LSD and First Episode Psychosis: Neural Overlap Revealed in Connectome Study
A 2026 connectome-based analysis finds partial neural overlap between LSD and first episode psychosis, challenging traditional psychotomimetic models and informing future research and policy.
New Evidence: LSD and First Episode Psychosis Share Partial Neural Overlap
A November 2026 study published via OpenAlex (source) provides direct evidence that the acute neural effects of lysergic acid diethylamide (LSD) partially overlap with those seen in medication-naïve, non-affective first episode psychosis (FEP), particularly within the dorsal attention network (DAN) and sensorimotor network (SMN). This finding is significant because it moves beyond long-standing assumptions that LSD is a straightforward 'psychotomimetic'—a substance that mimics psychosis—by demonstrating that the overlap is real but only partial and network-specific.
Mechanism and Context: Connectome-Based Classification Uncovers Selective Dysconnectivity
The study used connectome-based classification, a data-driven machine learning approach, to distinguish between the resting-state functional connectivity (RSFC) patterns of acute LSD administration and FEP. Classifiers were trained separately on LSD versus placebo datasets and on FEP patients versus matched healthy controls. These classifiers were then cross-applied to test whether the neural signatures of LSD could identify FEP and vice versa. The analysis revealed that both LSD and FEP show dysconnectivity in the DAN and SMN, suggesting a shared disruption in attention and sensorimotor processing. However, the overlap was not global—many neural changes in FEP were not mirrored by LSD, and vice versa, indicating distinct underlying mechanisms for many aspects of each condition.
This nuanced approach contrasts with earlier models that treated LSD as a broad model for psychosis, instead supporting a more differentiated view. Notably, the study's use of external validation for the LSD-placebo classifier strengthens the reliability of these findings, an important methodological advance over prior work that often relied on within-sample comparisons.
Policy and Research Implications: Rethinking Models and Risk Assessment
The demonstration of selective overlap between LSD and FEP has concrete implications for both psychiatric research and the development of psychedelic therapies. For researchers, the findings suggest that using LSD as a model for psychosis should be limited to specific neural circuits—particularly those involving attention and sensorimotor integration—rather than as a global proxy for the disorder. This may inform the design of future mechanistic studies and the interpretation of preclinical models.
For clinical trial designers and regulators, these results highlight the importance of screening for vulnerability to psychosis, especially in trials involving serotonergic psychedelics. The overlap in DAN and SMN dysconnectivity may serve as a biomarker for risk stratification or monitoring in early-phase studies. Additionally, the findings point to serotonergic dysfunction as a possible contributor to early psychosis, potentially opening new avenues for targeted intervention or prevention research.
- Trial Design: Inclusion and exclusion criteria for psychedelic trials may need to be refined to account for attention and sensorimotor network vulnerability.
- Regulatory Review: Agencies such as the U.S. Food and Drug Administration (FDA) and European Medicines Agency (EMA) may consider these neural markers when evaluating safety data for investigational psychedelic compounds.
Risks, Unknowns, and Real-World Failure Modes
While the study provides important new insights, several risks and unknowns remain. The partial overlap between LSD and FEP does not mean that LSD causes psychosis, nor does it validate the use of LSD as a comprehensive model for psychotic disorders. The sample included only medication-naïve, non-affective FEP patients, which may not generalize to chronic or affective psychoses. Furthermore, the acute effects of LSD are transient, whereas psychosis can be persistent and multifactorial.
One real-world failure mode not widely discussed in prior literature is the potential for over-reliance on neural overlap as a predictor of clinical risk. While DAN and SMN dysconnectivity may indicate increased vulnerability, the translation from network-level findings to individual patient outcomes is not straightforward. Overly broad exclusion of trial participants based on these biomarkers could inadvertently limit access to potentially beneficial interventions for otherwise suitable candidates.
Looking Ahead: Toward Nuanced Models and Safer Innovation
Future research should directly compare other serotonergic compounds and psychosis subtypes, and investigate longitudinal outcomes in individuals with overlapping neural signatures. The present findings underscore the need for nuanced models in both psychedelic medicine and psychiatric research, moving away from simplistic analogies toward precision neuroscience.
For policymakers, funders, and clinical leaders, the challenge will be to integrate these insights into trial protocols, risk assessment frameworks, and public communication, ensuring that both the promise and the limits of psychedelic research are clearly conveyed. As the field advances, ongoing collaboration between neuroscience, psychiatry, and regulatory science will be essential to maximize benefit while minimizing harm.
By Dr. Alex Morgan, PhD (Neuroscience). Reviewed by Dr. Emily Tran, MD, November 2026. Research based on primary analysis of OpenAlex study W7236406069 and direct review of classifier methodology.
Get tomorrow's briefing in your inbox
Policy, research, and regulatory signal — delivered on our publish cadence.