PsiConnect: Multimodal Neuroimaging of Context-Dependent Brain and Behaviour Dynamics under Psilocybin

Abstract

PsiConnect is a large-scale neuroimaging study designed to investigate context-dependent neural and subjective effects of psilocybin using multimodal neuroimaging. It combines functional, structural, and diffusion-weighted MRI with EEG to examine brain activity in 62 participants before and after a 19 mg dose of psilocybin. The design includes resting-state scans and three naturalistic conditions: guided meditation, music listening, and movie watching. Half of the cohort underwent an 8-week meditation training program, enabling exploration of interactions among meditation, psilocybin, and brain function. fMRI data was obtained through multi-echo fMRI, enhancing signal-to-noise ratio and reducing susceptibility artifacts to improve reliability. A comprehensive battery of behavioural and self-report measures captured acute and longitudinal cognitive and subjective effects, with follow-ups to one year post-administration. The large sample, multimodal imaging, contextual diversity, and behavioural follow-ups enable study of psilocybin-induced brain and behaviour changes with unprecedented comprehensiveness and reliability. Data is curated according to open science principles to ensure accessibility and compatibility with established neuroimaging pipelines, making PsiConnect a valuable, reusable resource for cognitive and computational neuroscience.

Background & Summary

PsiConnect (a portmanteau of Psilocybin, Connectivity, and Context) is a large-scale neuroimaging study examining psilocybin’s effects on the human brain. Conducted at Monash University, it provides the world’s largest single-site neuroimaging dataset on serotonergic psychedelics to date, with 62 participants scanned before and after receiving a 19 mg oral dose of psilocybin. The study is multimodal, involving functional, anatomical, and diffusion MRI, as well as EEG recordings, to deepen our understanding of psilocybin’s effects at multiple neurophysiological scales. Both MRI and EEG scans included resting-state and three naturalistic stimuli: guided meditation, music listening, and movie watching. These enable the study of how psilocybin influences both spontaneous and stimulus-driven brain activity. Additionally, half of the participants completed an 8-week meditation training program, providing the opportunity to investigate the interactions between meditation, psilocybin, and brain function. The neuroimaging data is complemented by an extensive set of behavioural questionnaires that capture subjective experiences and cognitive changes induced by psilocybin. Participants were assessed longitudinally, with follow-ups extending up to one year post-administration, enabling the study of both acute and sustained changes. These assessments include self-report measures of mental health, personality, mindfulness, and psychedelic experiences, enabling correlations between subjective reports and neuroimaging findings. The study design is illustrated in Fig. 1.

Fig. 1.

(Top) Simplified representation of study design. Participants were allocated (non-randomised, balanced) to 8 week mindfulness meditation course or immediately enrolled to multimodal (fMRI and EEG) imaging sessions. An open label imaging session was conducted approximately one week before imaging under administration of 19 mg psilocybin. Music, mindfulness, personality and psychedelic measures were collected before and longitudinally after psilocybin administration. Meditators received additional assessments before and after the mindfulness course. (Bottom) Data acquisition, preprocessing, and cleaning pipelines.

A key methodological feature of PsiConnect is the use of multi-echo fMRI, which enhances signal-to-noise ratio without sacrificing temporal resolution. Multi-echo acquisition generates multiple images per volume at different echo times spanning tens of milliseconds, covering the entire range of signal decay rates in different voxels across the brain. While this comes at the expense of additional complexity in acquisition and pre-processing and reduced spatial resolution, it offers several advantages: 1) The echoes can be optimally weighted and combined into a single time series to increase the BOLD signal-to-noise ratio, 2) reduced signal dropout due to susceptibility artifacts, and 3) improved test-retest reliability of resting-state functional connectivity^1,2. Furthermore, multi-echo independent component analysis (ME-ICA) can identify nuisance signal components that do not exhibit typical BOLD signal decay over echo times³, which can enhance denoising compared to single-echo acquisitions. Finally, effect size estimates can increase by more than 50% in brain regions prone to signal dropout⁴. The combination of high temporal resolution^5,6, robust acquisition methods, and sophisticated preprocessing techniques ensures that the dataset is well-suited for studies on functional connectivity, brain network dynamics, and computational modelling.

PsiConnect is the only fMRI study of psilocybin to combine resting-state scans with three distinct naturalistic stimuli. It is complementary to the recent dataset by Siegel et al.⁷, which emphasizes precision functional mapping with multiple time points and an active placebo condition, but has a sample size of 7 participants. While Siegel et al. provide a longitudinal perspective (2 psilocybin scans and a total of approximately 18 scans per participant), PsiConnect offers the largest sample size to date (N = 62 vs N = 7) and enables the study of psilocybin’s effects across diverse contexts and modalities. By incorporating naturalistic conditions that closely align with practices in clinical research and emerging therapeutic applications, PsiConnect provides a unique opportunity to examine psilocybin’s impact on large-scale brain dynamics in ecologically valid contexts.

The study design and methodology ensure broad applicability across cognitive neuroscience, clinical research, and computational modelling, enabling researchers across various disciplines to leverage PsiConnect to examine fundamental questions about subjective experience, brain connectivity, and altered states of consciousness. Finally, all data is open-access via OpenNeuro⁸. For these reasons, PsiConnect has high reuse potential for investigating psychedelic effects on the brain and behaviour.

Here, we introduce the PsiConnect dataset structure and neuroimaging acquisition protocols. We detail the organisation of neuroimaging and behavioural data according to open science best practices to facilitate interoperability with existing software and processing pipelines. Researchers can use this information to process the raw data according to the specific needs of their methods. For those who wish to use the data seamlessly without reprocessing it, we release the minimally preprocessed data and analysis-ready cleaned data, explaining the processing steps in detail and providing quality control assessments to evaluate the efficacy of the preprocessing pipeline. Our extensive quality control assessments confirms the high quality of the data and its value for future research.

