Dynamic Neural State Transitions Predict Psychedelic Phenomenal Richness: Magnitude, Not Direction, Drives Subjective Intensity

AUTHOR NOTE — files restricted July 2026 The files for this record have been restricted by the author following an internal audit. The analysis reported here should not be used or cited. Four defects were identified: The predictor described as "rate of CGI change" is in fact the single largest window-to-window jump (a maximum), which the paper's own Table 2 labels "Max rate of change." The outcome described as "intensity" is in fact phenomenal richness, a specific 5D-ASC composite. The Methods describe ρ as spectral entropy and φ as theta-band (4–8 Hz) coherence. The code that produced the results computes neither: ρ is Lempel-Ziv complexity of the Hilbert amplitude envelope, and φ is broadband (1–45 Hz) coherence. The reported result cannot be reproduced from the Methods as written. The paper combines two incompatible definitions of ρ in one document. Table 1's descriptive statistics use spectral-entropy ρ, under which ρ decreases significantly under DMT (t = 3.21, p = 0.003). The headline r = 0.378 correlation uses Lempel-Ziv ρ, under which ρ does not change significantly under DMT (t = 0.87, p = 0.39). The paper's "CGI" is therefore two physically different quantities in different places, disagreeing on the paper's own central premise.