MDMA enhances emotional empathy and prosocial behavior

Thirty-two healthy volunteers (16 men, 16 women; mean age 25 ± 3 years, range 20–31) were recruited. Exclusion criteria included personal or first-degree family history of psychiatric disorder, significant physical illness, current smoking, and substantial prior illicit-drug use; participants with recent illicit use were screened by repeated urine tests. Most participants were MDMA-naïve (n = 22); ten had fewer than five prior MDMA experiences. Female participants were tested during the follicular phase (days 2–14) to reduce menstrual-cycle effects. A double-blind, placebo-controlled, random-order cross-over design was used: each subject received MDMA (125 mg oral) and placebo in separate sessions, producing 64 total assessments. The washout period was at least 10 days. MDMA capsules contained 125 mg with mannitol filler; identical placebo capsules contained mannitol only. The mean administered dose corresponded to 1.89 ± 0.30 mg/kg body weight. Behavioural and subjective measures comprised: visual analogue scales (VAS) and the Adjective Mood Rating Scale (AMRS) for subjective prosocial mood; the Multifaceted Empathy Test (MET) to assess cognitive empathy (percentage correct in inferring mental states) and emotional empathy (explicit ratings of felt concern and implicit arousal ratings on 1–9 scales), using 40 photographs (20 positive, 20 negative) and 120 trials total; a paper-based Social Value Orientation (SVO) slider task to measure prosocial allocation preferences and to distinguish inequality aversion versus joint-gain motives; and a dynamic Face Emotion Recognition Task (FERT) measuring emotion-detection thresholds and recognition accuracy, performed with naturalistic dynamic stimuli. Timing of assessments was planned relative to pharmacodynamics: FERT at 2 hours (peak), MET at 3 hours, and SVO at 4 hours after administration. Endocrine sampling measured plasma oxytocin and copeptin before and at 1 and 2 hours post-dose, and cortisol, prolactin and testosterone before and 2 hours after drug administration; MDMA concentrations were assayed repeatedly by HPLC–tandem mass spectrometry to derive pharmacokinetic parameters (Cmax, AUC0–6h, Tmax). Statistical analyses used repeated-measures ANOVAs with drug as a within-subject factor and sex as a between-subject factor where appropriate. Trait empathy effects were examined by median-split IRI scores as between-subject factors. Tukey post hoc tests followed significant ANOVA effects; order and prior-drug experience were tested as potential confounders. Pearson correlations explored associations between behavioural, endocrine and pharmacokinetic measures.