Methods

Ethics and clinical trial registration

The study protocol was approved by the Monash University Human Research Ethics Committee. The trial was registered with the Australian New Zealand Clinical Trials Registry under the registration number ACTRN12621001375842. Written consent was obtained at the start of the first expression of interest online questionnaire, then verbal consent was obtained prior to completing both the phone screening and the Structured Clinical Interview for DSM-5, Research version (SCID-5-RV), which was completed via video conferencing. Written informed consent was obtained from all participants prior to baseline imaging.

Study design

The study was open-label and included two imaging sessions: a baseline (no-psilocybin) session and a session following the administration of a 19 mg dose of psilocybin (see Fig. 1). The dose in mg/kg varied between 0.152 − 0.388 (0.262 ± 0.057). Both sessions involved MRI and EEG scans, with four conditions repeated in each: resting state, guided meditation, music listening, and movie watching. In fMRI, conditions began ~80 minutes post-dose and followed a fixed order (rest  → meditation  → music  → movie). The sequence progressed from low-stimulus to high-stimulus contexts to prioritise safety and experiential coherence for psychedelic-naïve participants. To mitigate scanner noise, participants were fitted with S15 insert earphones using expanding foam tips to ensure effective passive attenuation. Audio levels were individually calibrated prior to scanning by first adjusting a normalised audio stimulus to a comfortable level, followed by presentation of a test sequence of scanner noise to confirm that the stimulus remained clearly audible over gradient sounds without discomfort. Volume was adjusted based on participant feedback. The EEG session started after a room transfer and 20–40 minutes of setup (cap placement and impedance checks), with the movie presented first, followed by the three eyes-closed conditions (movie  → rest  → meditation  → music). In the resting-state condition (8 min MRI, 5 min EEG), participants were instructed to relax and keep their eyes closed while remaining still. During the guided meditation (6:30 min MRI, 5 min EEG), participants received brief instructions via MRI-safe audio, followed by silent periods of meditation. For the music listening condition (11:24 min MRI, 7 min EEG), a curated playlist was designed to evoke emotional depth and resonance. In the naturalistic movie condition (6 min MRI, 5 min EEG), participants watched a video of moving clouds without audio. EEG blocks were intentionally shorter (total ≈22 min) to acquire all four contexts within the acute-effects window and to minimise fatigue. These conditions were studied in both the baseline (no-psilocybin) and psilocybin sessions and were repeated in both fMRI and EEG, allowing for comprehensive cross-modal and longitudinal analyses of psilocybin’s influence on brain activity and connectivity.

Half of the participants were assigned to an 8-week mindfulness-based cognitive therapy (MBCT) program Finding Peace in a Frantic World run by a trained and registered instructor, which involved weekly online group classes and daily independent practice; the other half were assigned to a control group with no mindfulness intervention. To maximise the sample size within the two study arms, active control interventions (e.g., journaling) were not implemented.

Participant allocation used a non-randomised, balanced procedure to account for scheduling and class size constraints in the meditation arm. In practice, allocating half of the total participants (n = 33) required the 8-week mindfulness program to be repeated across three separate subgroups (n = 10 in group one; n = 6 in group two; n = 17 in group three). This structure facilitated the scheduling of imaging sessions, allowing baseline scans to align with the endpoint of participants’ mindfulness training. The balancing procedure stratified participants by age and self-reported gender to ensure approximate demographic equivalence between the meditation and non-meditation groups.

Several behavioural measures were collected before and during the baseline (no-psilocybin) and psilocybin scans (see Behavioural Data Acquisition). The follow-up conducted the day after psilocybin administration included semi-structured, open-ended questions and experience ratings. Further follow-up measures were administered one week, and one, three, six and twelve months after psilocybin administration.

Participants

Sixty-five healthy adults aged 18–55 (37.3 ± 10.7) with no psychedelic experience were recruited for the study; participants reported their gender as 32 women, 1 non-binary person, and 32 men. Participants were required to have no formal meditation practice, and limited previous exposure to meditation. They were first screened via short online survey and then detailed screening was performed by a suitably trained staff member for excluding any psychopathology using the long form SCID-V⁹. Exclusion criteria included a history of psychiatric disorders or suicidality, a 5-year history of substance and/or alcohol use disorder, first-degree relatives with a diagnosed psychotic disorder, a history of major neurological disorders including stroke and epilepsy, use of hallucinogens within the last six months, contraindicated medications, a formal meditation practice in the last 6 months or extensive previous exposure to mindfulness meditation. Extensive experience was operationally defined as engaging in any formal training or regular practice routine. Incidental exposure, such as trying a technique once or twice without establishing a habit or focus, was permitted. Participants were also screened for MR contraindications and provided informed consent. We operationally defined no psychedelic experience as no prior serotonergic psychedelic use with subjective effects. During structured screening, three participants disclosed nominal or remote lifetime exposures without subjective effects (n = 2 self-reported ineffective microdoses; n = 1 self-reported remote exposure  >20 years ago with minimal or no recall). All other participants (n = 59) reported no lifetime use across online, phone, and SCID checks.

On the day of psilocybin administration, they were assisted by a study doctor, researchers, lab staff with relevant training, and volunteers from the community drug harm reduction support organisation Dancewize. Psilocybin was generally well tolerated, although some adverse effects were reported, including reports of transient headaches during the night (n = 3). Three participants received follow-up support from a clinical psychologist familiar with psychedelic integration. These follow-up calls were conducted over the phone, and no further support was required.

Participant attrition (MRI)

All the 65 enrolled participants completed the baseline MRI session (no psilocybin). Two participants were withdrawn prior to the psilocybin session due to meeting exclusion criteria identified during the study period, and their baseline data was not analysed. The remaining 63 participants received psilocybin. One did not undergo any post-dose imaging, and another only partially completed the resting-state condition of the fMRI scan. This partial resting-state fMRI was included in the technical validation presented here. In summary, the numbers of available post-psilocybin scans are: rest = 62, meditation = 61, music = 61, movie = 61.

Participant attrition (EEG)

All the 65 enrolled participants completed the baseline EEG session (no psilocybin). However, two missed event markers due to technical issues, so the full baseline EEG data including all conditions and event markers is available for 63 participants. Of the 63 participants who received psilocybin, 59 completed the EEG session. However, one missed event markers due to technical issues, so the full post-dose EEG data including all conditions and event markers is available for 58 participants.

