Psilocybin desynchronizes the human brain - PubMed (original) (raw)

. 2024 Aug;632(8023):131-138.

doi: 10.1038/s41586-024-07624-5. Epub 2024 Jul 17.

Subha Subramanian 2, Demetrius Perry 3, Benjamin P Kay 4, Evan M Gordon 5, Timothy O Laumann 3, T Rick Reneau 5, Nicholas V Metcalf 4, Ravi V Chacko 6, Caterina Gratton 7, Christine Horan 8, Samuel R Krimmel 4, Joshua S Shimony 5, Julie A Schweiger 3, Dean F Wong 5, David A Bender 3, Kristen M Scheidter 4, Forrest I Whiting 4, Jonah A Padawer-Curry 9, Russell T Shinohara 10 11 12, Yong Chen 12, Julia Moser 13 14, Essa Yacoub 15, Steven M Nelson 13 16, Luca Vizioli 15, Damien A Fair 13 14 15 16, Eric J Lenze 3, Robin Carhart-Harris 17 18, Charles L Raison 19 20, Marcus E Raichle 4 5 9 21 22, Abraham Z Snyder 4 5, Ginger E Nicol # 3, Nico U F Dosenbach # 4 5 9 21 23

Affiliations

Psilocybin desynchronizes the human brain

Joshua S Siegel et al. Nature. 2024 Aug.

Abstract

A single dose of psilocybin, a psychedelic that acutely causes distortions of space-time perception and ego dissolution, produces rapid and persistent therapeutic effects in human clinical trials1-4. In animal models, psilocybin induces neuroplasticity in cortex and hippocampus5-8. It remains unclear how human brain network changes relate to subjective and lasting effects of psychedelics. Here we tracked individual-specific brain changes with longitudinal precision functional mapping (roughly 18 magnetic resonance imaging visits per participant). Healthy adults were tracked before, during and for 3 weeks after high-dose psilocybin (25 mg) and methylphenidate (40 mg), and brought back for an additional psilocybin dose 6-12 months later. Psilocybin massively disrupted functional connectivity (FC) in cortex and subcortex, acutely causing more than threefold greater change than methylphenidate. These FC changes were driven by brain desynchronization across spatial scales (areal, global), which dissolved network distinctions by reducing correlations within and anticorrelations between networks. Psilocybin-driven FC changes were strongest in the default mode network, which is connected to the anterior hippocampus and is thought to create our sense of space, time and self. Individual differences in FC changes were strongly linked to the subjective psychedelic experience. Performing a perceptual task reduced psilocybin-driven FC changes. Psilocybin caused persistent decrease in FC between the anterior hippocampus and default mode network, lasting for weeks. Persistent reduction of hippocampal-default mode network connectivity may represent a neuroanatomical and mechanistic correlate of the proplasticity and therapeutic effects of psychedelics.

© 2024. The Author(s).

PubMed Disclaimer

Conflict of interest statement

Within the past year, J. S. Siegel has been an employee of Sumitomo Pharma America and received consulting fees from Longitude Capital. J. S. Siegel, N.U.F.D., T.O.L. and E.M.G. have submitted a provisional patent (patent no. 020949/US 15060-1787) for the use of precision functional mapping for measuring target engagement by experimental therapeutics. R.T.S. has received consulting compensation from Octave Bioscience and compensation for reviewership duties from the American Medical Association. C.L.R. serves as a consultant to Usona Institute and Novartis and receives research support from the Tiny Blue Dot Foundation. G.E.N. has received research support from Usona Institute (drug only). She has served as a paid consultant for Carelon, Alkermes, Inc., Sunovion Pharmaceuticals, Inc. and Novartis Pharmaceuticals Corp. T.O.L. holds a patent for taskless mapping of brain activity licenced to Sora Neurosciences and a patent for optimizing targets for neuromodulation, implant localization and ablation is pending. J. S. Siegel is a consultant and received stock options in Sora Neuroscience, and company that focuses on resting-state analysis. D.A.F. and N.U.F.D. are cofounders of Turing Medical Inc, have financial interest, may benefit financially if the company is successful in marketing FIRMM motion monitoring software products, may receive royalty income based on FIRMM technology developed at WUSOM and licenced to Turing Medical Inc. S.M.N., E.M.G. and T.O.L. have received consulting fees from Turing Medical Inc. D.F.W. is a consultant for Engrail Therapeutics and receives contract funds for WUSOM research studies from Eisai, Anavex and Roche. These potential conflicts of interest have been reviewed and are managed by WUSOM. The other authors declare no competing interests. All authors report no financial interest in psychedelics companies.

