The psychedelic state induced by ayahuasca modulates the activity and connectivity of the default mode network

Palhano-Fontes and colleagues frame their study within growing evidence that serotonergic psychedelics produce profound alterations in perception, cognition and self-related processing. The authors note that increased introspection is a hallmark of these states and that introspective, internally oriented mental activity has been consistently linked to the Default Mode Network (DMN), a set of brain regions more active at rest than during externally directed tasks. Previous work has shown DMN modulation in a variety of altered states (sleep, meditation, sedation) and after psilocybin; the authors therefore sought to test whether ayahuasca, a brew containing DMT plus monoamine oxidase inhibitors, similarly alters DMN activity and connectivity. This study set out to examine three related questions: whether acute ayahuasca intake changes DMN activity during an active task (verbal fluency), whether resting-state functional connectivity among DMN nodes is altered, and whether the relationship (orthogonality/anti-correlation) between the DMN and the task-positive network (TPN) is affected. The authors hypothesised that, given ayahuasca’s introspective effects and parallels with psilocybin and meditation, DMN activity would be reduced and DMN connectivity altered during the psychedelic state.

This was an fMRI study in experienced ayahuasca users. Ten healthy, right-handed volunteers with at least five years of regular ayahuasca use (initial sample: 5 males; mean age 29 years, range 24–48) were recruited from a Santo Daime church. One participant’s data were excluded for excessive head motion, leaving nine subjects (five women) for analysis. An independent control group of 26 individuals (11 males; mean age 37.39 years) was used to define DMN regions of interest (ROIs). Psychiatric screening (DSM-IV structured interview) excluded neurological or psychiatric comorbidity, and participants were medication-free for at least three months; they abstained from caffeine, nicotine and alcohol before scanning. Each participant underwent two MRI sessions on the same day: one before and one 40 minutes after oral ayahuasca (2.2 mL/kg). The brew used (provided by Santo Daime) contained 0.8 mg/mL DMT and 0.21 mg/mL harmine; harmaline was below detection. In each session subjects completed a block-design verbal fluency task (six rest blocks alternating with five 27.5 s silent word-generation blocks cued by letters) to evaluate task-related DMN activity, and a conventional resting-state acquisition (eyes closed) to assess functional connectivity. During rest blocks in the task participants were instructed to imagine a white wall. Subjective effects were measured with the Brief Psychiatric Rating Scale (BPRS) and the Young Mania Rating Scale (YMRS) at baseline, 40, 80 and 200 minutes after ingestion. Image acquisition used a 1.5 T scanner. For the verbal fluency task 66 EPI volumes (16 slices) were acquired with TR = 4600 ms; for resting state 150 volumes were acquired with TR = 1700 ms (other EPI parameters reported). High-resolution T1-weighted anatomical scans were also obtained. Preprocessing in FSL included slice-timing correction, head-motion correction, spatial smoothing (5 mm FWHM), and registration to the individual anatomy and to the MNI152 template. To avoid circularity, DMN ROIs were defined from the control group performing the same verbal fluency task. GLM analyses (rest > task contrast) were performed in SPM5 with a high-pass filter at 0.015 Hz; group maps thresholded at q[FDR] < 0.01 and cluster size ≥ 50 voxels were intersected with nine a priori DMN areas (ACC, PCC, precuneus, mPFC, left middle frontal gyrus, bilateral middle temporal gyri, bilateral inferior parietal lobules) to create the DMN mask. For task effects the experimental group’s rest > task contrasts entered a second-level paired (fixed-effect) comparison of before versus after ayahuasca; mean β-values were extracted from the nine ROIs and tested with Wilcoxon paired tests, Bonferroni-corrected for nine ROIs. Correlations between individual Δβ (after − before) and psychometric scores at 80 minutes were tested and corrected for multiple comparisons. Resting-state seed-based functional connectivity (fc-fMRI) was analysed in REST. Nuisance regressors (white matter, CSF, global signal and six motion parameters) were removed, then residuals band-pass filtered between 0.01 and 0.08 Hz. Ten-mm radius seeds were placed in two DMN hubs: PCC (MNI −4, −47, 45) and mPFC (MNI 0, 51, −14). Seed time courses were correlated voxel-wise, Fisher z-transformed and compared across sessions using paired t-tests within the predefined nine-ROI mask; separate analyses were also performed for each seed. To evaluate DMN–TPN orthogonality, three classical TPN seeds (IPS, FEF, MT) were used to derive a TPN mask; Pearson correlations between averaged DMN and TPN time series were compared before and after ayahuasca. Because global signal regression can influence negative correlations, analyses were performed both with and without global signal regression.

