Effects of serotonin 2A/1A receptor stimulation on social exclusion processing

Social ties are crucial for physical and mental health, and increased reactivity to social exclusion contributes to the development, progression, and treatment challenges of multiple psychiatric disorders. Previous work implicates the serotonin (5-HT) system in mood, affect and social cognition, and psilocybin (Psi) — a preferential 5-HT2A/1A receptor agonist — has been shown to attenuate processing of negative emotional stimuli and alter self-experience. However, research to date has largely examined non-interactive emotional stimuli, leaving unclear whether direct 5-HT2A/1A receptor stimulation modulates the neural and subjective processing of negative social interactions such as rejection or ostracism. Preller and colleagues designed a multimodal brain imaging study to address this gap. Using a double-blind, randomised, counterbalanced within-subject cross-over design, the investigators compared acute oral psilocybin (0.215 mg/kg) with placebo in 21 healthy volunteers while measuring functional MRI responses during an interactive social exclusion paradigm (Cyberball) and proton magnetic resonance spectroscopy (1H-MRS) in the dorsal anterior cingulate cortex (dACC). The primary aim was to determine whether Psi reduces neural and subjective responses to social exclusion and to explore whether changes in excitatory neurometabolites, particularly aspartate (Asp) and glutamate (Glu), relate to any observed effects. This study is notable for combining task-based fMRI of a real-time social interaction with spectroscopic probing of dACC biochemistry, permitting exploration of receptor-level modulation of social pain processing and its putative neurochemical substrates in humans.

Participants were 21 healthy adults (12 males, 9 females; mean age 26.48 y, SD 4.76, range 20–37). All provided written informed consent; psilocybin use was authorised by Swiss authorities and the study had ethical approval. Two subjects had MRS data excluded due to premature termination of acquisition, so MRS analyses were based on 19 participants. Further screening and eligibility details are reported in the supplementary information (SI Methods), which is cited but not reproduced in the extracted text. The investigation used a randomised, double-blind, placebo-controlled within-subject cross-over design. Each participant completed two sessions at least 10 days apart, receiving oral psilocybin (0.215 mg/kg) in one session and placebo (maltose) in the other. The chosen Psi dose was based on prior work indicating tolerability and emotion-modulatory effects. Participants completed behavioural questionnaires and physiological measures (blood pressure, pulse) alongside imaging. During fMRI, subjects played three separate runs of Cyberball, a virtual ball-tossing paradigm simulating interactive social inclusion and exclusion. The three conditions were: implicit social exclusion (ISE; participants watched because of an ostensible connectivity problem), inclusion (INCL; normal play), and explicit social exclusion (ESE; participants received five throws then were excluded for the remainder). Post-scan, participants completed a posttask questionnaire (PTQ) probing subjective feelings of exclusion and related constructs. MRI data were acquired on a Philips Achieva 3.0T scanner. Task fMRI used a 32-channel head coil; 1H-MRS used a different transmit–receive coil, requiring participants to be repositioned between modalities. A JPRESS (J-resolved) 1H-MRS protocol collected spectra from a dACC voxel (32 × 21 × 24 mm3). Spectra were quantified with Profit2 software to estimate up to 18 metabolites including Asp and Glu. fMRI analyses used an event-related general linear model implemented in SPM8. Statistical analyses of behavioural and physiological data used repeated-measures ANOVAs and paired t tests with Bonferroni correction where applicable; correlation analyses examined relationships among BOLD differences, subjective effects (5D-ASC, PANAS, PTQ), and metabolite concentrations.