Data acquisition

Structural MRI acquisition

Structural MRI data was acquired using a Siemens 3 Tesla Magnetom Skyra scanner at Monash Biomedical Imaging, Monash University, Australia. T1-weighted (T1w) anatomical images were obtained for each participant during two sessions: at baseline (no-psilocybin) and on the psilocybin administration day. The images were acquired using a 3D magnetisation-prepared rapid gradient-echo (MP-RAGE) sequence with a 32-channel head coil. The acquisition parameters were as follows: repetition time (TR) of 2300 ms, echo time (TE) of 2.07 ms, 192 slices per slab, 1 mm slice thickness, and 1 mm isotropic voxel size. The parallel acquisition technique mode used was GRAPPA, with an acceleration factor (PE) of 2 on the baseline day and 3 on the psilocybin administration day. A T2-weighted anatomical image was obtained only during the baseline session. The acquisition parameters were: TR of 3200 ms, TE of 452 ms, 176 slices per slab, 1 mm slice thickness, and 1 mm isotropic voxel size, with an acceleration factor of 2.

Functional MRI acquisition

Blood-oxygenation-level-dependent (BOLD) fMRI data was collected using a multi-echo, multi-band, echo-planar imaging, T2*-weighted sequence. The acquisition parameters were: TR of 910 ms, multi-echo TE of 12.60 ms, 29.23 ms, 45.86 ms, 62.49 ms, multi-band acceleration factor of 4, field of view of 206 mm, RL phase encoding direction, and 3.2 mm isotropic voxels. The scan durations were: resting state with eyes closed (8 minutes, 505 volumes), audio-guided meditation with eyes closed (6:30 minutes, 405 volumes), music listening with eyes closed (11:24 minutes, 728 volumes), movie watching (6:00 minutes, 372 volumes). The structural and functional MRI images acquired from the Siemens scanner were converted into the Neuroimaging Informatics Technology Initiative (NIfTI) format.

Diffusion-weighted imaging acquisition

Diffusion-weighted imaging data was acquired using a single-shell acquisition scheme with an echo-planar imaging sequence. The acquisition parameters were as follows: TR of 5900 ms, TE of 171.0 ms, 56 slices per slab, 2.5 mm slice thickness, and 2.5 mm isotropic voxel size. The field of view was 212 mm in the read direction and 100% in the phase direction. The Parallel Acquisition Technique mode used was GRAPPA, with an acceleration factor (PE) of 2 and 40 reference lines in the phase-encoding direction. Diffusion weighting was applied along 71 directions using a monopolar diffusion scheme. Two b-values were used: 0 s/mm² and 3000 s/mm². The EPI factor was set to 86, and the bandwidth was 1038 Hz/Px. Fat suppression was implemented using a fat saturation technique.

EEG acquisition

EEG data was recorded using a 64-channel BrainAmp MR Plus amplifier (Brain Products GmbH, Germany) with BrainVision Recorder software (version 1.22.0001, Brain Products GmbH, Germany) and Ag/AgCl electrodes embedded in an actiCAP slim cap according to the standardized 10-10 system¹⁰. The FCz electrode served as the reference and the FPz electrode served as the ground for online recording. The impedances were maintained below 10 kΩ using an abrasive paste (Nuprep, Weaver and Company, USA) followed by applying a conductive gel (Easycap GmbH, Germany). EEG signals were sampled at 500 Hz with a bandpass filter of 0.01–1000 Hz. Participants were comfortably seated in an acoustically absorbent room to minimise environmental noise and their chin strap was securely fastened to ensure head stability. They were instructed to remain as still as possible and avoid jaw clenching during the recording to minimise artifacts.

The EEG recording session included the same four conditions as the fMRI described above. The naturalistic movie (5 minutes) was presented first, followed by the three eyes-closed conditions: resting state (5 minutes), audio-guided meditation (5 minutes), and music listening (7 minutes). The raw data was acquired as a single continuous recording over all conditions and then split into four separate files using the event markers for the start and end of each condition. Due to technical issues, event markers were not recorded for two participants in the baseline session and one participant in the psilocybin session, so these recordings were excluded from pre-processing and analysis. In summary, pre-processed EEG is available for 63 participants in the baseline session and 58 in the psilocybin session.

Behavioural data acquisition

A synopsis of the behavioural measures and delivery times is provided in Table 2 and Supplementary Table 1. Survey presentation was randomised between participants, whereby all non-well-being related measures were presented in a random order first, excluding the Depression Anxiety and Stress Scale (DASS-21)¹¹ and Inventory of Depression and Anxiety Symptoms (IDAS-II)^12,13. Both these scales always appeared at the end, and were randomised within each other. Participants in the meditation arm completed the following measures prior to meditation training: DASS-21, selected scales from an abridged version of the Expanded Version of the IDAS-II, The Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS)¹⁴, and State Adult Attachment Measure (SAAM)¹⁵. Each week of the meditation course, participants were instructed to complete the Mindfulness Adherence Questionnaire (MAQ). They also completed additional questions about their meditation practice, as described in the metadata (JSON) files. Prior to their baseline scanning sessions, all participants completed the following measures at home: Political Perspective Questionnaire (PPQ-5), Nature Relatedness Scale (NR-6)¹⁶, The Attachment Style Questionnaire (ASQ)¹⁷, Levenson Multidimensional Locus of Control Scales (LOC)¹⁸, NEO Five-Factor Inventory-3 (NEO-FFI-3)¹⁹, Aesthetic Experiences Scale in Music (AES-M)²⁰, Absorption in Music Scale (AIMS)²¹, Barcelona Music Reward Questionnaire (BMRQ)²², Short Test of Musical Preferences (STOMP-R)²³, and Life Attitude Profile-Revised (LAP-R)²⁴.

Table 2.

Pre-scored surveys available for each time point in BIDS format, including accompanying metadata.

