From relaxed beliefs under psychedelics (REBUS) to revised beliefs after psychedelics (REBAS)

Psychedelic therapy has attracted growing interest because of its reported transdiagnostic effects on mental health, including reductions in anxiety, depression and suicidal ideation and increases in well-being. The Relaxed Beliefs Under pSychedelics (REBUS) model frames these effects within a Bayesian predictive-processing account: serotonergic psychedelics induce an ‘‘entropic’’ brain state that relaxes the precision-weighting of high-level predictive beliefs, which should be experienced subjectively as reduced confidence in beliefs or assumptions. Within this transient state of enhanced neural and cognitive plasticity, the model proposes that maladaptive, overweighted beliefs (for example, negative self-perceptions) become amenable to revision, potentially yielding therapeutic benefit. This study set out to test two psychological assumptions derived from REBUS. First, the investigators examined whether psilocybin acutely reduces confidence in personally held negative self-beliefs (belief relaxation) and whether such reductions persist 4 weeks later (belief revision, here termed REBAS). Second, they tested whether any relaxation and revision in negative self-belief confidence is associated with (a) neurobiological markers of REBUS—specifically increases in neural entropy measured with EEG, (b) the intensity of acute unitive experiences, and (c) changes in well-being four weeks after dosing. The study focused on healthy, psychedelic-naïve volunteers and used a within-subjects dosing contrast between a subthreshold (1 mg) and an active (25 mg) psilocybin dose.

The study used a single-blind, fixed-order, within-subjects design in which 11 healthy, psychedelic-naïve volunteers (4 female; mean age 42 years, SD 10.12) received two oral psilocybin sessions spaced four weeks apart: a 1 mg session first (intended as a sub-perceptual control) followed by a 25 mg session. Inclusion criteria required physical and mental health, English comprehension and minimal recent psychedelic exposure; exclusion criteria included current or past psychiatric disorder, significant medical conditions, pregnancy, excessive substance use and lack of email access. All participants provided written informed consent and sessions followed standard preparation and integration procedures in a comfortable, dimly lit setting with a music playlist, eyeshades and two guides present. Belief confidence was assessed using an idiographic approach: participants identified personally salient negative and positive beliefs and rated how confident they were that each belief was true on a 0–100 scale before, during (acute) and four weeks after dosing sessions. Acute unitive experience was measured using the oceanic boundlessness subscale of the Altered States of Consciousness questionnaire, and well-being was assessed with the 14-item Warwick–Edinburgh Mental Well-being Scale (WEMWBS) at baseline and four weeks after dosing. Resting-state EEG was recorded during dosing using a 24-channel wireless cap (DSI-24; 21 active channels) with Pz and FPz as reference and ground. EEG data were cleaned, low-pass filtered at 100 Hz, subsampled at 200 Hz and used to estimate neural signal entropy via the Context Tree Weighting (CTW) algorithm, an entropy estimator related to but distinguished from Lempel–Ziv complexity. For statistical analysis the investigators combined frequentist and Bayesian approaches. Changes in belief confidence across time and dose were examined with repeated-measures general linear model ANOVAs, with follow-up pairwise comparisons; Cohen’s dz quantified repeated-measures effect sizes. Pearson correlations tested associations between changes in belief confidence and (a) oceanic boundlessness and (b) changes in well-being. Because of limited available EEG data, no formal inferential tests were pre-specified for the CTW-based entropy measures; instead, exploratory Pearson correlations examined relationships between entropy at several time points (1, 2 and 4.5 hours post-administration) and changes in negative self-belief confidence. Bayesian analyses (Bayes Factors) accompanied key tests to quantify evidence for or against effects given the small sample.

