Psychedelic drugs: neurobiology and potential for treatment of psychiatric disorders

During the 1950s and 1960s, classic serotonergic psychedelics such as LSD and psilocybin were widely investigated in psychotherapeutic contexts, yielding a large but methodologically heterogeneous literature (more than 1,000 papers and reports involving roughly 40,000 subjects). Regulatory restrictions that followed the drugs' popularisation curtailed human research for several decades. With the advent of modern neuroimaging and molecular tools since the 1990s, experimental work has resumed and begun to clarify receptor-level actions, systems-level brain effects and cognitive–emotional consequences of psychedelics. Vollenweider and colleagues set out to review recent advances in the neurobiology and neuropharmacology of classic psychedelics and to consider mechanisms that might underlie their therapeutic effects. The review synthesises evidence from molecular pharmacology, animal electrophysiology and plasticity studies, human neuroimaging and electrophysiology, psychometric assessments of altered states, and recent clinical trials, and highlights discrepancies and key knowledge gaps while outlining directions for future research and clinical development.

Receptor pharmacology and cellular signalling. The review summarises that classic psychedelics fall into three chemical classes (indoleamines such as DMT and psilocin, phenylalkylamines such as mescaline and DOI, and ergolines such as LSD) and that a unifying molecular action is agonism at serotonin 5-HT2A receptors. Human positron emission tomography (PET) studies show a correlation between 5-HT2A receptor occupancy and subjective effects, and pre-treatment with the 5-HT2A antagonist ketanserin abolishes most subjective effects of psilocybin, LSD and DMT. Some 5-HT1A activity also appears to contribute to particular visual and attentional effects. Functional selectivity at 5-HT2A signalling (different intracellular pathways engaged by hallucinogenic versus non-hallucinogenic agonists) is noted as potentially important for designing therapeutics. Regulation of neuronal activity. Electrophysiological and in vivo animal work indicate that activation of cortical 5-HT2A receptors, richly expressed on apical dendrites of deep-layer pyramidal neurons in prefrontal cortex (PFC), leads to increases in glutamate release and postsynaptic AMPA activation. DOI produced a net-excitatory effect on many pyramidal neurons in rat PFC, including a reported 481% increase in mean firing rate in a subset of neurons, although some neurons were inhibited via interneuron-mediated mechanisms. The authors highlight that effects vary by cortical region, dose and receptor distribution, and that presynaptic 5-HT2A receptors on thalamocortical afferents also modulate NMDA-mediated transmission. Neuroplasticity. Preclinical studies show that psychedelics can increase cortical glutamate, brain-derived neurotrophic factor (BDNF) and downstream signalling through TRKB and mTOR pathways, promoting synaptogenesis and structural and functional changes in cortical neurons in rodents and in vitro. Separate mechanisms have been proposed for postsynaptic 5-HT2A gating of AMPA-dependent long-term depression and presynaptic 5-HT2A gating of NMDA-dependent associative learning. The authors caution that whether these plasticity effects translate to humans and account for durable clinical improvements remains to be established. Systems-level activity and connectivity. Neuroimaging and electrophysiological measures reveal several reproducible and some inconsistent patterns. Evidence supports altered thalamic gating within cortico‑striato‑thalamo‑cortical (CSTC) loops: LSD, psilocybin and DMT can increase thalamic functional connectivity with sensory cortical regions and alter effective connectivity (for example, increased thalamus→posterior cingulate cortex under LSD), consistent with models proposing increased bottom-up information flow. PET and blood‑flow studies reported regionally specific metabolic and perfusion changes, including prefrontal hypermetabolism in some studies; however, findings differ by drug, dose and preprocessing choices. Functional connectivity analyses have reported increased integration of sensory and somatomotor networks together with disintegration of associative networks (including reduced integrity within the default mode network, DMN), although between‑study results are heterogeneous and sensitive to preprocessing choices such as global signal regression (GSR). Electrophysiological and complexity measures show increased signal entropy or diversity (Lempel–Ziv complexity, sample entropy, Shannon entropy) under several psychedelics, a finding incorporated into the REBUS (relaxed beliefs under psychedelics) model linking reduced precision of high‑level priors with increased bottom‑up information flow. Altered cognition, emotion and social processing. Acute experimental and patient studies demonstrate that psychedelics alter emotional processing (for example, dose‑dependent reductions in recognition of negative facial expressions, reduced amygdala reactivity correlated with increased positive mood), self‑processing (loosening of self‑boundaries, ego‑dissolution/mystical‑type experiences) and social cognition (increased emotional empathy, reduced feelings of social exclusion). Some post‑acute effects on emotion and amygdala reactivity were reported up to 1 week in healthy volunteers, with return to baseline by 1 month in some studies, whereas other studies in patients with treatment‑resistant depression reported sustained improvements at 1 month. Self‑reported increases in interpersonal closeness and prosocial behaviour have been reported up to months after administration in some cohorts. Perceptual effects. Visual and other sensory changes are common and relate to electrophysiological findings such as decreased posterior alpha oscillations, increased excitability in the visual pathway, and altered event‑related potentials (increased P1, decreased N170 under psilocybin), consistent with an imbalance between early sensory processing and later integration. Clinical evidence from recent trials. Since 2010, clinical research has resumed, using a model of preparation, one or a few supervised drug sessions and post‑drug integration. Open‑label trials reported reductions in alcohol use sustained up to 53 weeks and nicotine cessation benefits up to 6 months following psilocybin‑assisted therapy, with outcomes linked to the intensity of mystical‑type experiences. Open‑label ayahuasca studies reported rapid antidepressant effects lasting up to 3 weeks. An open‑label trial of two psilocybin doses in treatment‑resistant depression showed significant reductions in self‑rated depressive symptoms lasting up to 6 months, with maximal effect at 5 weeks; acute pleasurable self‑dissolution predicted the week‑5 improvement. Several double‑blind, placebo‑controlled trials (including three psilocybin trials in patients with advanced cancer, an LSD trial in life‑threatening illness and a single‑dose ayahuasca trial in treatment‑resistant depression) reported reductions in anxiety and depressive symptoms, often with relationships between acute mystical‑type experience intensity and clinical benefit. The authors note heterogeneity across studies in design, sample size and imaging analyses.

