This secondary analysis from a DMT study explores the impact of intentional cognitive interruptions on psychedelic experiences. The study investigates whether increasing cognitive load during the experience affects subjective ratings, hypothesizing that higher task demands would lower these ratings. Additionally, it examines whether reduced task demands correlate with larger reductions in long-term depressive symptoms.
Under the psychedelic therapy paradigm, instead of looking at drug efficacy, researchers look at experience efficacy, defined as how certain experiences can be therapeutic. To test experience efficacy, researchers need to develop new research tools to manipulate the experience without changing the pharmacology. We suggest that intentional cognitive interruptions can help inquire into experience efficacy by experimentally interfering with the experience. To strengthen this suggestion, we present a secondary analysis from our 2023 N,N-dimethyltryptamine (DMT) study, investigating whether it is possible to interrupt the psychedelic experience by increasing cognitive load and whether an interrupted experience is associated with reduced long-term mental health changes. We hypothesized that subjective ratings of the psychedelic experience would be lower when task demands were higher and reductions in long-term depressive symptoms would be larger with fewer task demands during the experience.
Psychedelic therapy research increasingly treats the subjective experience induced by a psychedelic as the therapeutic agent, shifting focus from drug pharmacology to what the authors call “experience efficacy.” Previous work has linked particular experiential qualities — for example feelings of unity, spiritual-type experiences, insightfulness and altered meaning — with therapeutic outcomes, but methods to manipulate those experiences experimentally without changing the drug itself are limited. Roseman and colleagues propose that intentionally interrupting an ongoing psychedelic experience with cognitive tasks could serve as a tool to test experience efficacy. To illustrate this approach, they present a secondary analysis of data from their 2023 DMT study, testing whether increasing in-session cognitive load (operationalised here by asking participants to provide verbal intensity ratings every 60 seconds) attenuates subjective experience and whether such interruptions are associated with smaller improvements in depressive symptoms two weeks later. They hypothesised lower subjective experience ratings under higher task demands and larger reductions in depressive symptoms when task demands were absent during the experience.
Papers cited by this study that are also in Blossom
Schenberg, E. E. · Frontiers in Pharmacology (2018)
Timmermann, C., Roseman, L., Haridas, S. et al. · PNAS (2023)
Studerus, E., Gamma, A., Vollenweider, F. X. · PLOS ONE (2010)
Timmermann, C., Zeifman, R. J., Erritzoe, D. et al. · Scientific Reports (2024)
This work is reported as a quality improvement study, received ethical approval from the National Research Ethics Committee London‑Brent and the Health Research Authority, and follows the SQUIRE guideline. Healthy volunteers provided informed consent and attended two testing days separated by two weeks. On each testing day participants underwent two intravenous administrations (DMT or placebo) while being scanned with combined functional magnetic resonance imaging and electroencephalography in a randomised order. Each participant completed two scan conditions under the influence of DMT: a “rating” scan, in which they verbally rated subjective intensity of drug effects every 60 seconds in real time while lying with their eyes closed, and a “no rating” scan, in which they rested with eyes closed and did not provide in‑session ratings. Subjective experience was also measured retrospectively after drug effects subsided using the 11 Dimensions Altered States of Consciousness questionnaire (11D‑ASC). Depressive symptoms were assessed with the Quick Inventory of Depressive Symptomatology (QIDS) one day before and two weeks after DMT. To examine the effect of the in‑session interruption, the investigators compared within‑participant total ASC scores between rating and no‑rating scans and conducted exploratory analyses of the 11D‑ASC subfactors. For change in depressive symptoms they minimised potential carryover by analysing only participants’ first DMT session and tested time × group interactions using repeated‑measures analysis of variance. Details of some procedures are reported in a supplement, which was not fully reproduced in the extracted text.
Twenty volunteers were analysed (mean [SD] age 33.5 [7.9] years; 7 female, 35%). Using the total 11D‑ASC score, the investigators found a higher mean (SD) ASC score in the no‑rating condition (0.36 [0.08]) than in the rating condition (0.29 [0.10]); this difference was statistically significant (P < .001, within‑participant paired t test). In exploratory analyses of the 11D‑ASC subfactors, the authors report that not all dimensions differed equally by condition; dimensions previously associated with therapeutic effects (for example unity, spiritual experience, blissful state, insightfulness and changed meaning of percepts) tended to score higher in the no‑rating condition (the paper refers to a figure for these comparisons). To evaluate effects on depressive symptoms while avoiding carryover, only the first DMT session was analysed: 12 participants received DMT first in the rating condition and 8 received DMT first in the no‑rating condition. Repeated‑measures analysis of variance showed a significant group × time interaction (F1,18 = 5.9; P = .03; partial eta squared ηp2 = 0.247), indicating differential change in QIDS scores. The mean (SD) reduction in QIDS two weeks after DMT was larger following the no‑rating session (−1.25 [1.04]) than following the rating session (−0.17 [0.94]) (P = .01). Baseline QIDS scores did not differ significantly between groups (no‑rating 3.12 [1.46] v rating 1.92 [1.78]; P = .12, unpaired t test).