Figures

Fig. 1

Fig. 1. Acute psilocybin effects on functional brain organization.

FC change (Euclidean distance) was calculated across the cortex and subcortical structures. Effects of drug condition were tested with an LME model in n = 6 longitudinally sampled participants over ten sessions with psilocybin and six sessions with methylphenidate (MTP) (a and b are thresholded at P < 0.05 based on permutation testing with TFCE; see unthresholded statistical maps in Extended Data Fig. 2). **a**, Psilocybin-associated FC change, including in subcortex. a, anterior; p, posterior; L, left; R, right. **b**, MTP-associated FC change. **c**, Typical day-to-day variability as a control to the drug conditions (unthresholded: not included in LME model). **d**, Average FC change in individual-defined networks. Open circles represent individual participants. FC change is larger in DMN than other networks. Rotation-based null model (spin test,): ten psilocybin doses, 1,000 permutations, one-sided _P_spin < 0.001, (_P_spin > 0.05 for all other networks). **P < 0.001, uncorrected. e, Whole-brain FC change (Euclidean distance from baseline) for all rest scans across conditions. FC change for MTP, psilocybin and day-to-day are in comparison to same-participant baseline. White dots indicate median, vertical lines indicate quartiles. LME model predicting whole-brain FC change: ten psilocybin doses (275 observations), estimate (95% CI) = 15.83 (13.50, 18.15), t(266) = 13.39, P = 1.36 × 10−31, uncorrected. For the full FC distance matrix with session labels, see Extended Data Fig. 3. f,g, Comparison of the differences in FC change to differences in psychedelic experiences. f, Individual FC change maps and MEQ30 scores for two exemplars (see Extended Data Fig. 4 for all drug sessions). g, The relationship between whole-brain FC change and mystical experience rating is plotted for all drug sessions (psilocybin and MTP). The LME model demonstrated a significant relationship: 16 drug doses (ten psilocybin, six MTP), estimate (95% CI) = 69.78 (50.15, 89.41), t(13) = 7.68, P = 3.5 × 10−6, uncorrected. h, The relationship between FC change and MEQ30 (_r_2) is mapped across the cortical surface.

Fig. 2

Fig. 2. Data-driven clustering of brain network variability.

MDS blind to session labels was used to assess brain changes across conditions. a, In the scatter plots, each point represents whole-brain FC from a single 15 min scan, plotted in a multidimensional space on the basis of the similarity between scans. Dimensions 1 and 4 showed strong effects of psilocybin. The top shows scans coded on the basis of drug condition. Dark red denotes that the participant had an episode of emesis shortly after taking psilocybin. The bottom shows scans coloured on the basis of participant identity. Dimension 1 separates psilocybin from non-drug and MTP scans in most cases. See Extended Data Fig. 5 for the dimension 1–4 weight matrices. b, Visualization of dimension 1 weights. The top 1% of edges (connections) are projected onto the brain (green indicates connections that are increased by psilocybin). Cerebellar connections are included although the structure is not shown. c, Re-analysis of dimension 1 in extant datasets with intravenous psilocybin (left, ref. , paired two-sided _t_-test of change in dimension 1 score, n = 9, t(8) = 2.97, P = 0.0177, uncorrected) and LSD (right, ref. , paired two-sided _t_-test: n = 16, t(15) = 4.58, P = 3.63 × 10−4, uncorrected). *P < 0.05, **P < 0.001, uncorrected. d, Average effects of psilocybin on network FC, shown separately for within-network integration (left) and between network segregation (right). For network integration (left), blue indicates a loss of FC (correlations) between regions within the same network. For network segregation (right), blue indicates a loss of FC (anticorrelations) to all other regions in different networks; see Extended Data Fig. 6 for a full correlation matrix. Dissolution of functional brain organization corresponds to decreased within-network integration and decreased between network segregation.