Nine participants contributed usable data to the experimental analyses; ROI definition used an independent control group of 26 subjects. Task-based contrasts (rest > task) confirmed that the nine a priori DMN ROIs showed the expected rest > task signal difference both before and after ayahuasca. Comparing sessions, ayahuasca produced a significant decrease in DMN activity across most ROIs, including canonical hubs such as the posterior cingulate cortex/precuneus and the medial prefrontal cortex. Two ROIs—left middle frontal gyrus (LMFG) and left middle temporal gyrus (LMTG)—showed a significant signal increase after ayahuasca; the authors note these areas are language-related and could reflect engagement during the verbal fluency task, potentially confounding interpretation. Group-level ROI analyses reported statistically significant reductions in activity for the majority of DMN nodes (p < 0.01, Bonferroni corrected across nine ROIs). Resting-state seed-based connectivity analyses showed a significant decrease in functional connectivity within the PCC/precuneus after ayahuasca intake. Examining seeds separately indicated that the PCC seed contributed most to this observed connectivity reduction. Direct mPFC–PCC connectivity (seed-to-seed) did not reach significance (reported r values: before ≈ 0.13, after ≈ −0.02; p = 0.12). Correlation analyses relating individual Δβ to psychometric change revealed a trend-level negative association between Δβ in the ACC and YMRS scores at 80 minutes (r = −0.78, p = 0.072 after correction for 12 comparisons), suggesting greater ACC signal reduction was associated with larger YMRS change, but this did not meet conventional significance thresholds after correction. Analyses of DMN–TPN interactions produced no statistically significant change in DMN–TPN orthogonality after ayahuasca. When global signal was regressed out (a common preprocessing choice), the expected DMN–TPN anti-correlation was observed and remained statistically unchanged pre- versus post-ayahuasca. Repeating the analysis without global signal regression yielded overall positive DMN–TPN correlations both before and after intake, with a non-significant trend toward increased positive correlation after ayahuasca (p = 0.22). The authors therefore report no reliable change in the DMN–TPN relationship under ayahuasca in their dataset.

Palhano-Fontes and colleagues interpret their findings as evidence that acute ayahuasca intake attenuates activity in core DMN structures and reduces functional coupling within the posterior midline (PCC/precuneus). Two non‑exclusive explanatory hypotheses are offered. First, the authors propose that the ayahuasca state may functionally resemble a task-like condition for experienced users—an internally demanding ‘‘labor’’ that reduces canonical resting DMN activity because subjects are actively engaging with intense inner experience. Second, they note parallels with meditation: both states increase introspection and alter awareness of mind‑wandering, and prior work links meditation to decreased DMN activity. The authors suggest that while psychedelics and meditation differ in their effects on mind‑wandering (psychedelics may amplify it), both may increase meta-awareness of those thoughts, which could explain reduced DMN signal. Comparisons with previous findings are highlighted. The pattern seen here—particularly reduced posterior midline connectivity—resembles changes reported for psilocybin, sleep onset and light sedation, but differs in that mPFC–PCC coupling was not significantly reduced after ayahuasca as it has been reported after psilocybin. The authors attribute these differences to the unique phenomenology and pharmacology of ayahuasca: beyond serotonergic agonism, ayahuasca contains DMT with affinities for multiple receptor systems and monoamine oxidase inhibitors that alter monoamine levels, and somatic/sedation effects tend to be stronger than with psilocybin. The authors acknowledge several limitations that constrain interpretation and generalisability. The sample comprised experienced users only, chosen to minimise vomiting and movement in the scanner; this prevents disentangling drug effects from the influence of prior experience. No placebo control was used, because a credible placebo for experienced users is difficult to create; the hospital scanning context makes a simple placebo explanation less likely, but not impossible. Analyses used fixed‑effects group statistics, which restricts inference to the studied group rather than the population level. Order effects were not controlled because both scans were acquired the same day two hours apart, a design choice the authors argue was necessary for practical and safety reasons but that limits counterbalancing. Finally, the small sample size was implicit in several analyses and some correlations only approached significance after correction. Taking these caveats together, the authors conclude that acute ayahuasca is associated with diminished DMN activation and reduced posterior midline connectivity, a pattern consonant with altered states such as those produced by psilocybin, meditation and early sleep stages, while also recognising substance‑specific differences and the need for larger, placebo‑controlled and more diverse studies.