Persisting effects of ayahuasca on empathy, creative thinking, decentering, personality, and well-being

Ayahuasca is a traditional Amazonian psychoactive brew whose combination of Banisteriopsis caapi (monoamine oxidase inhibitors) and Psychotria viridis (DMT) produces robust psychedelic effects. Previous naturalistic and experimental studies have reported that ayahuasca can alter cognition and emotion and may ameliorate symptoms of stress-related psychopathologies. However, evidence remains limited and sometimes mixed regarding which cognitive and affective processes are affected and how long effects persist; in particular, findings on creative thinking, empathy, mindfulness-related capacities (such as decentering), personality traits, and subjective well-being are sparse or inconsistent. Kiraga and colleagues therefore carried out a naturalistic, observational study to characterise sub-acute (the morning after) and persisting (7 days after) changes following attendance at ayahuasca ceremonies. By administering a comprehensive test battery at baseline, within 24 hours, and at 1 week post-ceremony, the investigators aimed to distinguish direct and more enduring effects on flexible thinking (convergent and divergent creativity), cognitive and emotional empathy, decentering, life satisfaction, and personality, and to explore relationships between changes in empathy and well-being. The authors hypothesised post-ceremonial increases in creative and empathic abilities, greater life satisfaction and decentering, and potential personality changes such as increased openness and decreased neuroticism.

This was a prospective naturalistic observational study of volunteers attending ayahuasca retreats at four locations in the Netherlands between 2017 and 2019. Recruitment occurred on-site or via email after retreat organisers informed attendees about the study. Eligible volunteers were aged 18 or older and proficient in English; participation was voluntary and unpaid. Baseline assessments were completed in the week before the ceremony (online or on-site), the sub-acute assessment was completed on-site within 24 hours after the first ayahuasca ingestion, and the follow-up assessment was completed online 7 days after the ceremony. The research team did not participate in the screening, preparation, administration or supervision of the ceremonies. Ceremony settings were described as neoshamanic group retreats with facilitators who performed personal intakes and preparatory instructions but were not clinically trained. Ceremonies varied in rituals, music, duration (2–4 days), and facilitator practices; participants ingested ayahuasca in the evening and generally met the next day to discuss experiences. Four ayahuasca samples (one per site) were collected and analysed by LC-MS to quantify DMT, harmine and harmaline concentrations; mean amounts consumed were reported with standard deviations. The primary measures formed a multi-domain battery administered mostly at all three time points. Behavioural tests included the Multifaceted Empathy Test (MET) for cognitive empathy (CE) and implicit/explicit emotional empathy (EE), and the Picture Concept Test (PCT) for convergent thinking (correct solutions) and divergent thinking (fluency, originality, and an originality/fluency ratio). Questionnaires comprised the Satisfaction With Life Scale (SWLS), the Experiences Questionnaire—Decentering subscale (EQ-D), the Big Five Inventory (BFI; assessed at baseline and 1 week), and the Persisting Effects Questionnaire (PEQ; shortened 90-item version administered at 7 days). Acute subjective experience was assessed once after the ceremony using the Ego Dissolution Inventory (EDI) and ten Visual Analogue Scales (VAS). Statistical analyses used ordinary least squares multi-linear regression models with Session as a three-level dummy variable (baseline, morning after, 7 days after). When a main effect of Session was found, planned contrasts compared baseline to sub-acute and follow-up. Valence-specific MET analyses (positive versus negative pictures) were conducted when appropriate. Ayahuasca-related VAS scores were tested against zero with one-sample t-tests. Hypothesis-driven correlations (one-sided Pearson) tested relationships between empathy changes and satisfaction with life using change scores (24 h–baseline and 7 days–baseline). The alpha threshold was p ≤ 0.05, and Cohen's d effect sizes were reported for significant results.