Survey	Folder	Scales (TSV and JSON format)
No psilocybin	ses-01	AES, AIMS, ASC11, ASQ, BMRQ, CLOUDS, DASS, EQ, FFMQ, IDAS, LAPR, LOC, LRO, NEO-FFI-3, NR, PPQ, SAAM, SCS, STOMP-R, WEMWBS
Psilocybin	ses-02	AES, ASC11, ASC5, CLOUDS, EDI, LRO, MEQ30
Follow-up	followup-1 day	MINDSET, MOSTMEANING, INTENSITY
	followup-1 week	AES, EBI, FFMQ, LCI-R, LOC, SAAM, SCS
	followup-1 month	ASQ, DASS, EQ, FFMQ, IDAS, LAPR, LOC, NEO-FFI-3, NR, PEQ60, PPQ, SCS
	followup-3 month	FFMQ, LOC, SCS
	followup-6 month	FFMQ, LAPR, LOC, SCS
	followup-12 month	FFMQ, LAPR, NR, PPQ, SCS
Meditation training group only	meditation	DASS, EQ, FFMQ, IDAS, MAQ, MEDEQ, MEDITATION, SAAM, SCS, WEMWBS

On the baseline scan day, participants completed several measures after the MRI. They first completed a measure evaluating perceptions of patterns, faces, and imagery observed during the video of clouds (referred to as “clouds” or “clouds measure”). This was followed by a measure assessing ratings of music experience variables (liking, resonance, and openness)²⁵. This measure is referred to here as LRO.

During EEG setup, participants listened to six music samples, rating each using the LRO before advancing to the next sample. The order of the samples was randomised. The first EEG recording was the video of clouds. Following the video, participants completed the clouds measure. After this, EEG recording continued with rest, meditation, and music conditions. After EEG recording was completed, they completed the LRO. Shortly after, participants completed a series of psychological measures, including the Five Facet Mindfulness Questionnaire (FFMQ-SF)²⁶, DASS-21, IDAS-II, WEMWBS, and the SAAM. Additionally, a 42-item version of the 5-Dimension Altered States of Consciousness (5D-ASC) scale was administered, allowing for the derivation of the 11-Dimension ASC (11D-ASC) solution, as described in^27,28. Participants in the meditation condition also completed the Meditation Depth Questionnaire (MEDEQ)²⁹.

On the administration day, participants again first completed the LRO and clouds measure after the MRI, followed by the Ego Dissolution Inventory (EDI)³⁰. The order of these measures was kept consistent. During EEG setup, participants listened to six music samples, rating each using the LRO before advancing to the next sample. The order of the samples was randomised. The first EEG recording was the video of clouds. Following the video, participants completed the clouds measure. After this, EEG recording continued with rest, meditation, and music conditions. After the completion of EEG recording, the LRO was again administered. At the end of the session, participants completed the 94-item 5D-ASC (with 94 items used to construct the 5D-ASC and 42 items used to construct the 11D-ASC), the Mystical Experience Questionnaire - Revised (MEQ30)^31,32, the AES-M, and a second administration of the EDI. The order of the measures was randomised. The metadata (JSON) files for all sessions provide the questions asked and more details.

The one-day follow-up included semi-structured, open-ended questions, experience ratings, and meditation evaluations for the meditation arm. Ratings of personal meaningfulness and spiritual significance³³, subjective intensity, and anxiety at times throughout the day were also included. Further follow-up measures were administered one week, and one, three, six and twelve months after psilocybin administration. At each follow-up time point, participants completed the Self-Compassion Scale (SCS)³⁴, FFMQ-SF, and meditation participants additionally completed the MAQ. The one-week follow-up also included the Life Changes Inventory - Revised (LCI-R)³⁵, AES-M, Emotional Breakthrough Inventory (EBI)³⁶, LOC, and SAAM. The one-month follow-up also included: Persisting Effects Questionnaire Revised (PEQ60)³³, Experiences Questionnaire (EQ)³⁷, LAP-R, PPQ-5, NR-6, ASQ, LOC, DASS-21, IDAS-II, NEO-FFI-3, and meditation participants completed the Meditation Depth Questionnaire (MEDEQ) instead of the MAQ. The three-month follow-up also included LOC. The six-month follow-up also included LOC and LAP-R. Finally, the 12-month follow-up also included: LAP-R, PPQ-5, and NR-6.

Data preprocessing and cleaning

BIDS

The Digital Imaging and Communications in Medicine (DICOM) images acquired from the Siemens (Skyra) scanner were converted into the NIfTI format and organized according to the Brain Imaging Data Structure (BIDS) 1.7.0 using BIDScoin³⁸. Following the BIDS guidelines, each data file is accompanied by a corresponding metadata file in JSON format. These contain extensive lists of neuroimaging acquisition parameters, and the necessary information to interpret the behavioural measures. To de-identify the anatomical images, the facial features were removed using PyDeface (https://github.com/poldracklab/pydeface). The EEG data was also organised according to BIDS 1.7.0.

MRI anatomical data preprocessing

Anatomical MRI preprocessing was performed using fMRIPrep 22.0.2^39,40 (RRID:SCR_016216), which is based on Nipype 1.8.5^41,42 (RRID:SCR_002502). The baseline and administration T1-weighted (T1w) images were corrected for intensity non-uniformity (INU) with N4 BiasFieldCorrection⁴³, distributed with ANTs 2.3.3⁴⁴ (RRID:SCR_004757). The T1w-reference was then skull-stripped with a Nipype implementation of the antsBrainExtraction.sh workflow (from ANTs), using OASIS30ANTs as the target template. Brain tissue segmentation of cerebrospinal fluid (CSF), white matter (WM) and gray matter (GM) was performed on the brain-extracted T1w using fast⁴⁵ (FSL 6.0.5.1:57b01774, RRID:SCR_002823). A T1w-reference map was computed after registration of 2 T1w images (after INU-correction) using mri_robust_template⁴⁶ (FreeSurfer 7.2.0). Brain surfaces were reconstructed using recon-all⁴⁷ (FreeSurfer 7.2.0, RRID:SCR_001847), and the brain mask estimated previously was refined with a custom variation of the method to reconcile ANTs-derived and FreeSurfer-derived segmentations of the cortical gray-matter of Mindboggle⁴⁸ (RRID:SCR_002438). Volume-based spatial normalization to one standard space (MNI152NLin2009cAsym) was performed through nonlinear registration with antsRegistration (ANTs 2.3.3), using brain-extracted versions of both T1w reference and the T1w template. The following template was selected for spatial normalization: ICBM 152 Nonlinear Asymmetrical template version 2009c⁴⁹ (RRID:SCR_008796; TemplateFlow ID: MNI152NLin2009cAsym).

