Trait Openness and serotonin 2A receptors in healthy volunteers: A positron emission tomography study

Psilocybin and related serotonergic psychedelics produce profound alterations in perception and meaning, sometimes described as mystical-type experiences, and these effects are primarily mediated by agonism at the serotonin 2A receptor (5-HT2A R). The 5-HT2A R is an excitatory receptor densely expressed across the cerebral cortex, and its dysfunction has been implicated in psychiatric conditions such as depression and schizophrenia. Prior studies have reported that acute and subacute exposure to psilocybin can produce lasting increases in the personality domain Openness; however, it remained unclear whether individual differences in baseline 5-HT2A R availability in psychedelic‑naïve healthy people are related to trait Openness. Stenbaek and colleagues set out to test whether in vivo neocortical 5-HT2A R availability, measured with PET, is associated with trait Openness. To do so they analysed data from 159 healthy participants who had undergone PET imaging with either the antagonist radioligand [18F]altanserin or the agonist radioligand [11C]Cimbi-36, alongside self-reported NEO PI-R scores for Openness. The study aims to determine whether baseline 5-HT2A R binding correlates with individual variation in Openness, thereby addressing a mechanistic question relevant to how psychedelics may relate to personality change.

Participants were drawn from the Cimbi PET database. From an initial pool of 200 healthy individuals with PET data and NEO PI-R assessments, the investigators excluded people with more than 365 days between PET and personality testing (n = 5) and those with body mass index >30 (n = 36), leaving 159 participants (53 females). Eight participants had completed both [18F]altanserin and [11C]Cimbi-36 scans. Self-reported lifetime psychedelic use was available for 134 participants, all of whom reported being psychedelic-naïve; for 33 [18F]altanserin datasets this information was not systematically recorded in the archive. Personality was measured with the Danish version of the NEO PI-R (240 items), and the analyses used the summed raw score for trait Openness and its six subfacets (Fantasy, Esthetics, Feelings, Actions, Ideas, Values). Two item responses were missing across the sample and were replaced with the neutral response "2" on the 0–4 Likert scale. Internal consistency for the Openness scale was high (Cronbach's α = 0.87). Educational attainment was recorded on a five‑point scale, though the extracted text does not clearly state whether education was included as a covariate in all models. MRI and PET acquisitions followed established Cimbi protocols. Structural T1-weighted MRI was used for segmentation and ROI delineation. PET imaging was performed on either a high-resolution HRRT scanner (approximate in-plane resolution 2 mm) or a GE-Advance scanner (approximate in-plane resolution 6 mm). [18F]altanserin was delivered as a bolus followed by continuous infusion to achieve steady state; [11C]Cimbi-36 imaging procedures were as previously described by the group. Plasma measurements and metabolite assessments were obtained according to standard methods (details referenced elsewhere). Image preprocessing comprised motion correction, within-frame Gaussian smoothing (10 mm for HRRT, 12 mm for GE-Advance before alignment), coregistration to the participant's MRI (manual methods for some GE-Advance [18F]altanserin scans, SPM otherwise), and automatic ROI delineation using PVElab. A large neocortical ROI was chosen because 5-HT2A R expression is low subcortically and highly correlated across neocortical subregions; the neocortex ROI included occipital, orbitofrontal, parietal, pre/postcentral, middle/inferior frontal, middle/inferior temporal, superior frontal, and superior temporal gyri. For [18F]altanserin the primary outcome was binding potential relative to total plasma concentration (BPP), which accounts for radiolabelled metabolites crossing the blood–brain barrier. For [11C]Cimbi-36 the primary outcome was nondisplaceable binding potential (BPND) derived from two‑tissue compartment modelling with arterial input; cerebellum served as the reference region for both tracers. Statistical analysis used separate ordinary least-squares linear regression models for the [18F]altanserin and [11C]Cimbi-36 groups, with neocortical 5-HT2A R binding as the predictor and trait Openness as the outcome. Age and sex were included a priori as covariates. Scanner-specific effects for [18F]altanserin (HRRT vs GE-Advance) were examined and found to yield similar effects, so pooled results are reported. The extracted text indicates the influence of sex on the association was also investigated; details on additional covariates (for example education) are not clearly reported in the provided extraction.

Across 159 healthy participants, no significant association was observed between neocortical 5-HT2A R availability and trait Openness for either radioligand. Specifically, in the [18F]altanserin group the association was nonsignificant (p = 0.5), and in the [11C]Cimbi-36 group it was also nonsignificant (p = 0.8). A combined model pooling data from both tracers did not materially change the outcome (p = 0.4). Interaction tests did not reveal significant effects of sex on the relationship (p > 0.35). The standardized regression coefficient reported in the discussion was −0.07 with a 95% confidence interval of −0.27 to 0.13, indicating a small effect estimate centred near zero. The NEO PI-R Openness scale showed acceptable psychometric properties in this sample (Cronbach's α = 0.87). Power calculations presented by the authors indicate that with N = 159 the study had approximately 80% power to detect correlations of r ≥ 0.2 (small-to-medium effects); the authors note they were underpowered to detect smaller effect sizes. No additional primary or secondary outcome effects, subgroup findings, or adverse events relevant to the primary question are reported in the extracted text.

Stenbaek and colleagues report that, in the largest PET investigation to date of a serotonin marker and trait Openness, there was no evidence supporting an association between baseline neocortical 5-HT2A R availability and Openness measured by the NEO PI-R. The null findings were consistent across two different PET radioligands—[18F]altanserin (antagonist) and [11C]Cimbi-36 (agonist)—and when data were pooled. The authors therefore conclude that individual differences in 5-HT2A R availability are not directly related to variation in trait Openness in healthy, psychiatrically screened participants. Putting these results in context, the investigators note that prior PET studies have reported associations between other serotonergic markers and personality dimensions—most notably negative correlations between 5-HT transporter (5-HTT) binding and traits related to Openness or Self‑transcendence, and mixed findings for 5-HT1A and 5-HT4 receptors. The current null result contrasts with some smaller and heterogeneous reports and, by virtue of sample size, provides relatively strong evidence against a moderate or larger association between 5-HT2A R availability and Openness at baseline. The authors discuss a possible reconciliation with experimental findings that psilocybin can increase Openness: changes in Openness after psychedelic exposure may be driven by acute 5-HT2A R agonism and downstream plasticity rather than by baseline receptor availability per se. They also reference evidence that the magnitude of psilocybin-induced mystical experiences—rather than baseline Openness—predicts increases in Openness, and that mystical experiences scale with psilocybin dose; thus baseline 5-HT2A R may not determine the trait but may be part of the pharmacological mechanism that enables acute effects. Key limitations acknowledged include limited sensitivity to detect very small effects despite the relatively large sample (80% power to detect r ≥ 0.2), possible sample bias due to exclusion of individuals with personal or family psychiatric history, and inherent limitations of self-report measures such as the NEO PI-R (social desirability, censorship, or response manipulation). The authors also note that even if a statistically significant association of the observed magnitude were present, its practical relevance would be limited given the small effect estimate. In sum, the study applied PET imaging to a large healthy sample and found no support for an association between 5-HT2A R availability and trait Openness, while recommending future work to explore 5-HT2A R–mediated mechanisms of personality change following psychedelic administration.