Fig. 3

Fig. 3. Spatial desynchronization of cortical activity during psilocybin.

a, NGSC captures the complexity of brain activity patterns. It is derived from the number of spatial principal components needed to explain the underlying structure. Higher entropy equals desynchronized activity. On the right is variance explained by subsequent principal components for psilocybin in red, MTP in blue and no drug in grey for P6. b, Whole-brain entropy (NGSC) is shown for every fMRI scan for a single participant (P6). At right, increases during psilocybin were present in all participants. Sample sizes are provided in Supplementary Table 1. Grey bars indicate condition means. c, Parcel entropy (computed on individual-specific parcels) within functional brain areas shows similar psilocybin-driven increases as whole-brain entropy. d, Psilocybin-associated spatial entropy (individual-specific parcels, averaged across participants) is visualized on the cortical surface. Psilocybin-associated increases in entropy were largest in association cortex. e, LSD-associated increases in spatial entropy were similar to those induced by psilocybin (using data from ref. ). f, Increases corresponded spatially to 5-HT2A receptor density. In b–d, n = 6 participants, 272 observations (scans). For e, n = 16 participants.

Fig. 4

Fig. 4. Effects of perceptual task performance on psilocybin-associated FC change and desynchronization.

a, Psilocybin-associated FC change from resting scans (left) and from task scans (right). b, Regional NGSC change (psilocybin minus baseline) from rest scans (left) and from task scans (right). Bar graphs on the bottom indicate the corresponding whole-brain FC change (a) and whole-brain NGSC values (b) during rest and task for baseline and drug conditions. LME models indicated an interaction of task × psilocybin on FC change (n = 7 with task data on psilocybin, estimate (95% CI) = −6.48 (−9.59, −3.37), t(265) = −6.48, P = 5.49 × 10−5, uncorrected) and an interaction of task × psilocybin on NGSC (n = 7 with task data on psilocybin, estimate (95% CI) = −0.042 (−0.056, −0.027), t(265) = −5.62, P = 4.82 × 10−8, uncorrected). Bars indicate mean and error bars indicate s.e.m.. **P < 0.001, uncorrected.

Fig. 5

Fig. 5. Persistent effects of psilocybin on hippocampal-cortical FC.

a, Hippocampus FC change maps (left hippocampus; unthresholded _t_-maps, as in Extended Data Fig. 2). Acute psilocybin FC change is shown on top and persistent FC change (3 weeks after psilocybin) on the bottom. b, Each dot represents the FC change score for the anterior hippocampus for a single scan before (left) and after (right) psilocybin for every participant (coloured as in Fig. 2). Participants showed a post-psilocybin increase in FC change in the anterior hippocampus (LME model, pre- versus post-psilocybin; n = 6 participants, 186 observations, estimate (95% CI) = 0.095 (0.032, 0.168), t(182) = 2.97, P = 0.0033, uncorrected). c, Connectivity from an anterior hippocampus seed (Montreal Neurological Institute coordinates −24, −22, −16 and 24, −18, −16) pre-psilocybin (left), post-psilocybin (middle) and persistent change (post- minus pre-) for an exemplar participant (P3). The red border on the right-most brain outlines the individual-specific DMN. A decrease in hippocampal FC with parietal and frontal components of the DMN is seen. d, Time course of anterior hippocampus minus DMN for all participants and scans (participant colours as in b). A moving average is shown in black. e, Schematic of hippocampal-cortical circuits, reproduced from ref. ,

CC BY 4.0

.