MRI functional data preprocessing

Functional MRI preprocessing was also performed using fMRIPrep 22.0.2^39,40. For each of the 8 BOLD runs per participant (across all tasks and sessions), the following preprocessing was performed. First, a reference volume and its skull-stripped version were generated from the shortest echo of the BOLD run using a custom methodology of fMRIPrep. Head-motion parameters with respect to the BOLD reference (transformation matrices, and six corresponding rotation and translation parameters) were estimated before any spatiotemporal filtering using mcflirt⁵⁰ (FSL 6.0.5.1:57b01774). BOLD runs were slice-time corrected to 0.401 s (0.5 of slice acquisition range 0s-0.802 s) using 3dTshift from AFNI⁵¹ (RRID:SCR_005927). The BOLD time-series (including slice-timing correction) were resampled onto their original, native space by applying the transforms to correct for head-motion. These resampled BOLD time-series will be referred to as preprocessed BOLD in original space, or just preprocessed BOLD. A T2^⋆ map was estimated from the preprocessed EPI echoes, by voxel-wise fitting the maximal number of echoes with reliable signal in that voxel to a monoexponential signal decay model with nonlinear regression. The T2^⋆/S0 estimates from a log-linear regression fit were used for initial values. The calculated T2^⋆ map was then used to optimally combine preprocessed BOLD across echoes following the method described in⁵². The optimally combined time series was carried forward as the preprocessed BOLD. The BOLD reference was then co-registered to the T1w reference using bbregister (FreeSurfer) which implements boundary-based registration⁵³. Co-registration was configured with six degrees of freedom. Several confounding time series were calculated based on the preprocessed BOLD: framewise displacement (FD), DVARS and three region-wise global signals. FD was computed using two formulations following Power (absolute sum of relative motions,⁵⁴) and Jenkinson recommendations (relative root mean square displacement between affines⁵⁰). FD and DVARS are calculated for each functional run, both using their implementations in Nipype, following the definitions by⁵⁴. The three global signals are extracted within the CSF, the WM, and the whole-brain masks. The BOLD time series were resampled in standard space, generating a preprocessed BOLD run in MNI152NLin2009cAsym space. All resamplings were performed with a single interpolation step by composing all the pertinent transformations (i.e. head-motion transform matrices and co-registrations to anatomical and output spaces). Gridded (volumetric) resamplings were performed using antsApplyTransforms (ANTs), configured with Lanczos interpolation to minimize the smoothing effects of other kernels⁵⁵. Non-gridded (surface) resamplings were performed using mri_vol2surf (FreeSurfer). Many internal operations of fMRIPrep use Nilearn 0.9.1⁵⁶ (RRID:SCR_001362), mostly within the functional processing workflow. For more details of the pipeline, see the section corresponding to workflows in fMRIPrep’s documentation.

Additional available regressors

Additionally, a set of physiological regressors were extracted to allow for component-based noise correction (CompCor⁵⁷). Principal components are estimated after high-pass filtering the preprocessed BOLD time series (using a discrete cosine filter with 128 s cut-off) for the two CompCor variants: temporal (tCompCor) and anatomical (aCompCor). tCompCor components are then calculated from the top 2% variable voxels within the brain mask. For aCompCor, three probabilistic masks (CSF, WM and combined CSF+WM) are generated in anatomical space. The implementation differs from that of Behzadi et al. in that instead of eroding the masks by 2 pixels on BOLD space, a mask of pixels that likely contain a volume fraction of GM is subtracted from the aCompCor masks. This mask is obtained by dilating a GM mask extracted from the FreeSurfer’s aseg segmentation, and it ensures components are not extracted from voxels containing a minimal fraction of GM. Finally, these masks are resampled into BOLD space and binarized by thresholding at 0.99 (as in the original implementation). Components are also calculated separately within the WM and CSF masks. For each CompCor decomposition, the k components with the largest singular values are retained, such that the retained components’ time series are sufficient to explain 50% of variance across the nuisance mask (CSF, WM, combined, or temporal). The remaining components are dropped from consideration.

The head-motion estimates calculated in the correction step were also placed within the corresponding confounds file. The confound time series derived from head motion estimates and global signals were expanded with the inclusion of temporal derivatives and quadratic terms for each⁵⁸. Frames that exceeded a threshold of 0.5 mm FD or 1.5 standardized DVARS were annotated as motion outliers. Additional nuisance time series are calculated by means of principal components analysis of the signal found within a thin band (crown) of voxels around the edge of the brain, as proposed by⁵⁹.

fMRI data cleaning

The preprocessed and optimally combined data produced by fMRIPrep was cleaned via a single regression in SPM12⁶⁰ using the general linear model. The regressors were: the white matter and cerebrospinal fluid signals computed by fMRIPrep, the frame-wise displacement, and the non-BOLD components identified via multi-echo ICA performed using tedana (version 0.0.12)⁶¹. In short, multi-echo ICA identifies non-BOLD components that are independent of the echo time³. Finally, to facilitate surface-based analyses, the cleaned volumetric data was projected to the FreeSurfer left-right-symmetric cortical surface template with 32000 vertices for each hemisphere (fsLR32k).