The investigators interpret these secondary‑analysis results as preliminary evidence that contextual manipulation of the psychedelic session — here, interrupting the experience with frequent subjective intensity ratings — can attenuate the reported acute phenomenology and may reduce short‑term improvements in depressive symptoms. They present this approach as a potential experimental tool to probe ‘‘experience efficacy’’ without altering the pharmacological intervention. The authors acknowledge important limitations: the sample was small, the interruption used (verbal ratings every minute) was relatively mild as a cognitive manipulation, and the analysis was secondary to a study not originally designed to test this hypothesis. They therefore recommend that future studies employ more demanding, intentional interruptions to test the concept rigorously. Finally, they suggest that incorporating contextual manipulations into future psychedelic research could help to examine both safety and efficacy, and to clarify the role of particular experiential dimensions in therapeutic outcomes.
This quality improvement study was approved by the National Research Ethics Committee London-Brent and Health Research Authority and is reported following the SQUIRE reporting guideline. Participants underwent 2 scans under the influence of N,N-dimethyltryptamine (DMT) while lying with eyes closed, performing ratings of subjective intensity every 60 seconds in 1 scan. Subjective ratings were measured retrospectively using the 11 Dimensions Altered States of Consciousness questionnaire (eMethods in Supplement 1). With the exception of elementary imagery, significantly diminished dimensions of experience were those previously found to be associated with therapeutic outcomes of psychedelic therapy (experience of unity, spiritual experience, blissful state, insightfulness, and changed meaning of percepts). Participants provided informed consent (eMethods in Supplement 1). Healthy volunteers participated in 2 testing days separated by 2 weeks. On each testing day, participants underwent 2 separate intravenous DMT or placebo administrations while being scanned with combined functional magnetic resonance imaging and electroencephalography in a randomized fashion.In the rating scan, participants were asked to verbally rate the subjective intensity of drug effects every minute in real time, while the no rating scan had no ratings and participants merely rested. Therefore, we were able to assess the association of interruptions from the rating procedure with the DMT experience. Subjective ratings of the experience were measured retrospectively using the 11 Dimensions Altered States of Consciousness (11D-ASC) questionnaireafter drug effects subsided. The Quick Inventory of Depressive Symptomatology (QIDS) was measured 1 day before and 2 weeks after DMT.
The analysis included 20 volunteers (mean [SD] age, 33.5 [7.9] years; 7 female [35.0%]) based on previous publications.Using the total ASC score, we confirmed that the total mean (SD) ASC score was higher for the no rating (0.36 [0.08]) than rating (0.29 [0.10]) condition (P < .001; withinparticipant paired t test). In further exploratory analysis of 11D-ASC factors, not all factors had equal differences by rating group. Factors previously identified as moderators of clinical efficacyhad higher scores in the no rating condition (Figure ). We further tested if decreases in depressive symptom ratings after DMT were larger for the no rating condition. To avoid carryover effects, we looked only at changes after the first scan and compared the 12 participants who received DMT first in the rating condition with 8 participants who received DMT first in the no rating condition. Repeated-measure analysis of variance was used to test for an interaction between time (before vs after DMT) and group (rating vs no rating). An interaction was observed for group × time [F 1,18 = 5.9; P = .03; ηp 2 = 0.247], with a greater mean (SD) reduction in QIDS after the no rating (-1.25 [1.04]) than rating (-0.17 [0.94]) scan (P = .01). No significant differences were observed for baseline mean (SD) QIDS measures of the no rating (3.12 [1.46]) compared with the rating (1.92 [1.78]) group (P = .12; unpaired t test).
This quality improvement study's results suggest that contextual manipulation, such as cognitive interruption (via giving subjective ratings in this study), may be used to study experience efficacy. Conclusions are limited by a small sample size, relatively mild interruption, and status as a secondary analysis of data not intended to test this hypothesis. For future studies, we suggest that an intentional interruption should be more cognitively demanding than merely asking for ratings. We hope that future psychedelic studies will include contextual manipulation as a procedure to examine safety and efficacy.
Create a free account to open full-text PDFs.
Kangaslampi, S. · Journal of Psychedelic Studies (2023)
Roseman, L., Nutt, D. J., Carhart-Harris, R. L. · Frontiers in Pharmacology (2018)
Papers in Blossom that reference this study
Marek, G. J., Makai-Bölöni, S., Umbricht, D. et al. · Journal of Psychopharmacology (2025)
Egger, K., Redondo, J. J., Müller, J. et al. · Biomedicine & Pharmacotherapy (2025)
Sanders, J. W., Milliere, R., Daily, Z. G. et al. · Preprints (2024)