Skepticism about Recent Evidence That Psilocybin “Liberates” Depressed Minds

A critique of a Nature Medicine paper claiming psilocybin therapy reduces brain network modularity in depressed patients, an effect not seen with the SSRI S-citalopram. The authors identify multiple problems: inconsistent reporting of the primary clinical outcome, statistical flaws including a one-tailed test and a nonsignificant interaction, regression to the mean, ambiguous interpretation of resting-state fMRI data, and a missing reference to a similar study that undermines the justification for a one-tailed test. These issues cast doubt on the uniqueness and impact of the original findings and the media hype they generated.

Study at a glance

Abstract

A recent paper in Nature Medicine found that psilocybin therapy in patients with depression decreased brain network modularity (measured with task-free functional magnetic resonance imaging), an effect supposedly not found with the selective serotonin reuptake inhibitor S-citalopram. This decrease in network modularity also correlated with depression. Here, we raise several issues with this paper, including inconsistencies in reports of the primary clinical outcome, statistical flaws including a one-tailed test, nonsignificant interaction, and regression to the mean, the ambiguity and overinterpretation of "resting state" data, and a missing reference for a conceptually similar study that exemplifies why a one-tailed test cannot be justified. Together, these issues make us question the uniqueness and impact of these findings, as well as the unwarranted media hype that they generated.