Vollenweider and colleagues interpret the assembled evidence as converging on a central role for 5-HT2A receptor activation in producing the acute subjective effects of classic psychedelics and implicating this receptor system in modulation of self‑representation, social functioning and mood regulation. They propose that downstream consequences of 5-HT2A activation — including glutamate release, engagement of AMPA/NMDA receptors, and activation of neurotrophic signalling (BDNF, TRKB, mTOR) — could mediate synaptic remodelling that contributes to the sustained clinical benefits reported after only a few administrations. The review situates recent neuroimaging findings within mechanistic models: disrupted thalamic gating and altered cortico‑cortical integration may explain sensory flooding and ego‑dissolution, while increased signal entropy and decreased precision of high‑level priors (the REBUS framework) offer a unifying account linking neuronal complexity changes to altered perception, cognition and affect. The authors stress that these explanatory models are not mutually exclusive and likely reflect different facets of how psychedelics change neuronal dynamics. Key limitations and uncertainties are acknowledged. Many neuroimaging studies are small (<20 participants), use disparate preprocessing pipelines (notably GSR decisions that can qualitatively affect results), and produce inconsistent findings across analytic approaches. Translational gaps remain between animal plasticity findings and human clinical outcomes: it is not established whether observed neuroplastic changes in animals occur in humans or whether they causally underlie therapeutic effects. The necessity of the subjective psychedelic experience for clinical efficacy is unresolved, as is the identification of individual predictors or biomarkers of response. The authors also note the heterogeneity of psychotherapeutic support used across trials and the need to define the contribution of integration work to sustained benefits. Implications for future research and practice are outlined. Larger, placebo‑controlled and longitudinal clinical trials are needed to replicate promising early therapeutic signals and to characterise temporal dynamics of functional connectivity and neuroplasticity markers. Experimentally manipulating receptor subtypes and signalling pathways may yield therapeutics that dissociate hallucinogenic effects from putative beneficial plasticity. Finally, the authors suggest that integrating psychedelics into clinical practice should be conceptualised as pharmacologically assisted psychotherapy, and they call for systematic evaluation of psychotherapeutic approaches, biomarkers of efficacy and safety across disorders.

The authors conclude that renewed neuroscientific and clinical research has begun to map how classic psychedelics act from receptors to systems to subjective experience, revealing potential trans‑diagnostic mechanisms (5‑HT2A‑mediated modulation of emotional, social and self‑processing; and putative glutamate‑driven neuroplasticity) that might be harnessed therapeutically. They emphasise that psychedelics represent a distinct treatment model — rapid‑acting, with lasting effects after few doses — but that many mechanistic questions and clinical uncertainties remain and require larger, rigorous trials and careful study of psychotherapeutic contexts.