A coordinated study across five laboratories at the University of California, Berkeley, UCSF and Stanford University found no reproducible evidence that a single dose of psilocybin produces lasting behavioral improvements in mice. The findings were published in Nature Neuroscience on November 2, 2026.
The researchers used harmonized protocols, identical mouse strains and a dose of 2 milligrams per kilogram administered intraperitoneally. The experiments assessed locomotion, avoidance, social behavior, fear learning and stress-coping measures, using approximately 200 male and female mice per experiment across the participating laboratories.
Psilocybin produced consistent acute effects. Shortly after administration, mice displayed the characteristic head-twitch response, a temporary increase in avoidance-related behavior and reduced expression of learned fear. These effects were replicated across the study sites.
The results differed when the animals were tested 24 hours later. The pooled data showed no statistically reproducible reductions in anxiety- or depression-like behaviors, no changes in social preference or social reward learning, and no consistent improvement in fear extinction or fear retention. The consortium also found no evidence that psilocybin reopened a proposed developmental period for social reward learning in adult mice.
Results from individual laboratories sometimes differed in statistical significance. In one example involving a standard behavioral despair test, two laboratories recorded significant improvements while three did not. The researchers said that relying on only the positive results could contribute to publication bias, particularly when studies use small sample sizes.
The study’s authors also highlighted the limitations of translating rodent behavioral tests into human depression research. Measures such as immobility in a forced swim test cannot fully represent the psychological complexity of human depression or the subjective experiences associated with guided psychedelic treatment.
The researchers noted that human psychedelic experiences are influenced by factors including expectations, emotional state and social environment. They also pointed to the difficulty of representing these elements in isolated animal experiments, as well as the differences between traditional communal uses of psychedelics and reductionist laboratory models.
According to the study, variations in factors such as animal sourcing, handler sex, habituation procedures, enclosure size and testing time did not explain the lack of persistent effects in the combined results. The authors described the findings as a recalibration for basic neuroscience rather than evidence against psilocybin’s potential therapeutic value in humans.
The consortium argues that identifying reliably reproducible effects, including the acute changes in fear expression, can help researchers focus on the neural circuits associated with those responses and improve the design of future psychedelic studies.
Sources: Neuroscience News.