EEG data preprocessing and cleaning

The raw EEG signals were preprocessed using the default parameters of the automated RELAX pipeline⁶² implemented in MATLAB using functions from EEGLAB⁶³ and FieldTrip⁶⁴ toolboxes. First, the data was bandpass filtered between 0.25 and 80 Hz using a fourth-order Butterworth filter with zero phase, with a notch filter applied between 47 and 53 Hz to reduce line noise. Subsequently, bad channels were removed by applying a multistep process incorporating the “findNoisyChannels” function of the PREP pipeline⁶⁵. This was followed by the initial reduction of the artefacts related to eye movements, muscle activity and drift using a multi-channel Wiener filter⁶⁶, after which the data were re-referenced to the robust average reference⁶⁵. Residual artefacts were then removed by Wavelet-enhanced Independent Component Analysis (WICA) with the artefactual components identified by the automated ICLabel classifier⁶⁷. Electrodes rejected during the cleaning process were then restored to the data through spherical interpolation. Finally, all pre-processed data were visually inspected for data quality.

Behavioural data processing

Questionnaires were administered online using Qualtrics. The raw data from all questionnaires were collected in their original Likert scale format, as defined by the authors of each measure. Raw data spreadsheets are provided for each time point. Subscale scores were computed according to the authors’ guidelines, using the referenced papers listed in the metadata (JSON) file for each time point. If fewer than 50% of the required items for a scale were missing, the missing values were imputed using the median of the available items. Both the imputed and not imputed values are released as part of the dataset.

Data Records

The data can be accessed from the OpenNeuro public repository (access number: ds006110), organised according to the Brain Imaging Data Structure (BIDS) specification, version 1.7.0⁶⁸. Specifically, the defaced raw data and the MRI and EEG derivatives are organised in the subfolders indicated in Table 1. The MRI derivatives are listed in the same order as they were computed via the preprocessing and cleaning pipeline: cortical surface reconstruction via FreeSurfer, minimal preprocessing via fMRIPrep, multi-echo ICA via tedana, and cleaning via a GLM performed using SPM12. The following sections will provide additional information about naming conventions and available formats for the files contained in the folders listed in Table 1.

Table 1.

Paths to defaced raw data and derivatives organised according to the Brain Imaging Data Structure (BIDS) 1.7.0. The MRI derivatives are listed in the same order as they were computed via the preprocessing and cleaning pipeline.

Data type	Folder	Participant ID	Session	Modality or format
Defaced raw data	/bids	/sub-PC0*, /sub-PC2*	/ses-01, /ses-02	/anat, /eeg, /fmap, /func
Surface reconstruction	/derivatives/freesurfer-7.2	/sub-PC0*, /sub-PC2*
Minimally preprocessed MRI	/derivatives/fmriprep-22.0.2	/sub-PC0*, /sub-PC2*	/ses-01, /ses-02	/anat, /func
Multi-echo ICA	/derivatives/tedana-0.0.12	/sub-PC0*, /sub-PC2*	/ses-01, /ses-02	/func
Cleaned fMRI	/derivatives/tedana-0.0.12-GLM	/sub-PC0*, /sub-PC2*	/ses-01, /ses-02	/func
Cleaned EEG	/derivatives/EEG/cleaned_RELAX			/FieldTrip_format

Raw MRI data organised according to BIDS

De-identified MRI data (with coded subject IDs and defaced structural images) is organised in individual participant directories named “sub-<ID>”. The IDs starting with “PC2” refer to participants who underwent the 8-week meditation training before the neuroimaging sessions. All other IDs start with “PC0”.

Preprocessed and cleaned data

Surface reconstruction data is stored in /derivatives/freesurfer-7.2. Minimally preprocessed MRI data is stored in /derivatives/fmriprep-22.0.2, which also contains individual participant reports, each named sub-<ID>.html. The ME-ICA output is stored in /derivatives/tedana-0.0.12. For each participant, there are individual HTML reports, a table listing the accepted and rejected independent components (named desc-tedana_metrics.tsv), and a single NIFTI file with the BOLD signal obtained by optimally combining the four echoes (file names ending in optcomb_bold.nii.gz). This optimally combined file is cleaned via GLM, and the cleaned data is stored in /derivatives/tedana-0.0.12-GLM, with a file name ending in optcomGLM_bold.nii.gz. To facilitate analysis in surface space, this is also projected to the fsLR32k template (name ending in optcomGLM_fsLR_32k.mat). Cleaned EEG data is stored in /derivatives/EEG/cleaned_RELAX and is available in FieldTrip format.

Audio and video stimuli

All audio and video stimuli were prepared and presented using open-source software. To comply with the media formatting requirements of the presentation software PsychoPy3 (version 2021.2.3), used during MRI and EEG scanning, all audio recordings were converted to the WAV format (sampling rate: 44100 Hz; bit depth: 16-bit; stereo), and the silent (clouds) video file was converted to the M4V format with an RGB24 codec (resolution: 1920 × 1080 pixels; frame rate: 30.00 fps) using functions in VLC media player (version 3.0.12). Audio stimuli were additionally normalized to a peak amplitude of −1.0 dB using functions in Audacity (version 3.1.3). A complete list of stimuli used in MRI and EEG sessions is provided below, including both the calibration audio for volume adjustments and the instructional audio presented prior to the (clouds) video. Due to copyright restrictions, certain audio files are omitted from the repository but are documented here.

MRI

• Task: Audio calibration, Track: India Song (Piano), Author: D’Alessio, C., Duration: 166.58 s, Notes: Subject to copyright (and thus omitted from repository).

• Task: Meditation, Filename: meditation.wav, Author: Stoliker, D., Duration: 390.10 s.

• Task: Music (part 1), Track: Sunset from the Ethereal Shore (Lauge Rework), Author: Spacecraft, Lauge, Duration: 351.56 s, Notes: Subject to copyright (and thus omitted from repository).

• Task: Music (part 2), Track: Sollys (Applefish Rework), Author: Lauge, Applefish, Duration: 332.38 s, Notes: Subject to copyright (and thus omitted from repository).

• Task: Instructions (open eyes and note patterns), Filename: instructions.wav, Author: Greaves, M. D., Duration: 18.54 s.

