Brain serotonin 2A receptor binding predicts subjective temporal and mystical effects of psilocybin in healthy humans

Psilocybin is a serotonergic psychedelic that produces an acute altered state of consciousness typically described in three temporal phases: an onset of psychoactive effects, a peak plateau, and a return to normal waking consciousness. These acute effects frequently include perceptual alterations and changes in selfhood, affect and meaning, and can include so-called mystical-type experiences characterised by unity, profound positive mood, transcendence of time and space, and ineffability. Previous research indicates that dose, personality, and set and setting influence psilocybin responses, and that psilocybin’s effects are primarily mediated by agonism at the serotonin 2A receptor (5-HT2AR). However, it is unknown whether pre-drug individual differences in brain 5-HT2AR binding predict temporal features of the acute subjective drug intensity (SDI) time course or the magnitude of retrospective mystical-type experiences. Stenbaek and colleagues set out to test whether baseline neocortical 5-HT2AR binding, measured with the agonist PET tracer [11C]Cimbi-36, predicts three modelled temporal parameters of SDI (onset slope, duration of peak plateau, and return slope) and retrospective scores on the Mystical Experience Questionnaire (MEQ) following a single oral psilocybin session in healthy volunteers. The study therefore investigates a candidate neurobiological marker that could help explain individual differences in the phenomenology of the psychedelic experience and potentially inform future clinical research on predictors of therapeutic response.

Design and participants: Sixteen healthy volunteers (mean age 30.3 ± 6 years, range 24.2–49.8; six female) were recruited and screened. Key exclusion criteria included present or past psychiatric disorder (or immediate family history), neurological or significant somatic illness, recent use of hallucinogens (within six months) or other intoxicants (within one month), pregnancy or breastfeeding, MRI contraindications, and certain cardiovascular or haematological conditions. Data were pooled from two open-label studies from the same group that used the same psilocybin regimen and baseline PET imaging; all participants provided written informed consent and the studies had ethical approval. Imaging and predictor variable: Baseline brain 5-HT2AR binding was measured prior to the psilocybin session using [11C]Cimbi-36 PET. Dynamic PET scans (120 min) were acquired and analysed with a cerebellar reference region to derive non-displaceable binding potential (BPND) using the simplified reference tissue model (SRTM). The primary predictor was mean BPND across a neocortex region of interest comprising multiple cortical areas; subcortical signal was not considered due to low/noisy signal. High-resolution T1-weighted MRI scans were used for ROI delineation and PET–MRI coregistration. Intervention and psychometrics: On a separate day, participants received an oral weight-adjusted psilocybin dose of either 0.2 mg/kg (n = 7) or 0.3 mg/kg (n = 9), administered in 3 mg capsules. Two trained psychologists provided interpersonal support throughout the session, and a standard music playlist was used. Subjective drug intensity (SDI) was assessed approximately every 20 minutes during the session using a 0–10 Likert scale question (“How intense is your experience right now?”), answered orally. About 6 hours after ingestion participants completed the revised Mystical Experience Questionnaire (MEQ), a 30-item self-report instrument with four subscales (Mystical quality, Positive mood, Transcendence of time and space, Ineffability) and a total score. Modelling and statistics: For each participant the SDI time course was modelled with segmented linear regression containing two breakpoints to represent the three phases (onset, peak plateau, return). This produced three participant-level outcomes: onset slope (SDI change per 10 min), peak plateau duration (minutes), and return slope (SDI change per 10 min). One participant had missing SDI data and could not be modelled. Associations between neocortex BPND and each of the three SDI outcomes were tested in separate linear regressions with Bonferroni family-wise error rate correction for three tests; an additional regression tested BPND against MEQ total score. Covariates considered (dose, sex, PET setting, prior psychedelic experience) were retained only if associated with outcomes; they were not included in final models because they were not statistically significant. Exploratory analysis examined the association between BPND and time-to-peak (TMAX) plasma psilocin in a subset with available pharmacokinetic data (n = 6). Model assumptions were assessed visually and analyses were performed in R.

