This study aggregated a large preclinical dataset (N=693 mice) to examine behavioural responses to a single high dose of psilocybin across multiple laboratories and experimental conditions. Behavioral endpoints were divided into acute measures—head twitch responding and locomotion—and post-acute measures that are commonly used to index antidepressant-like or anxiolytic-like effects: sucrose preference, the elevated plus maze, and novelty-suppressed feeding. The authors used data-driven analyses to probe how biological factors (sex, age, strain) and experimental factors (stress protocol, 5-HT1B signaling manipulations) influenced outcomes.
Across the combined dataset, psilocybin produced a consistent and robust acute behavioral signature, as assessed by head twitch responding and locomotor activity. The acute response to a single high dose of psilocybin was more reproducible than later, post-acute endpoints in this pooled sample. This robust acute effect contrasts with the greater heterogeneity observed in assays conducted after the acute phase.
Post-acute behavioral measures showed a variable profile across the dataset. Among the post-acute outcomes evaluated, sucrose preference exhibited the most robust and consistent signal associated with psilocybin treatment. Other post-acute readouts—elevated plus maze and novelty-suppressed feeding—were more inconsistent across the sample. Overall, the post-acute behavioral profile following psilocybin was heterogeneous, with effect size and direction depending on additional biological and experimental factors.
The dataset was examined for systematic influences of sex, age, and mouse strain, as well as the impact of different stress protocols. The analyses identified sex and the specific stress model used as significant modulators of post-acute antidepressant-like and anxiolytic-like behavioral outcomes following psilocybin. Age and strain were included among the variables analyzed; the abstract reports that these biological and experimental factors modulated behavioral readouts, but detailed effect estimates, subgroup sizes, and statistical values are not reported in the source abstract.
Modulation of 5-HT1B receptor (5-HT1BR) signaling emerged as a key experimental factor influencing post-acute behavioral outcomes. Activation of 5-HT1BR was specifically mentioned as a significant modulator of the antidepressant-like and anxiolytic-like effects that were assessed after the acute phase. The abstract does not provide mechanistic detail or quantitative effect sizes for 5-HT1BR manipulations; those details may be available in the full manuscript or supplementary materials.
To assess whether post-acute behavioral outcomes together with biological and experimental variables could reliably indicate drug treatment, the authors applied a random forest classification approach. The model was only narrowly able to predict psilocybin treatment based on the available post-acute behavioral data and covariates, indicating that, despite some consistent signals (notably in sucrose preference), substantial heterogeneity remains. This limited predictive performance underscores the influence of multiple interacting factors on post-acute behavioral readouts in mice.
The study's central interpretation is that heterogeneity in mouse behavioral responses to psilocybin reflects meaningful biological and experimental variables rather than purely random noise. Accounting for factors such as sex, stress paradigm, and 5-HT1BR activation could improve reproducibility across laboratories and help identify conditions under which therapeutic-like effects of psilocybin are most robust. The authors suggest that acknowledging and systematically testing these modulators can support more reproducible preclinical models that are better suited to inform neural mechanisms underlying persisting behavioral effects of psilocybin and to enhance translational relevance for potential clinical applications.
Notes and limitations
This report is a preprint and has not undergone peer review. The abstract summarizes the dataset and primary findings but does not include detailed statistics, subgroup sample sizes, or full methodological parameters; those may be provided in the full text or supplementary materials. Funders disclosed in the source include Dartmouth College and the International Center for Responsible Gaming.