Subjective mood measures showed robust acute MDMA effects: VAS ratings increased for 'happy' (F1,31 = 120.59, P < 0.001), 'open' (F1,31 = 80.77, P < 0.001) and 'close to others' (F1,31 = 67.52, P < 0.001). MDMA also increased multiple AMRS subscales including activity, extroversion and well-being (all P < 0.01). A sex × treatment interaction reached significance for the 'happy' VAS, but post hoc tests did not reveal between-sex differences after MDMA. On the MET, MDMA significantly increased both explicit and implicit emotional empathy across all stimuli (explicit: F1,31 = 6.05, P = 0.019; implicit: F1,31 = 4.29, P = 0.047). These increases were driven by positive-valence stimuli (explicit: F1,31 = 8.60, P = 0.006; implicit: F1,31 = 5.02, P = 0.032) and not by negative-valence stimuli. A treatment × sex interaction was found for implicit emotional empathy (F1,31 = 4.68, P = 0.039) with a trend for explicit empathy; post hoc tests showed that the increases occurred only in men (explicit P = 0.025; implicit P = 0.022) and not in women. Cognitive empathy (MET accuracy) was not affected by MDMA. Behavioural prosociality measured by the SVO showed a significant MDMA treatment effect on the SVO angle (F1,31 = 4.42, P = 0.044) with a treatment × sex interaction (F1,31 = 5.52, P = 0.026); post hoc comparisons indicated a significant MDMA-induced increase in prosocial behaviour in men (P = 0.008) but not in women. There was a tendency for MDMA to reduce the inequality-aversion index among subjects with prosocial orientation (F1,20 = 3.39, P = 0.079), suggesting a shift toward fairness preference. In the FERT, MDMA impaired overall emotion-recognition accuracy (F1,31 = 28.63, P < 0.001) with a sex × treatment interaction (F1,31 = 6.04, P = 0.020). Women performed significantly worse after MDMA (P < 0.001), whereas men did not show a significant change. Valence-specific deficits after MDMA included reduced correct recognition of fearful (F1,31 = 14.90, P < 0.001), angry (F1,31 = 18.60, P < 0.001), disgusted (F1,31 = 5.81, P = 0.022) and surprised (F1,31 = 9.79, P = 0.004) faces; recognition of happy faces was unchanged. Impairments for fearful and sad faces differed by sex (fear: F1,31 = 6.61, P = 0.015; sad: F1,31 = 9.42, P = 0.005), with significant deficits in fear, anger and sadness recognition occurring only in women. In women, fear-recognition accuracy correlated inversely with MDMA Cmax and AUC0–6h (r = −0.62, P = 0.014; r = −0.66, P = 0.006, n = 16). MDMA also increased the detection threshold for fearful faces (F1,31 = 4.92, P = 0.032). Endocrine findings showed significant MDMA-induced increases in plasma oxytocin (F1,31 = 19.84, P < 0.001), cortisol (F1,31 = 98.70, P < 0.001) and prolactin (F1,31 = 127.81, P < 0.001), whereas copeptin and testosterone were unchanged. No correlations were observed between these neuroendocrine changes and the empathogenic or prosocial behavioural effects. Pharmacokinetics indicated greater MDMA exposure in women: Cmax and AUC0–6h were higher in women than men (Cmax: men 209 ± 6.4 ng/ml vs women 269.9 ± 9.3 ng/ml; AUC0–6h: men 926.8 ± 29.3 ng·ml−1·h vs women 1146.5 ± 37.9 ng·ml−1·h), and relative dose (mg/kg) was higher in women. Tmax averaged ~2.44 h overall and differed by sex (men ~2.10 h; women ~2.75 h). Apart from the inverse relationship between MDMA exposure and fear recognition in women, no other pharmacokinetic–pharmacodynamic correlations were reported.

The study found that MDMA selectively enhanced emotional empathy and prosocial behaviour while impairing recognition of negative facial emotions, with notable sex differences. Specifically, MDMA increased explicit and implicit emotional empathy and increased prosocial allocations in the SVO test in men, raising male empathy and prosociality to levels similar to placebo-treated women. Cognitive empathy as measured by MET accuracy was unchanged. In contrast, MDMA reduced accuracy for several basic emotions—especially fear, anger and sadness—with the largest deficits observed in women and a dose–exposure relationship for fear-recognition impairment in women. The investigators related these behavioural effects to MDMA’s primary pharmacology as a serotonin and norepinephrine releaser and discussed possible downstream involvement of oxytocin. MDMA increased plasma oxytocin alongside cortisol and prolactin, and the pattern of enhanced emotional but not cognitive empathy resembled effects previously observed after oxytocin administration. Nevertheless, no correlations were found between circulating oxytocin (or other measured hormones) and empathogenic or prosocial outcomes; the authors suggested several reasons for this null finding, including the imperfect relation between peripheral and central neurohormone levels, the timing of blood draws relative to tasks, and use of a single relatively high dose that may have produced ceiling effects. Relating results to prior work, the discussion noted consistency with studies showing reduced recognition of negative emotions after serotonergic or noradrenergic manipulations and with imaging data indicating MDMA-related attenuation of amygdala responses to negative faces and enhanced reward-related responses to positive faces. The authors proposed that the valence-specific shift—reduced sensitivity to negative signals and greater responsiveness to positive social cues—could underlie both the recreational sociability effects of MDMA and potential therapeutic utility when combined with psychotherapy for conditions involving social threat (for example, post-traumatic stress disorder). Key limitations acknowledged by the study team included testing only a single, relatively high MDMA dose (raising blinding concerns because subjective effects revealed treatment), potential bias from unblinding, and multiple statistical comparisons across many social-cognitive measures. The authors recommended future work using varied doses, active placebos, investigation of sex differences in clinical populations, and genetic or pharmacological approaches to probe the role of oxytocin and other mediators.