Sample and descriptive data: The analysed sample comprised 16 participants (mean age 30.3 ± 6 years, six female); 13 were psychedelic-naïve and three had past psychedelic experience. Baseline PET scans were acquired a median of 25 days (range 1–68) before the psilocybin session. No adverse drug reactions requiring crisis support were reported. SDI time course modelling: The segmented two-breakpoint model fitted individual SDI time courses well, explaining on average 94% of the variance (range 87–98%). Across participants the mean time to SDI peak was 71 minutes post-ingestion (range 43–167 min), mean peak plateau duration was 82 minutes (range 32–143 min), and mean time to return from peak to normal waking consciousness was 143 minutes (range 77–333 min). SDI ratings affected by short disruptions (for example visits to the toilet) and repeated end-of-session scores were excluded from the modelling. Associations with pre-drug 5-HT2AR binding: Pre-drug neocortex [11C]Cimbi-36 BPND was statistically significantly negatively associated with peak plateau duration: estimate −143.6 minutes per unit BPND (95% CI −238 to −49.6), pUNC = 0.006, pFWER = 0.018. Lower cortical BPND therefore predicted a longer peak plateau. BPND was also positively associated with the return slope coefficient: 1.2 SDI units/10 min per unit BPND (95% CI 0.6 to 1.9), pUNC = 0.001, pFWER = 0.003, indicating that higher BPND was associated with a faster decrease in SDI after the peak. There was no significant association between BPND and the onset slope (pUNC = 0.9, pFWER = 1). Associations with mystical-type experiences: Neocortex BPND was statistically significantly negatively associated with MEQ total score: −3.4 MEQ units per unit BPND (95% CI −6.2 to −0.7), pUNC = 0.02. Post-hoc (unadjusted) inspection of MEQ subscales showed similar negative point estimates for Positive mood (−3.3), Mystical (−3.8) and Transcendence of time and space (−3.5), while Ineffability showed a smaller effect (−1.9) with a confidence interval including zero. These subscale tests were post-hoc and p-values were not reported. Exploratory pharmacokinetic analysis: In the small subset with plasma psilocin TMAX data (n = 6), there was little evidence of an association with neocortex BPND. The standardised regression coefficient was 0.005 with R2 < 0.0001; the extracted estimate and confidence interval were imprecise, reflecting the very small sample.

Stenbaek and colleagues interpret these findings as evidence that individual differences in baseline neocortical 5-HT2AR binding are related to the temporal dynamics of the acute psilocybin experience and to the retrospective magnitude of mystical-type experiences. Specifically, lower baseline 5-HT2AR binding was associated with a longer plateau at peak subjective intensity, a more rapid decline from peak, and higher MEQ scores. The investigators argue that these results reinforce 5-HT2AR’s central role in shaping both the temporal profile and certain qualitative features of the psychedelic state. The authors note that the segmented SDI modelling provided a parsimonious, within-subject description of the psychedelic time course that correlated with MEQ outcomes; in particular, a longer peak plateau combined with a faster return to normal waking consciousness appeared to characterise experiences reported as more mystical. They caution that plateau length and return slope were strongly negatively correlated and therefore may not be independent components but rather jointly describe aspects of the finite time course of the experience. Mechanistic possibilities discussed by the investigators include differences in receptor density versus receptor occupancy dynamics. Under the assumption that lower BPND reflects lower receptor density, one hypothesis is that individuals with lower 5-HT2AR levels may have different temporal dynamics of psilocin–receptor interaction, for example faster peak occupancy and more rapid clearance from interstitial fluid, leading to prolonged plateau and abrupt return. An alternative is that baseline 5-HT2AR binding relates to broader serotonergic tone or pharmacokinetics; previous work linking plasma psilocin to SDI motivated an exploratory analysis of TMAX, but no association with BPND was observed in the small subset with pharmacokinetic data. The authors therefore recommend future studies routinely measure plasma psilocin to disentangle pharmacokinetic from receptor-level effects. Key limitations acknowledged by the investigators include the modest sample size typical of PET studies, which limits precision and generalisability and necessitates replication. They also highlight that SDI is a simple, content-agnostic measure and does not specify which experiential domains drove participants’ intensity ratings; nonetheless, its strong correlation with MEQ and with prior PET occupancy measures supports its utility as a minimally intrusive real-time metric. Finally, the authors underline that pharmacologically induced changes in 5-HT2AR (for example after antidepressant treatment) may not be equivalent to the stable, possibly genetically determined individual differences captured by baseline PET. In conclusion, the authors propose that baseline neocortical 5-HT2AR binding is a promising neurobiological marker linked to important temporal and phenomenological aspects of the psilocybin experience and suggest further research to test whether pre-drug 5-HT2AR predicts therapeutic outcomes in clinical populations.