• Task: Video, Filename: clouds.m4v, Author: Stoliker, D., Duration: 360.08 s.

EEG

• Task: Video, Filename: clouds_short.m4v, Author: Stoliker, D., Duration: 300.00 s.

• Task: Meditation, Filename: meditation_short.wav, Author: Stoliker, D., Duration: 300.00 s.

• Task: Music (part 1), Track: Sunset from the Ethereal Shore (Lauge Rework), Author: Spacecraft, Lauge, Abbreviated track duration., Notes: Subject to copyright (and thus omitted from repository).

• Task: Music (part 2), Track: Sollys (Applefish Rework), Author: Lauge, Applefish, Abbreviated track duration, Notes: Subject to copyright (and thus omitted from repository).

Behavioural data

Following BIDS conventions, behavioural data is stored in the phenotype folder, with subfolders for different surveys: those specific to the meditation arm (pre-training, weekly measures during the 8-week course, and post-training measures collected after both the 8-week course and psilocybin administration), as well as baseline, psilocybin administration, and follow-up surveys (Table 2). Each session folder contains one or more TSV files. Each TSV file refers to one scale and contains the scores for all the subscales involved. The scales are listed in Table 2 and described in Supplementary Table 1. Each TSV file is accompanied by a metadata (JSON) file that contains the description of the scale, a reference to the relevant literature, and information about how the items were aggregated into a score for each subscale (for example, by summation, average, or rescaling to the 0–100 range). In addition, each folder contains a JSON file listing the specific question asked for each item in the questionnaire. The raw data is provided in the phenotype/rawdata folder, which contains one spreadsheet per questionnaire in TSV format, each accompanied by a sidecar JSON file listing the item prompts.

Technical Validation

MRI quality control

Quality control was performed using the MRIQC BIDS app⁶⁹, which uses structural and functional MRI images to compute several quality metrics (see Supplementary Figs. 1 and 2 for a comprehensive overview). These include the temporal signal-to-noise ratio and the frame-wise displacement that quantifies head motion⁵⁴. Results and group reports can be found in the derivatives/mriqc-22.0.6 folder. Principal component analysis (PCA) of these QC metrics revealed clusters corresponding to the four echos, and helped identify some outliers in the structural and functional images (see Fig. 2), which were further confirmed by visual inspection. In the case of fMRI, we found that head motion was the dominant factor causing outliers on the PCA plot, as confirmed by the square markers in Fig. 2a that indicate scans with large head motion (mean frame-wise displacement greater than 0.5, which is often used as an exclusion threshold⁷⁰). Indeed, head motion, quantified via the mean framewise displacement, was the feature with the largest contribution to the second principal component (see Supplementary Fig. 3). A detailed analysis of the mean frame-wise displacement showed that participants moved their head more during the psilocybin MRI session than in the previous session without psilocybin (Fig. 2b,c).

Fig. 2.

Principal components analysis (PCA) of quality control metrics computed using MRIQC captures important features of the data and flags outliers for further inspection. (a) Functional MRI quality control metrics. Each point represents a single echo, and colour-coding the points by echo number showed that PCA was able to cluster the four echoes. Points that are further away from the cluster centres are outliers in terms of multiple metrics. However, head motion was the dominant factor in determining these outliers, as shown by the square markers that indicate scans with high head motion (mean frame-wise displacement greater than 0.5). (b) Colour-coding the same points by session (no psilocybin and psilocybin) shows that the outliers belong to the psilocybin session. (c) Participants moved their head more during the psilocybin MRI session, in alignment with the findings presented in first two panels. Head motion was quantified using the mean frame-wise displacement (in mm). (d) Anatomical (T1w) MRI quality control metrics. Square markers indicate outliers that were confirmed by visual inspection of the anatomical scans.

While we retained and released all the data, we used the above process to generate a table of recommended exclusions for users who do not wish to re-run a dedicated quality control analysis (see Supplementary Table 2). There are no suggested exclusions for the fMRI images collected during baseline. Among the functional MRI images collected during psilocybin, we recommended excluding specific conditions for 7 participants: 2 rest, 2 meditation, 5 music, and 4 movie scans, as indicated in Fig. 2a and Supplementary Table 2. If these suggestions conditions were excluded from analysis, in addition to the exclusions and attrition detailed in the Participants section above, the fMRI samples during psilocybin would be: rest = 60, meditation = 59, music = 56, and movie = 57.

Among the anatomical MRI files, we excluded 8 structural T1w images from the psilocybin session but retained all the T1w images from the baseline session, as indicated in Fig. 2d. These exclusions were applied in our exploratory analysis of the PsiConnect MRI data⁷¹.

Cleaning the data removed slow drifts in the global signal

Functional MRI recordings often show a drift over time, due to magnetic field fluctuations caused by the scanner hardware, or by physiological factors such as slow variations in respiration or cardiac cycles that can introduce low-frequency fluctuations in the BOLD signal⁷². This drift is evident in the raw data but is largely removed after cleaning via ME-ICA, and even more so after including white matter and CSF signals as regressors in the GLM (see Supplementary Fig. 4). Since cleaning didn’t include global signal regression nor high-pass filtering, it is reasonable that some of these slow physiological drifts were captured by the confounding signals that were regressed.

Signal-to-noise ratio

The second echo was used as a proxy for single-echo data since the second echo time matches the typical value used in single-echo acquisitions (~30 ms). The temporal SNR (tSNR) of this pseudo single-echo data was compared to the optimally combined multi-echo data and to the cleaned ME-ICA data, both at the subject- and at the voxel-level. For each scan, the median tSNR across all voxels is higher in the optimally combined multi-echo data than in the single-echo data, and it further increases after ME-ICA cleaning (Fig. 3a). This first result reveals the overall trend but ignores the spatial distribution of the tSNR across voxels. In order to locate these tSNR changes, we computed the median difference between tSNR values of each voxel (by subtracting the single-echo tSNR value from the optimally combined tSNR). The spatial maps shown in Fig. 3b demonstrate that the largest improvement is achieved in brain regions prone to signal dropout, including the temporal pole, ventromedial prefrontal cortex, and subcortical areas. In addition, Fig. 3c shows that the largest SNR improvement of ME-ICA over the optimally combined data is obtained in superficial grey matter voxels. Our results successfully replicate the median subject-level tSNR reported in² and confirm that the benefits of the multi-echo data and ME-ICA procedures are region-specific. Importantly, the optimally combined multi-echo data shows higher tSNR in regions prone to signal dropout and in subcortical areas that are particularly relevant in the context of psychedelics. However, we must be careful not to understate the drawbacks of multi-echo fMRI sequences. The higher complexity in the acquisition and processing stages were mentioned above, as well as the reduced spatial resolution. In addition, the louder scanner noise could be distressing in psychedelic studies and might complicate task acquisitions involving complex auditory stimuli.

