Serotonergic psychedelics temporarily modify information transfer in humans

This study (n=10) investigated the brain’s directed functional connectivity (FC) under the influence of ayahuasca, and found that neural hierarchies were temporarily disrupted with decreased top-down control and increased bottom-up information transfer.

Abstract

“Background: Psychedelics induce intense modifications in the sensorium, the sense of “self,” and the experience of reality. Despite advances in our understanding of the molecular and cellular level mechanisms of these drugs, knowledge of their actions on global brain dynamics is still incomplete. Recent imaging studies have found changes in functional coupling between frontal and parietal brain structures, suggesting a modification in information flow between brain regions during acute effects. Methods: Here we assessed the psychedelic-induced changes in directionality of information flow during the acute effects of a psychedelic in humans. We measured modifications in connectivity of brain oscillations using transfer entropy, a nonlinear measure of directed functional connectivity based on information theory. Ten healthy male volunteers with prior experience with psychedelics participated in 2 experimental sessions. They received a placebo or a dose of ayahuasca, a psychedelic preparation containing the serotonergic 5-HT2A agonist N,N-dimethyltryptamine. Results: The analysis showed significant changes in the coupling of brain oscillations between anterior and posterior recording sites. Transfer entropy analysis showed that frontal sources decreased their influence over central, parietal, and occipital sites. Conversely, sources in posterior locations increased their influence over signals measured at anterior locations. Exploratory correlations found that anterior-to-posterior transfer entropy decreases were correlated with the intensity of subjective effects, while the imbalance between anterior-to-posterior and posterior-to-anterior transfer entropy correlated with the degree of incapacitation experienced. Conclusions: These results suggest that psychedelics induce a temporary disruption of neural hierarchies by reducing top-down control and increasing bottom-up information transfer in the human brain.”

Authors: Joan F. Alonso, Sergio Romero, Miquel A. Mañanas & Jordi Riba

Summary

Psychedelics induce intense modifications in the sense of self and the experience of reality. Recent imaging studies suggest a modification in information flow between brain regions.

Introduction

Growing interest in psychedelics as models of disease and potential therapeutic agents in psychiatry has stimulated new research into their neural mechanisms. Ayahuasca is one of the most widely used psychedelics, with effects that last 1.5 to 2 hours.

Recent studies have shown that the medial aspects of the frontal cortex and several regions in the parietal lobes are targeted by psychedelics, and that nuclear medicine techniques show few or no effects at more posterior regions.

Drug-induced changes in parieto-occipital areas have been evidenced using electrophysiological measures. MRI studies have shown that psilocybin decreases functional coupling between prefrontal and parietal cortical midline regions, including the anterior and posterior cingulate cortices.

We hypothesized that psychedelics modify neural hierarchies and the flow of information around the brain. We studied directed functional connectivity of brain oscillations using transfer entropy.

Materials and Methods

Participants and Drug Administration

We recruited 10 healthy male volunteers with previous experience in psychedelic drug use, who underwent a medical examination, gave written informed consent, and participated in the study.

The study was conducted according to a double-blind, randomized, balanced crossover design. Participants received an oral dose of ayahuasca equivalent to 0.75 mg DMT/kg body weight in a double-blind randomized fashion.

EEG Recording and Processing

Three-minute EEG recordings with eyes closed were obtained from 19 standard scalp leads before, 15, 30, and 45 minutes, and 1, 1.5, 2, 2.5, 3, and 4 hours after drug administration.

Prior to parameter calculation, EEG artifacts were minimized and rejected using blind source separation and a type-II Chebyshev filter.

TE Calculation

Following artifact reduction, TE was calculated to measure the amount of uncertainty reduced in the future values of x by taking into account the past values of y.

TE was estimated using a nonparametric methodology based on equiquantal binning, with 6000 data points. This procedure is recommended when sufficient data points are available.

Subjective Effect Measures and Blood Samples

Subjective effects were measured using a visual analog scale and were rated at 15, 30, 45, 1, 1.5, 2, 2.5, 3, and 4 hours after dosing.

Volunteers were also asked to answer 2 questionnaires measuring psychedelic-induced subjective effects, including the Hallucinogen Rating Scale (HRS), which includes 6 subscales.

The Altered States of Consciousness Questionnaire is a Spanish version of the APZ and includes 3 subscales: oceanic boundlessness, dread of ego-dissolution, and visionary restructuralization.

