Ritualistic use of ayahuasca enhances a shared functional connectome identity with others

This pre-print fMRI study (n=21) of people who regularly use ayahuasca within the Santo Daime church finds that during the experience, changes in functional connectivity (FC, how brain areas communicate) indicate on the scans more similarity between them (so less ‘unique’ FC). The authors use an analogy that each person still has their own hairstyle, but instead of wearing different colour t-shirts, all are now wearing the same shirt.

Abstract

“The knowledge that brain functional connectomes are both unique and reliable has enabled behaviourally relevant inferences at a subject level. However, it is unknown whether such fingerprints persist under altered states of consciousness. Ayahuasca is a potent serotonergic psychedelic which elicits a widespread dysregulation of functional connectivity. Used communally in religious ceremonies, its shared use may highlight relevant novel interactions between mental state and FC inherency. Using 7T fMRI, we assessed resting-state static and dynamic FCs for 21 Santo Daime members after collective ayahuasca intake in an acute, within-subject study. Here, connectome fingerprinting revealed a shared functional space, accompanied by a spatiotemporal reallocation of keypoint edges. Importantly, we show that interindividual differences in higher-order FCs motifs are relevant to experiential phenotypes, given that they can predict perceptual drug effects. Collectively, our findings offer an example as to how individualised connectivity markers can be used to trace a subjects functional connectome across altered states of consciousness.”

Authors: Pablo Mallaroni, Natasha L. Mason, Lilian Kloft, Johannes T. Reckweg, Kim van Oorsouw, Stefan W. Toennes, Hanna M. Tolle, Enrico Amico & Johannes G. Ramaekers

Summary of Ritualistic use of ayahuasca…

Introduction

Functional connectivity patterns derived from fMRI data are stable across a lifetime and can be used to predict complex behavioural phenotypes such as cognition, demographics, traits such as fluid intelligence, and even clinical outcomes.

Recent work has demonstrated that an individual’s connectome fingerprint across sessions can be separated into signalling motifs reflecting both trait intra-subject and state-dependent inter-subject variance. These fingerprints may also reflect the temporal quality of functional network architecture.

Much work has been done to understand how the brain’s inherent connectivity may be differentially altered according to a particular individual or mental state, but there is little evidence bridging these lines of research. Altered states of consciousness may provide a new means to probe the interdependency of unique spontaneous brain activity and functional brain organisation.

The religious use of ayahuasca, a psychedelic brew produced by a combination of two different plant sources, might provide a useful means by which to investigate the orthogonality between trait and state FC under conditions in which an individual transitions from a normal, waking state of consciousness to a shared altered state.

Here, we sought to understand how ayahuasca brew consumption might alter a subject’s functional connectivity, and how this might explain aspects of their subjective experience.

Results

Experienced members of Santo Daime were enrolled in a fixed-order, within-subject, observational study that included resting-state fMRI, pharmacokinetic sampling, and questionnaires pertaining to retrospective drug effects and aspects of “work” during resting-state.

Acute effects of Ayahuasca

Ayahuasca intake was associated with increased ratings on all (sub)dimensions of the 5D-ASC and the EDI, and serum concentrations of DMT were significantly greater than zero at both 60 and 160 minutes after intake.

Quantifying whole-brain fingerprints

Connectome fingerprinting provides a window into the “uniqueness” of one’s functional connectivity by measuring the difference between each subject’s FC self-similarity against the other subjects’ FCs.

The differential identifiability of each participant was significantly diminished under ayahuasca, and this effect was driven by a significantly increased Iothers score. However, subjects continued to retain high Iself and SR scores.

We examined how dissimilar a subject’s fingerprint might be under ayahuasca. We found that their fingerprint makeup is reconstituted, with a greater dissimilarity between a subject’s functional connectome under Ayahuasca and baseline.

Dynamic identifiability increased steadily with longer window lengths, and was significantly reduced under Ayahuasca in a temporally selective manner. Iothers increased across select frames, and Iself remained stable at all timescales.

We observed global reductions in ICC scores under Ayahuasca, but individual RSNs had varying levels of importance for fingerprints and may be differentially affected.

