Decreased directed functional connectivity in the psychedelic state

Introduction

Barnett and colleagues frame the study within ongoing efforts to understand how classical psychedelics (notably LSD and psilocybin) and sub‑anesthetic ketamine alter large‑scale neural dynamics underlying changes in consciousness. Previous work has linked subjective psychedelic effects to receptor‑level actions (for example 5HT2A partial agonism for classical psychedelics and NMDA antagonism for ketamine) and to reorganisations of functional connectivity (FC) measured with fMRI, but the authors note important limitations in fMRI for assessing fast temporal dynamics. They argue that high‑temporal‑resolution electrophysiological recordings (MEG/EEG) permit simultaneous comparison of undirected FC measures (e.g. correlation, coherence, mutual information) and directed FC measures (e.g. Granger causality, transfer entropy), which may yield complementary insights into shared information versus information flow across brain regions. The study therefore set out to characterise changes in both directed and undirected FC in the psychedelic state using source‑localised steady‑state MEG recordings previously collected from healthy volunteers after controlled administration of LSD, psilocybin, and low‑dose ketamine, with tiagabine (a non‑psychedelic GABA reuptake inhibitor) included as an additional comparator. Analyses were exploratory and performed on drug versus placebo recordings within subjects, with functional connectivity estimated at the level of 90 AAL cortical sources and aggregated into larger ROIs where appropriate. The central question was whether psychedelics produce reliable changes in directed and undirected FC relative to placebo and relative to the non‑psychedelic control, and whether directed and undirected measures move together or independently in these states. This work is positioned as an exploratory, cross‑drug comparison using high‑temporal‑resolution MEG to probe information flow in the psychedelic state and to evaluate whether changes in directed FC can be dissociated from changes in undirected FC and from simple confounds such as spectral power or signal‑to‑noise ratio (SNR).