likewise, acute SSRI effects differ in that they tend to increase perception and responses to both positive and negative stimuli, whereas MDMA reduced negative perception while amplifying positive responding. Pharmacological differences, including MDMA's serotonergic action alongside other effects, are offered as possible explanations. Several limitations are acknowledged. The sample was relatively small, homogeneous and screened to exclude psychiatric disorders and heavy substance use, which may limit generalisability and reduce sensitivity to abuse‑related effects. All participants had prior MDMA exposure, so effects in drug‑naïve individuals (for example in therapeutic contexts) remain uncertain. The laboratory environment and brief, positively framed interaction may not fully capture the social contexts in which MDMA is used or therapeutic processes occur. Finally, the single‑item measure of desire to take the drug again provides only a crude proxy of abuse liability; more detailed behavioural choice or valuation measures might yield different conclusions. In the authors' view, the pattern of effects—reduced sensitivity to negative social cues combined with enhanced positive responding and more positive social behaviour—may both help explain recreational appeal and plausibly contribute to MDMA's utility as an adjunct to psychotherapy, while not appearing tightly coupled to self‑reported desire for future use in this screened sample.