We investigated how edges important to normative connectome fingerprinting might shift in importance under Ayahuasca. We found that edges normally driving a subject’s identifiability are no longer significant contributors, and are instead replaced by edges pertinent to the DMN.

We examined the spatial ICC patterns of RSN fingerprints as a function of time under ayahuasca, and found global reductions in dFC stability across all measured timescales.

Network-based analyses revealed diffuse changes to the stability of dynamic functional connectivity under ayahuasca. These changes were primarily ascribed to edges involved in between-network SM and VIS connectivity, and to previously identified static increases in SM-L connectivity stability.

We next asked whether altered fingerprint dynamics under ayahuasca could also be reflected at a regional level of brain organisation. We found that the temporal gradient of ICC maxima was inverted following intake.

Connectome fingerprints are predictive of perceptual drug effects.

We built an iterative multilinear model approach to predict subjective experience of ayahuasca using PCA components of subsets of edges. This model performed best for Visual Restructuralisation and Auditory Alterations.

The model for VR and AA was improved by including PCA3, which was found to be the most predictive edge. This finding is consistent with previous literature indicating that sensorimotor networks exhibit lower inter-subject functional connectivity variability than associative networks 45.

Additional control analyses

We performed a series of quality controls on our primary identifiability analysis, and found that our findings are robust to motion and replicable across different denoising strategies.

Discussion

The authors used connectome fingerprinting to study the functional brain connectivity of 21 Santo Daime members taking part in the religious use of ayahuasca. We found that the inherent features of a subject’s functional connectome might transition into a collective altered state of consciousness.

The collective use of ayahuasca yields shared functional traits

While the Idiff of everyone’s connectome was diminished under ayahuasca, this reduction could be attributed to the increased contribution of Iothers, suggesting that a subject’s brain fingerprint holds a larger number of shared functional traits under ayahuasca, diminishing its overall differentiability.

The inter-subject variability of a sample is diminished when engaging in a task battery, proportionally to cognitive load 14,31. This suggests that shared functional traits may reside in previously identified timeframes at which complex cognition emerges 25.

Constituents of connectome self-identity are mutable under ayahuasca

Studies have repeatedly demonstrated the remarkable consistency of inherent functional connectivity patterns across participants and mental states. Ayahuasca does not affect this consistency, and instead produces a general functional reconfiguration of inherent signalling traits.

Local shifts in functional connectivity stability drive altered connectome fingerprints

We observed global reductions in edge stability at all measured timescales under ayahuasca, striking proxy of the patterns of functional change under psychedelics, and increased signal complexity, which may consequently limit the temporal concordance of edge pairs.

Under ayahuasca, the importance of a subset of 250 edges to maximally define a subject’s fingerprint dropped, and the frontotemporal DMN nodes emerged as the focal point for a subject’s identifiability, expressing greater stability.

The temporal organization of brain activity under ayahuasca was similar to that observed under LSD and psilocybin, with unimodal regions (parietal operculum, visual cortex) peaking in stability at longer timescales and transmodal regions (prefrontal cortex, posterior cingulate cortex) exhibiting longer, integrative firing patterns maximal at shorter timescales.

Connectome fingerprints are relevant to the subjective ayahuasca experience

To explore the hypothesis that subject-level shifts in functional connectivity might help predict overlaying subjective experiences, we devised a data driven PCA approach. In practice, PCA offers the opportunity to separate subject-level and group-level functional connectivity information carried in low and high-variance components. Higher-order PC deviations capturing individual variation in FC were most relevant to the visual and auditory effects of ayahuasca.

Limitations

The present work comes with several limitations, including the fact that Santo Daime members are not reflective of the general population and that 5-HT2A agonists are potent psychoplastogens.

The study of inter-individual differences in functional connectivity is subject to confounding effects regarding dosage, blinding, sample inclusion criteria and expectancy. Future studies should aim to replicate this workflow using framewise approaches such as dynamic conditional correlations or phase coherence estimation. Per prior work 25, preprocessing was performed with a pipeline comprising Global Signal regression (GSR). However, the suitability of GSR alongside other preprocessing steps is widely debated.