Extended Data Fig. 1

Extended Data Fig. 1. Quantifying psilocybin effects with precision functional mapping: design.

a) Schematic illustrating the study protocol of the individual-specific precision functional mapping study of acute and persistent effects of psilocybin (single dose: 25 mg). Repeated longitudinal study visits enabled high-fidelity individual brain mapping, measurement of day-to-day variance, and acclimation to the scanner. The open label replication protocol 6-12 months later included one or two scans each of baseline, psilocybin, and after drug. b) Timeline of imaging visits for 7 participants. c) Head motion comparisons across psychedelics studies,. Average head motion (FD, framewise displacement, in mm) off and on drug compared between our dataset and prior psychedelic fMRI studies. Unpaired two-sided _t_-test: n PSIL 2012 = 15, n LSD,2016 = 20, n PSIL(PFM) = 7; off drug PSIL2012-PSIL(PFM) t (274) = −4.57, P uncorr = 7.33 × 10−5; off drug LSD2016-PSIL(PFM) t (286) = −4.03, P uncorr = 7.34 × 10−6; on drug PSIL2012-PSIL(PFM) t (46) = −1.80, _P_uncorr = 0.079; on drug LSD2016-PSIL(PFM) t (88) = −0.73, P uncorr = 0.46. Dotted line at FD of 0.2 mm. Dark gray bars indicate quartiles, light gray violins indicate distribution. *P uncorr < 0.05, +P < 0.1. d) Timeline for an example participant (P1). e) Participants reported significantly higher scores on all dimensions of the mystical experience questionnaire during psilocybin (red) than placebo (40 mg methylphenidate; blue). Paired two-tailed _t_-test, n = 7; Mystical t (6) = −3.64, P uncorr = 0.011; Positive Mood t (6) = −5.44, P uncorr = 0.0016; Transcendence t (6) = −4.98, P uncorr = 0.0025; Ineffability t (6) = −2.54, P uncorr = 0.044. Error bars indicate SEM.

Extended Data Fig. 2

Extended Data Fig. 2. Unthresholded vertex-wise FC change maps.

T-statistic maps, resulting from the linear mixed effects (LME) model based on vertex-wise FC change (Euclidean distance from baseline scans) across the cortex and subcortical structures for every scan. Higher t values indicate a larger change from baseline (pre-drug) scans. Effects of drug condition (baseline, psilocybin, methylphenidate, post-psilocybin, post-methylphenidate), were modeled as fixed effects. For example, if drug 1 was psilocybin and drug 2 was methylphenidate, then scans between drug visits were labeled post-psilocybin and scans after drug 2 were labeled post-methylphenidate.

Extended Data Fig. 3

Extended Data Fig. 3. Functional connectivity (FC) distance and condition matrices for all fMRI scans.

Following Gratton et al., we compared FC matrices between rs-fMRI sessions to quantify contributors to variability in whole-brain FC. Under this approach, the effects of group, individual, session, and drug (as well as their interactions) are examined by first calculating the Euclidean distance among every pair of FC matrices (i.e., distance among the linearized upper triangles). a) In the resulting second-order distance matrix, each row and column show whole-brain FC from a single study visit. The colours in the matrix indicate distance between functional networks for a pair of visits (i.e., Euclidean distance between the linearized upper triangles of two FC matrices). Panels b and c demonstrate how the distance matrix was subdivided to compare different conditions. b) Black triangles represent distinct individuals. Replication protocol visits are listed at the end. c) Task and rest scans are shown in white and orange, respectively. Note that psilocybin scans (black arrows pointing to P1 psilocybin scans in panel a are very dissimilar to no-drug scans from the same individual (left arrow; in a) but have heightened similarity to psilocybin scans from other individuals (right arrow in a).

Extended Data Fig. 4

Extended Data Fig. 4. Participant-specific FC change maps for drug sessions.

Individual participant methylphenidate (MTP) and psilocybin (PSIL) FC change maps. Left most column shows individuals’ functional networks. Right 3 columns show FC change maps, generated by calculating Euclidean distance from baseline seedmaps for each vertex. For each session the total score on the Mystical Experience Questionnaire (MEQ30: out of a maximum of 150) is given in the upper right corner. *P5 had an episode of emesis 30 minutes after drug ingestion during PSIL2.

Extended Data Fig. 5

Extended Data Fig. 5. Multi-dimensional scaling, dimension edge weights.

a) Group parcellation (324 cortical and 61 subcortical parcels)b) Weights from the first 4 dimensions generated by multi-dimensional scaling of the full dataset. The color of each pixel in the plot represents the weight of a given edge. Dimension 1 captures the loss of network integration (on diagonal boxes) and segregation (off diagonal boxes) of psilocybin. Dimensions 2 and 3 primarily explain individual differences and do not show network patterns as clearly. Dimension 4 captures shared effects of psilocybin (PSIL) and methylphenidate (MTP) on sensorimotor systems (suspected arousal effects).