Fig. 3.

(a) Comparison of temporal signal-to-noise ratio (tSNR) between single-echo data, optimally combined multi-echo data, and cleaned data via the multi-echo independent component analysis (ME-ICA). (b) Spatial distribution of temporal signal-to-noise ratio (SNR) differences. Largest tSNR increase in the optimally combined multi-echo data compared to single-echo data. c) Further tSNR increase when using cleaned data via multi-echo ICA compared to the optimally combined data (thresholded at the top quartile to locate the voxels with the largest tSNR increase).

Behavioural scores

The acute effects assessed using the 11-Dimension Altered States of Consciousness (11-D ASC) scale²⁸ and the Mystical Experience Questionnaire (MEQ30)³² illustrate substantial group-level subjective effects intensity from the standardised 19mg dose of psilocybin (Fig. 4a,b). To provide an overview of the other behavioural scores available in the dataset, we present the matrices of pairwise Pearson correlation between the scores acquired during the psilocybin and no-psilocybin sessions, as well as during the meditation training and the longitudinal follow-ups (Fig. 4c,d,e,f). The matrices have been clustered to highlight sets of highly intercorrelated behavioural scores, which reflect key theoretical and conceptual distinctions in psychedelic effects. For example, the clusters identified during the psilocybin session (Fig. 4d) distinguish between: positively felt self- and boundary-dissolving effects, which characterise the high-level associative, abstract and existential experiential qualities (e.g., transcendence, bliss, unity); sensory-hallucinogenic effects (e.g., visual imagery, synaesthesia); and dysphoric effects (e.g., impaired control and cognition, and anxiety). We also note that negative effects, particularly anxiety, were infrequent or rated as low by the participants. Two clusters also emerge in the follow-up questionnaires and in the meditation-related scores (Fig. 4e,f), separating positive (purpose, meaning, coherence, well-being) and negative scores (stress, dysphoria, anxiety, discomfort).

Fig. 4.

(a) Radar plot of the 11-Dimension Altered States of Consciousness (ASC) Questionnaire, showing minimum-maximum range (shaded) and mean scores (solid line) across participants in both the psilocybin and no-psilocybin sessions. (b) Radar plot of Mystical Experiences Questionnaire (MEQ30) showing minimum-maximum range (shaded) and mean scores (solid line) for 4 dimensions across participants. (c) Pearson correlation between all pairs of behavioural scores acquired during baseline session (no psilocybin). (d) Pearson correlation between all pairs of behavioural scores acquired after psilocybin administration. (e) Pearson correlation between all pairs of behavioural scores acquired during longitudinal follow-ups. (f) Pearson correlation between all pairs of behavioural scores related to the meditation scans and the 8-week meditation training (for half of the participants).

Usage Notes

Quality control of structural and functional data:

1. mriqc_singularity.sh (run MRIqc)

2. mriqc_PCA.m (obtain the plots in Fig. 2 based on the T1w and BOLD reports generated by MRIqc)

Pre-processing:

1. fmriprep_singularity.sh

Cleaning:

1. tedana.sh (identify non-BOLD independent components using ME-ICA)

2. glm_tedana_cleaning.m (fit a general linear model to regress head motion, cerebrospinal fluid, white matter, and non-BOLD independent components labelled as artifacts during the previous ME-ICA step. There is also the option to regress the global signal)

Author contributions

L.N.: Methodology, Software, Validation, Formal analysis, Data Curation, Visualization, Writing - Original Draft.D.S.: Conceptualization, Methodology, Formal analysis, Investigation, Data Curation, Writing - Review & Editing, Project administration. T.B.: Data Curation, Writing - Review & Editing. M.D.G.: Methodology, Software, Investigation, Data Curation, Writing - Review & Editing. S.C.: Methodology, Writing - Review & Editing. J.J.: Investigation. J.K.: Investigation, Writing - Review & Editing. J.C.P.: Software. M.L.W.: Investigation, Writing - Review & Editing. A.R.: Conceptualization, Methodology, Investigation, Writing - Review & Editing, Supervision, Project administration, Funding acquisition.

Data availability

The data can be accessed from the OpenNeuro public repository: https://openneuro.org/datasets/ds006110.

Code availability

Open-source code for all data analysis pipelines will be made available on GitHub upon publication (https://github.com/razilab/psiconnect). These pipelines used the following software packages: fMRIprep 22.0.2 (https://fmriprep.org), tedana 0.0.12 (https://tedana.readthedocs.io), MRIqc 22.0.6 (https://mriqc.readthedocs.io), MATLAB R2022a (https://www.mathworks.com), SPM12 (https://www.fil.ion.ucl.ac.uk/spm/software/spm12), Freesurfer 7.2 (https://surfer.nmr.mgh.harvard.edu), FSL v.6.0.7 (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki), FieldTrip 20240916 (https://www.fieldtriptoolbox.org). LZ complexity was computed by adapting open-source code by Fernando Rosas and Pedro Mediano (https://information-dynamics.github.io/complexity/information/2019/06/26/lempel-ziv.html). Projection of volumetric fMRI data to the cortical surface was performed by adapting open-source code by Pang et al. available on GitHub (https://github.com/NSBLab/BrainEigenmodes).

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41597-026-07312-1.