Conclusion

The ritualistic use of ayahuasca produces a shared functional space, marked by a spatiotemporal reconfiguration of brain connectivity traits. This blurred connectome fingerprint may be caused by the synergy of a psychoactive sacrament with the interpersonal dynamics of ritualism.

Participants

Twenty-four volunteers were enrolled in a within-subject, fixed-order observational study. Three volunteers were excluded from analyses due to excessive head motion leaving a final sample of 21 subjects (10 females) of ages 29 to 64 (M: 54.48, SD: 10.55).

Study procedures

Participants self-administered ayahuasca equivalent to their usual dose, and were tested at the same window of time as to minimise diurnal variation.

A urine drug screen, breath alcohol test, pregnancy test, and 1h MRI scanning session were performed on each subject. They were also given a 5-D-ASC scale and Ego Dissolution Inventory 360 minutes after drug intake to assess the subjective experience after drug intake.

Participants underwent fMRI, structural MRI, and proton MRS at various times following treatment.

T1-weighted anatomical images were acquired using a magnetisation-prepared 2 rapid acquisition gradient-echo (MP2RAGE) sequence, followed by 500 whole brain echo planar imaging (EPI) volumes acquired at rest.

MP2RAGE images were first denoised, bias-field corrected, skullstripped, segmented, warped in functional space, normalized, demeaning, linear detrended, and regression of 18 signals were performed on BOLD fMRI data. A bandpass first-order Butterworth filter was applied to all BOLD timeseries at the voxel level. We kept track of the fMRI volumes that were highly influenced by head motion, using three different metrics as a scrubbing index: frame displacement, DVARS, and SD. No volume censoring was performed using these metrics, and functional connectomes were highly similar between and within conditions.

A 2mm cortical Schaefer parcellation was projected into each subject’s T1 space and subsequently their native EPI space. FSL boundary-based-registration was also applied to improve the registration of the structural masks and parcellation to the functional volumes.

We devised two separate workflows for functional connectivity assessment: a static functional connectivity matrix was calculated for each pair of ROIs, and a dynamic functional connectivity matrix was assessed by balancing the number of time points for a stable dFC computation.

We split each scan into two corresponding halves, and then take the upper triangle of each matrix to extract unique elements. These elements can then be compared using Pearson correlation.

A static FCs identifier matrix A has dimensions N2 and the average of the main diagonal elements is Iself, while the average of the off-diagonal elements is Iothers. The differential identifier is Idiff.

We calculated the difference between the average within-subject FCs similarity and the average between-subject FCs similarity, and also calculated the distance of each participant fingerprint under Ayahuasca from their respective normative state from their baseline. We also measured the Success-rate of the identification procedure.

We can extend this principle to dynamic functional connectomes (dFC) by calculating each measure across each dynamic frame of connectivity. The resulting dynamic identifiability matrix is a block diagonal matrix.

When a subject is other than a particular subject, the dynamic Iothers can be defined as the sum of the dynamic Iothers for all subjects other than the particular subject.

Edgewise connectome identifiability

To understand which edges were key contributors to changes in connectome reliability, we quantified the edge-wise reliability of individual connectomes using intraclass correlation analysis as a reference.

We assumed that highly synchronous, or stable, edges between test and control regions were major determinants of each state’s connectome frequency. We then calculated the ICC of the edges between test and control regions.

We calculated the ICC estimates for each timescale and then sorted the functional connectivity frames based on their similarity, and then recalculated the ICC estimates for each timescale.

We sorted static edges according to their thresholded ICC values and added edges in a descending fashion, recalculating Idiff at each iteration of 50 edges.

PCA is an unsupervised exploratory approach that is typically used for dimensionality reduction and pattern recognition. A MLR was built using three PCA components ranked according to explained variance for each subjective effect measure, alongside two covariates: singing and scrubbing. We strengthened the reliability of our model using k-fold cross-validation106, where the Spearman’s correlation coefficient between predicted and actual inventory values was calculated. This performance score was assessed against surrogate models.