LSD reconfigs cortex thru faster brain rhythms and increased fractal dimension

Using source-resolved magnetoencephalography, explicit spectral parameterization, temporal-complexity metrics, and interpretable machine learning in a placebo-controlled design with and without music, this work maps how LSD reconfigures cortical dynamics. Key findings are robust increases in alpha- and beta-band peak frequencies co-occurring with genuine reductions in oscillatory power, plus a flattening of the aperiodic 1/f spectral slope. Temporal-complexity measures rise under LSD: Lempel-Ziv complexity increases and fractal structure (Higuchi fractal dimension) becomes more pronounced, indicating richer and more scale-free neural activity. These effects are spatially structured, preferentially affecting sensory, language, emotion, and imagery-related networks while largely sparing motor cortex. Explicitly accounting for peak-frequency shifts rules out misattributing frequency migration to mere desynchronization.

Interpretable machine-learning analyses show that peak-frequency shifts, aperiodic parameters, and complexity metrics are the strongest discriminators of the psychedelic state, suggesting these features form a distinct electrophysiological signature of LSD that differs from other serotonergic psychedelics. Music does not amplify these neural signatures and trends toward attenuating them. Overall, the findings provide a detailed electrophysiological account of how LSD speeds cortical rhythms and increases temporal complexity, offering concrete biomarkers for differentiating drug states and for linking altered dynamics to perceptual and affective phenomenology.