Sample and general approach: Eleven healthy participants completed both sessions and provided self-report and (for most) EEG data. Repeated-measures ANOVAs and follow-up comparisons examined belief-confidence changes across the 1 mg and 25 mg sessions; effect sizes and Bayes Factors were reported alongside p-values. Primary behavioural outcome—negative self-belief confidence: Administration of 25 mg psilocybin was associated with decreases in confidence in negative self-beliefs measured four weeks after dosing, whereas the 1 mg session produced no such change. Acute decreases in negative self-belief confidence following 25 mg showed a moderate effect size but were not statistically significant in the acute window. For positive self-beliefs, the investigators observed increases in confidence at four weeks after 25 mg; acute changes were not statistically significant. Changes in other-belief categories were generally smaller or absent, with evidence against change for some categories, suggesting the effect was not globally applied across all belief types. Associations with unitive experience: The relationship between oceanic boundlessness and decreases in negative self-belief confidence after 25 mg grew stronger over time. Acutely, the correlation between oceanic boundlessness and immediate decreases in negative self-belief confidence was r = -0.502 (p = 0.116; BF10 = 1.374), which did not reach significance. At four-week follow-up the association was larger and significant (r = -0.725, p = 0.012; BF10 = 4.750). By contrast, oceanic boundlessness showed small, non-significant associations with belief changes after 1 mg (BFs ≈ 0.6 in favour of the null). Associations with well-being: Decreases in negative self-belief confidence after 25 mg were strongly associated with increases in well-being at four weeks. Acute decreases in negative self-belief confidence correlated with later increases in WEMWBS scores (r = -0.815, p = 0.002; BF10 = 11.368), and decreases measured at the 4-week follow-up correlated even more strongly with well-being change (r = -0.870, p < 0.001; BF10 = 24.808). For the 1 mg session correlations were small and non-significant, with Bayes Factors generally not supporting an effect. Associations with neural entropy: Exploratory correlations linking CTW-derived neural entropy and changes in negative self-belief confidence after 25 mg showed medium-to-large negative associations at multiple acute time points. Decreases in acute negative self-belief confidence correlated with entropy at 1 hour (r = -0.837; BF10 = 9.640), 2 hours (r = -0.785; BF10 = 4.067) and 4.5 hours (r = -0.603; BF10 = 2.121) post-25 mg. Neural entropy measured acutely also related to decreases in negative self-belief confidence observed at 4 weeks, most strongly at 1 hour (r = -0.721; BF10 = 3.584), with weaker evidence at 2 hours (r = -0.652; BF10 = 1.928) and none at 4.5 hours (r = -0.385; BF10 = 0.948). In contrast, corresponding correlations after 1 mg were small and accompanied by Bayes Factors indicating anecdotal support for the null. The authors note that inferential EEG analyses were limited by the small number of subjects with usable EEG data and therefore treated these results as exploratory.

Zeifman and colleagues interpret the findings as preliminary psychological support for key REBUS hypotheses: that psychedelics can acutely relax the felt confidence of strongly held beliefs and facilitate post-acute revision of maladaptive beliefs, and that these processes relate to both subjective phenomenology and neurophysiological indices. Specifically, the study found that a clinically relevant 25 mg psilocybin dose produced decreases in confidence for self-identified negative self-beliefs measured four weeks after dosing, while a subthreshold 1 mg dose did not. The investigators also observed that the intensity of the acute unitive experience and increases in neural entropy during the acute state were associated with both immediate and sustained reductions in negative self-belief confidence, and that these reductions correlated with meaningful increases in well-being four weeks later. The authors situate these results relative to earlier research showing reductions in pessimism and hopelessness after psychedelic administration and extend prior work by using an idiographic measure of personally salient beliefs. They suggest that relaxation of high-level priors may sensitise individuals to new evidence, enabling belief updating that—when the context is supportive—tends to be adaptive. The unitive experience is proposed as a phenomenological proxy for the entropic state conducive to belief relaxation; the data showed that stronger unitive experiences were associated with larger later decreases in negative self-belief confidence. Key limitations acknowledged by the investigators include the small sample size of healthy volunteers, which limits generalisability to clinical populations and increases uncertainty in effect estimates. Blinding is also a challenge given the overt subjective effects of psilocybin; the study mitigated expectancy to some extent by enrolling psychedelic-naïve participants and including a very low-dose control, but the authors recommend future studies employ other active control agents and a wider range of doses to better characterise dose–response relationships. Finally, the authors note that psychological processes mediating the transition from acute belief relaxation (REBUS) to sustained belief revision (REBAS)—for example reflective re-evaluation and integration—remain to be elucidated. To aid future research, the investigators present the Relaxed BEliefs Questionnaire (REB-Q) and its self-report variant as structured tools for measuring idiographic belief confidence change across studies. They conclude that, while the current study provides initial empirical evidence linking psilocybin-induced belief relaxation and revision to neural entropy, unitive experience and improved well-being, replication in larger and clinical samples is required to confirm and extend these findings.

In this preliminary within-subjects study, administration of 25 mg psilocybin to healthy, psychedelic-naïve volunteers was associated with specific decreases in confidence in personally identified negative self-beliefs that persisted four weeks after dosing, and these decreases were strongly related to acute unitive experience, increases in neural entropy and improvements in well-being. The authors view these results as initial psychological support for the REBUS-to-REBAS framework but emphasise the need for replication, larger samples and clinical investigations, and further work to characterise the psychological processes that mediate acute belief relaxation into durable belief revision.