Subjective and physiological effects: Psilocybin produced robust changes on the Altered States of Consciousness (5D-ASC) questionnaire. A repeated-measures ANOVA showed significant main effects of treatment [F(1,20) = 93.54, P < 0.001], scale [F(10,200) = 18.98, P < 0.001], and a treatment × scale interaction [F(10,200) = 19.67, P < 0.001], with simple effects indicating increased ratings on all 5D-ASC scales after Psi versus placebo (all P < 0.05). The authors report low anxiety ratings on the 5D-ASC (mean 7.5%, maximum 46%), and an increase in basic positive mood measures; detailed PANAS results are provided in the supplementary information. Systolic and diastolic blood pressure and pulse showed slight but significant increases after Psi compared with placebo (all P < 0.05). Behavioural response to Cyberball: On the posttask questionnaire participants reported feeling less excluded after Psi than after placebo [t(20) = 2.71, P < 0.01]. Other PTQ items (self-esteem, meaningful existence, control, inclusion, belongingness, liking) did not differ significantly between conditions (all P > 0.1). Estimates of the number of throws received did not differ between Psi and placebo, indicating comparable awareness of exclusion across treatments. fMRI task activation and drug effects: In the placebo condition, the explicit social exclusion (ESE > INCL) contrast activated regions typically linked to social pain: bilateral anterior midcingulate cortex (aMCC), posterior MCC (pMCC), left inferior orbitofrontal cortex, and bilateral middle frontal gyrus (MFG). Comparing placebo versus psilocybin for ESE > INCL revealed significantly reduced activation under Psi in the right aMCC and left MFG; no regions showed greater activation under Psi for this contrast. For the implicit exclusion contrast (ISE > INCL), placebo showed activation in left anterior/posterior MCC and right pregenual ACC; Psi reduced activation bilaterally in anterior/posterior MCC and in the left pregenual ACC versus placebo. The context-specific contrast (not receiving INCL > receiving INCL) recruited subgenual ACC under placebo but showed no significant Psi versus placebo differences, suggesting Psi's effects were specific to exclusion contexts rather than to not receiving the ball per se. Correlations with subjective effects: A correlation analysis examined differences in BOLD responses (placebo > Psi) and subjective drug effects. A significant positive correlation emerged between reduced activation in the right MCC and the 5D-ASC 'experience of unity' scale in the Psi condition (r = 0.53, P < 0.02). No other significant correlations between BOLD responses and 5D-ASC, PANAS or PTQ items were reported (all P > 0.08 to P > 0.1). Relationships with MRS metabolites: The dACC voxel used for MRS overlapped the region where Psi reduced activation for ESE > INCL. Changes in dACC activity (placebo minus Psi) were strongly correlated with baseline-corrected aspartate concentrations after Psi (r = 0.80, P < 0.001, n = 19). Additionally, Asp concentration at the Psi follow-up measurement correlated with dACC activity (r = -0.56, P < 0.02, n = 19). No significant correlations were observed for other metabolites or metabolite measures (all P > 0.17).

Preller and colleagues interpret their multimodal results as evidence that acute 5-HT2A/1A receptor stimulation with psilocybin attenuates the neural and subjective experience of social pain. Task-based fMRI showed that Psi reduced exclusion-related activation in regions centrally implicated in social pain processing — notably the dorsal anterior cingulate cortex (dACC/aMCC) and middle frontal gyrus — and participants reported feeling less excluded during the Psi condition. The authors propose that reduced activation in the MFG may reflect a diminished need for frontal regulatory processes because affective distress was lower under Psi. The investigators further link neural changes to alterations in self-experience: decreased dACC responses correlated with higher scores on the 5D-ASC 'experience of unity' scale, suggesting that changes in the sense of self or boundary dissolution may modify interpersonal processing and make negative social experiences more tolerable. In addition, spectroscopic findings associated reductions in dACC BOLD responses with decreases in aspartate (Asp) concentrations; the authors discuss Asp as an excitatory amino acid that preferentially activates NMDA receptors and consider possible ties to glial–neuronal metabolic pathways, including the malate–aspartate shuttle and a proposed neuron-to-astrocyte Asp transcellular pathway relevant to glutamate–glutamine cycling. Several caveats are acknowledged. The complex resonance pattern of Asp challenges in vivo quantification, although the authors report JPRESS acquisition and fitting parameters that yielded acceptable Cramér–Rao lower bounds and data quality. They note that MRS cannot distinguish synaptic from extrasynaptic pools, so metabolite concentration changes do not directly prove altered neurotransmission. Potential confounds from drug effects on cerebral vasoactivity are also discussed; prior work cited by the authors found no evidence that non-neural vascular changes account for psilocybin-related BOLD effects, and the task-specific nature of contrasts in this study argues against a global vascular explanation. The authors further recognise conflicting PET versus fMRI findings in the literature regarding resting ACC activity after psilocybin and emphasise that their task contrasts do not permit conclusions about global increases or decreases in ACC activity. In terms of broader relevance, the authors suggest that modulating 5-HT2A/1A receptors — and potentially targeting Glu/Asp-related metabolic pathways — could be pertinent to disorders characterised by heightened rejection sensitivity, such as major depression and social anxiety, and that Psi's capacity to reduce negative social bias could relate to putative antidepressant effects. They caution, however, that the present findings derive from healthy volunteers and may not directly generalise to clinical populations. Finally, the study is presented as the first multimodal human imaging demonstration that direct 5-HT2A/1A receptor stimulation can regulate social pain processing and that Asp-related biochemistry in the dACC is associated with these effects.