DMT plasma concentrations were determined immediately prior to drug administration and at 0.5, 1, 1.5, 2, 2.5, 3, and 4 hours thereafter.

Statistical Analysis

Ayahuasca induced changes in TE in 342 connections. The difference in TE between placebo and ayahuasca was calculated.

We used two-sided Wilcoxon signed-rank tests to search for differences at the different time-points. The results were graphically presented using schematic representations of electrodes on the scalp, and the significance of the functional disintegration was assessed using an omnibus statistic. The statistic was calculated using a binomial distribution under the null hypothesis of no changes anywhere, and significant changes were only reported when their number exceeded a threshold.

TE (transmission efficiency) was calculated for each electrode and 2 additional directionality maps were drawn to show sources and sinks. These maps were calculated only if the prior connection maps were considered significant.

Results were compared between placebo and ayahuasca using Student’s t tests, and correlations between TE changes and subjective data and DMT plasma levels were explored using Pearson’s correlation coefficient.

Results

EEG Signals, TE

Ayahuasca induced significant TE changes in many electrode pairs at 1.5, 2, and 2.5 hours following drug administration.

The directionality of change in TE following ayahuasca administration was shown in the figure below the connection maps. The pattern of anterior-to-posterior TE decreases and posterior-to-anterior TE increases was observed at all time points between 45 minutes and 4 hours following drug administration.

Subjective Effects and DMT Plasma Levels

Ayahuasca induced statistically significant increases in all 6 subscales of the HRS and APZ questionnaires, while nonsignificant elevations were observed in the OSE and AIA subscales.

Correlation Analysis

We calculated mean TE values for electrode pairs that showed statistically significant differences from placebo and found that TE decreases were significantly correlated with scores on the HRS and APZ questionnaires. No correlations were found for TE increases.

Data Reanalysis Using Granger Causality

To assess the robustness of the TE findings, we conducted a reanalysis of the data using Granger Causality (GC). The results show that there were significant increases in postero-anterior connections, although the number of significant connections varied slightly between model order.

Discussion

The directed functional connectivity analysis showed that ayahuasca altered brain dynamics, as measured by TE. This is different from other connectivity measures such as coherence or mutual information, which do not take information flow from other regions into account.

Ayahuasca acutely decreased anterior-to-posterior information transfer and increased posterior-to-anterior information transfer, which suggests that psychedelics transiently disrupt neural hierarchies by modifying information transfer between brain regions.

The present results combine high temporal resolution neurophysiological measures with directionality information provided by TE and add to anatomical data available from neuroimaging techniques. They highlight increased blood perfusion and glucose metabolism in several frontomedial regions after acute ayahuasca administration.

Despite converging evidence from the various radiotracer studies, analysis of drug-induced changes in brain oscillations has yielded only partially overlapping results. Here, we replicate the results of previous studies and find similar results.

Ayahuasca causes paradoxical effects involving energy decreases in EEG and MEG and simultaneously enhanced visual activity. These effects can be explained by increased activation of areas involved in visual processing and cognitive-emotional processing.

In the psychedelic state, the functional coupling of EEG signals is altered, and predominantly so along the anterior-to-posterior axis. This is in line with the partial findings of more classical techniques, which show that psychedelics temporarily modify the dynamics of the interaction between the higher order frontal regions and the more sensory-selective posterior areas.

Using time-evolutionary modeling, we found that the hierarchies governing the flow of information observed after placebo are transiently disrupted by psychedelics, and that the frontal cortex is at the top of the hierarchy, sending backward projections that modulate incoming information according to a series of constraints.

When top-down constraints are reduced and sensory excitability is increased, weak endogenous activity or “system noise” in visual and auditory cortices can reach higher levels in the hierarchy, explaining visual phenomena commonly reported for psychedelics. In the study by Muthukumuraswarmy and coworkers (2013), increased excitability of deep layer pyramidal neurons can be explained by increased 5HT 2A receptors, which fits comfortably with our findings in relation to TE. This may explain the overwhelming nature of the experience.

If this disruption of hierarchies were to linger after discontinuation of the drug, problems could emerge in the normal processing of sensory information, which could explain some of the adverse consequences associated with psychedelic drug use.

Acute administration of psychedelics disrupted the functional connectivity of brain oscillations in humans, which may explain the profound modifications of perception, cognitive processes, and experience of reality typically induced by these drugs.