Extended Data Fig. 6

Extended Data Fig. 6. Average functional connectivity (FC) matrices by condition.

a) Group parcellation (324 cortical and 61 subcortical parcels). b) Average FC matrices and condition differences. Top left shows the group average FC adjacency matrix. Bottom left shows the effect of psilocybin, e.g. increased correlation between dorsal attention, fronto-parietal, and default mode network to each other and to other cortical, limbic, and cerebellar systems. Top right shows effect of methylphenidate. For comparison and validation, we compared methylphenidate to the main effect of stimulant use within the last 24 hours (bottom right, n = 487 yes, n = 7992 no) in ABCD rs-fMRI data (bottom right).

Extended Data Fig. 7

Extended Data Fig. 7. Correlations with mystical experience scores.

Comparison of MEQ30 score (y-axes) to global desynchronization (top left; NGSC change, drug minus baseline), head motion (bottom left; framewise displacement (FD) in mm), heart rate change (top right; drug minus baseline), and respiratory rate change (bottom right; drug minus baseline), for all drug sessions. Statistics (rho, P) are based on bivariate correlation, two-sided, uncorrected. In the case of Δ NGSC, statistics are reported before and after the removal of an outlier point (> 2 SD lower than mean, indicated by the gray arrow).

Extended Data Fig. 8

Extended Data Fig. 8. Auditory-visual matching fMRI task.

a) Schematic of auditory/visual matching task design. b) Comparison of performance (‘No Drug’ and psilocybin conditions are at ceiling). Lines indicate means and standard deviation across sessions. Number of task sessions are indicated in Supplementary Table 1. c) Comparison of reaction time (RT). Lines indicate mean and standard deviation across all trials (48 trials per session). d) Task fMRI activation maps (beta weights) and e) contrasts (simple subtraction) using the canonical hemodynamic response function (HRF). f) Eight a priori regions of interest for timecourse analyses. g) Average timecourses from the regions of interest shown in panel f, calculated using finite impulse response model over 13 TR x 1.761 s/TR = 22.89 seconds, for all trials. Shaded area around each line indicates SEM. ANOVAN of Condition x HRF Beta (Main effect of all trials) magnitude testing effect of drug, two-sided: Left V1, F (2,40) = 3.91, P = 0.030; Right V1, F (2,40) = 4.40, P = 0.020; Left M1 hand, F (2,40) = 0.40, P = 0.68; Left Auditory A1, F (2,40) = 0.22, P = 0.81; Right Auditory A1, F (2,40) = 0.77, P = 0.47; Left Language, F (2,40) = 0.025, P = 0.98; Left DMN, F (2,40) = 1.15, P = 0.33; Right DMN, F (2,40) = 0.14, P = 0.87. *P < 0.05. P-values are uncorrected for multiple comparisons.

Similar articles

Cited by

References

    1. Goodwin, G. M. et al. Single-dose psilocybin for a treatment-resistant episode of major depression. N. Engl. J. Med.387, 1637–1648 (2022). 10.1056/NEJMoa2206443 - DOI - PubMed
    1. Raison, C. L. et al. Single-dose psilocybin treatment for major depressive disorder: a randomized clinical trial. JAMA330, 843–853 (2023). - PMC - PubMed
    1. Carhart-Harris, R. et al. Trial of psilocybin versus escitalopram for depression. N. Engl. J. Med.384, 1402–1411 (2021). 10.1056/NEJMoa2032994 - DOI - PubMed
    1. Bogenschutz, M. P. et al. Percentage of heavy drinking days following psilocybin-assisted psychotherapy vs placebo in the treatment of adult patients with alcohol use disorder: a randomized clinical trial. JAMA Psychiatry79, 953–962 (2022). - PMC - PubMed
    1. Cameron, L. P. et al. A non-hallucinogenic psychedelic analogue with therapeutic potential. Nature589, 474–479 (2020). - PMC - PubMed

MeSH terms

Substances

Grants and funding

LinkOut - more resources