Participants and sample characteristics: Sixty-four volunteers provided informed consent. Completion rates varied across measures and time points due to dropout and missing data; the extracted text reports that 43 participants completed parts of the baseline assessment, 20 completed the morning-after assessments, and 19 completed the 1-week follow-up for some measures, while other instruments had different Ns (detailed below). The mean age was 39 ± 11.3 years; 55% identified as male, 42% as female, and 3% did not disclose sex. Most participants were European (78%), 61% had completed academic education, 23.4% had prior ayahuasca use, and 52% had used other psychedelics. Based on reported consumption and analysis of the four samples, mean amounts consumed were DMT 57.44 ± 25.77 mg, harmine 127.08 ± 97.12 mg, and harmaline 59.84 ± 73.27 mg. Cognitive empathy (MET): For the MET, 41 participants at baseline, 25 at the morning after, and 12 at 1 week completed all parts used in the analysis. A significant main effect of Session on cognitive empathy was observed (F(2,75) = 12.77; p < 0.001). Compared with baseline, recognition accuracy increased the morning after (p < 0.001; d = 0.93) and remained elevated at 1 week (p = 0.01; d = 0.79). Valence-specific analyses showed improved recognition of both positive and negative emotions the morning after (positive p < 0.001; d = 0.79; negative p < 0.001; d = 0.85). At follow-up, recognition of positive emotions remained higher than baseline (p = 0.01; d = 0.80), while the change for negative emotions was not statistically significant (p = 0.056; d = 0.59). Emotional empathy (MET implicit/explicit): A main effect of Session on implicit emotional empathy (arousal ratings) was found (F(2,75) = 4.42; p = 0.04). Implicit EE was significantly higher at the 1-week follow-up versus baseline (p = 0.047; d = 0.62); no significant change was reported from baseline to the sub-acute measurement (p = 0.266). Explicit EE (“concern”) showed no main effect (F(2,75) = 0.572; p = 0.452). Valence analyses indicated higher arousal toward positive pictures at 1 week (p = 0.012; d = 0.78) but no change toward negative pictures. Convergent and divergent thinking (PCT): Convergent thinking data were available for 43, 20, and 19 participants at baseline, sub-acute, and follow-up, respectively. A significant Session effect on convergent thinking emerged (F(2,79) = 12.08; p < 0.001). Correct solutions did not change significantly the morning after (p = 0.58) but improved significantly at 1 week (p < 0.001; d = 0.83). Divergent thinking measures showed no Session effects on originality (F(2,79) = 0.858; p = 0.357) or fluency (F(2,79) = 0.717; p = 0.4). However, the originality/fluency ratio—an estimator of divergent-thinking quality—decreased significantly both the morning after (p = 0.038; d = 0.5) and at 1 week (p < 0.001; d = 1.27) relative to baseline. Satisfaction with life (SWLS) and decentering (EQ-D): SWLS was completed by 43 participants at baseline, 27 the morning after, and 24 at follow-up. Session exerted a significant effect on life satisfaction (F(2,91) = 17.0; p < 0.001). Compared with baseline, SWLS increased the morning after (p = 0.023; d = 0.5) and further at 1 week (p < 0.001; d = 0.9). Decentering scores (N = 42 baseline, 21 sub-acute, 22 follow-up) showed a pronounced linear increase across sessions (F(2,82) = 60.67; p < 0.001), with large effects at 24 h (p < 0.001; d = 1.0) and 7 days (p < 0.001; d = 1.76) versus baseline. Acute subjective experience and VAS: Thirty-one participants completed the Ego Dissolution Inventory the morning after; mean score was 45.8 (SD 33.7), range 0–95.6. Twenty participants completed the VAS; one-sample t-tests versus zero showed significant increases for all VAS items (t(19) = 5.15–7.95; p ≤ 0.000), with very large effect sizes (d = 1.63–2.51). Personality (BFI) and persisting effects (PEQ): Personality was assessed at baseline (N = 33) and 1 week (N = 8). A Session effect was reported for neuroticism (F(1,39) = 4.68; p = 0.037; d = 0.98), with lower neuroticism at 1 week; no significant changes were found for openness, extraversion, agreeableness, or conscientiousness. The PEQ was returned by 17 participants at 7 days; notable findings included that 11.8% rated the experience as the single most meaningful of their lives and 47.1% rated it the most spiritually significant for them. Reports of psychological challenge and insightfulness were distributed across severity/meaningfulness categories as detailed in the extraction. Correlations: A hypothesis-driven one-sided Pearson correlation (n = 10) showed that increases in implicit EE toward positive stimuli correlated with 7-day increases in satisfaction with life (r = 0.584; p = 0.038).

Kiraga and colleagues interpret their findings as evidence that a single ayahuasca ceremony in a naturalistic group setting was associated with sub-acute and short-term improvements in several psychological domains. The most robust and consistent changes were increases in cognitive empathy (ability to correctly recognise emotions), higher implicit emotional empathy toward positive stimuli at 1 week, greater life satisfaction at 24 hours and 7 days, and enhanced decentering across the same time frame. Convergent thinking improved at 1 week, whereas a measure of divergent-thinking quality (originality weighted for fluency) decreased after the ceremony. A decrease in trait neuroticism at 1 week was also reported, while openness and other Big Five traits did not change. The authors situate these results within prior literature showing acute and persisting psychological benefits after psychedelics, noting that the present study is among the first to demonstrate substantial sub-acute and 1-week increases in cognitive empathy following ayahuasca. They acknowledge that previous psychedelic studies have produced mixed effects on emotion recognition and empathy, and suggest that the observed empathy and decentering changes might be therapeutically relevant for stress-related disorders that involve social difficulties and rigid negative thought patterns. A hypothesis-driven correlation between early increases in positive implicit empathy and later life satisfaction offers limited evidence for temporal directionality, but the authors emphasise that causality cannot be inferred from their data. Several limitations and uncertainties are acknowledged. The naturalistic design yields high ecological validity but introduces self-selection bias, heterogeneous ceremony contexts, variable alkaloid doses across sites, and substantial attrition, which complicate generalisability. The absence of a control or placebo group prevents firm causal attribution; practice effects on cognitive tasks cannot be entirely excluded despite use of parallel test versions. Small and variable sample sizes for some measures (notably BFI at follow-up and PEQ) limit statistical power. The authors also note broader heterogeneity in the psychedelic literature (differences in substance, setting, timing of assessment) which hampers synthesis across studies. For future research, the investigators recommend placebo-controlled experimental designs to disentangle non-pharmacological influences (set and setting) from drug effects, and longitudinal naturalistic studies and clinical trials to evaluate therapeutic potential in clinical populations. They highlight decentering as a promising mechanism to study further given its links to reduced rumination and adaptive coping, and suggest incorporating measures such as sleep quality which may mediate or confound outcomes.