This within-subject MRI study (n=11) finds that inhaled DMT increases the mean population receptive field (pRF) sizes in the peripheral visual field of the primary visual cortex (V1). Documented by the Hallucinogen Rating Scale (HRS), this effect explains visual perceptual distortions like field blurring and tunnel vision, and supports the role of 5-HT2A receptor activation in controlling visual cortex activity.
N, N-dimethyltryptamine (DMT) is a psychedelic tryptamine acting on 5-HT2A serotonin receptors, which is associated with intense visual hallucinatory phenomena and perceptual changes such as distortions in visual space. The neural underpinnings of these effects remain unknown. We hypothesised that changes in population receptive field (pRF) properties in the primary visual cortex (V1) might underlie visual perceptual experience. We tested this hypothesis using magnetic resonance imaging (MRI) in a within-subject design. We used a technique called pRF mapping, which measures neural population visual response properties and retinotopic maps in early visual areas. We show that in the presence of visual effects, as documented by the Hallucinogen Rating Scale (HRS), the mean pRF sizes in V1 significantly increase in the peripheral visual field for active condition (inhaled DMT) compared to the control. Eye and head movement differences were absent across conditions. This evidence for short-term effects of DMT in pRF may explain perceptual distortions induced by psychedelics such as field blurring, tunnel vision (peripheral vision becoming blurred while central vision remains sharp) and the enlargement of nearby visual space, particularly at the visual locations surrounding the fovea. Our findings are also consistent with a mechanistic framework whereby gain control of ongoing and evoked activity in the visual cortex is controlled by activation of 5-HT2A receptors.
DMT (N,N-dimethyltryptamine) is a classical tryptamine psychedelic that produces intense visual hallucinations and marked alterations in visual perception. Earlier studies have described phenomena such as enlargement of perceived nearby space, blurring and difficulty focusing, and a bias towards central (foveal) versus peripheral processing, but the neural mechanisms that could account for these perceptual distortions remain unclear. The population receptive field (pRF) mapping approach with fMRI permits estimation of the spatial tuning of neural populations across retinotopic cortex and therefore offers a way to test whether changes in cortical receptive field properties could underlie psychedelic visual phenomena. Pais and colleagues set out to test the hypothesis that acute inhaled DMT produces rapid changes in pRF properties of early visual cortex that would be consistent with perceptual distortions such as peripheral blurring and “tunnel vision”. They used a within-subject design (control and DMT sessions) with pRF mapping in V1 and concurrent subjective assessment with the Hallucinogen Rating Scale (HRS) to relate physiological changes to reported visual experience. The study aims to determine whether mean pRF sizes in V1 differ between conditions across eccentricity and whether such differences align with participants’ reported visual effects.
Papers cited by this study that are also in Blossom
Aday, J. S., Bloesch, E. K., Wood, J. R. et al. · Reviews In The Neuroscience (2021)
De Araujo, D. B., Ribeiro, S., Cecchi, G. A. et al. · Human Brain Mapping (2011)
Castelhano, J. M., Lima, G. M., Teixeira, M. et al. · Frontiers in Pharmacology (2021)
Johnson, M. W., Richards, W. A., Griffiths, R. R. · Journal of Psychopharmacology (2008)
Eleven healthy, experienced psychedelic users (four female, mean age 37 ± 12.4 years) completed a within-subject study comprising a control session and an active inhaled DMT session on different days. Exclusion criteria included personal or family history of major psychiatric disorders, neurological disease, significant medical comorbidity, current psychoactive medications, and pregnancy. The sessions were not counterbalanced across subjects (control and active order is not clearly reported), but the investigators introduced a long washout period between visits to reduce carry-over; the extracted text does not clearly state the exact inter-session interval beyond an imprecise “4 to weeks” phrase. Ethical approval and written informed consent were obtained. For the active condition participants self-administered a changa preparation (Mimosa hostilis extract) via inhalation immediately before MRI. HPLC-DAD quantification on the changa showed 30.92% DMT by the reported procedure; participants estimated pipe loads of ~50–70 mg. MRI acquisition was timed to the pharmacokinetics of inhaled DMT: resting-state and functional/anatomical blocks were acquired first and pRF mapping was performed after the expected subjective peak, around 35 minutes post-inhalation. The HRS questionnaire and a phenomenological debriefing were administered after effects subsided. MRI data were collected on a 3T Siemens MAGNETOM Prisma with a 64-channel head coil. Structural T1-weighted images were acquired at 1 × 1 × 1 mm3. Functional T2*-weighted EPI data were acquired at approximately 2.0 × 2.0 × 2.0 mm3 voxel size with TR = 2000 ms, TE = 30 ms and 29 slices. Eye position and fixation were continuously monitored inside the scanner using EyeLink hardware (EyeLink 1000 Plus, sample rate 500 Hz); sessions with inadequate fixation were excluded (two sessions excluded from pRF analysis for eye movement). Head motion was monitored and a cut-off of ≤ 3 mm translation was applied. Visual stimuli for pRF mapping consisted of high-contrast checkerboard bars moving in two perpendicular orientations, presented on an LCD viewed via mirror; two runs (~5 minutes each) were obtained per session. Participants maintained central fixation on a dot that changed colour intermittently. pRF estimation was performed in BrainVoyager using a circular two-dimensional Gaussian model: predicted neural responses (over 174 binary stimulus frames) were convolved with a canonical two-gamma HRF to model the BOLD time course. Voxels were retained if model explained variance exceeded R2 > 0.09. V1 was delineated manually from polar angle maps and used as the region of interest. The primary physiological measure was mean pRF size across eccentricity. Eccentricity was divided into four non-overlapping 2° bins starting at 1°; mean pRF sizes were evaluated at central points 1.25°, 3.75°, 6.25° and 8.75°. Statistical analyses comprised a two-way repeated-measures ANOVA (factors: Condition [Control, DMT] × Eccentricity bin [4 levels]), post-hoc paired sample t-tests (two-tailed), and Ordinary Least Squares linear regression to compare relations of pRF size with eccentricity between conditions. HRS perceptual scores and a visual-domain subset (items 57–68) were analysed with repeated-measures ANOVA.
Mean pRF sizes in V1 increased with eccentricity as expected, and inhaled DMT produced a statistically significant overall increase in mean pRF size relative to control. The two-way repeated-measures ANOVA showed a main effect of eccentricity bins (F(3,24) = 20.457, p = 8.448 × 10⁻7) and a main effect of Group/Condition (F(1,8) = 21.702, p = 0.002). Post-hoc paired t-tests identified significant differences in the peripheral bins (3.75°, 6.25° and 8.75°), indicating that the DMT-related pRF enlargement was concentrated in parafoveal and perifoveal locations (> ~3° eccentricity). Complementary OLS regression of mean pRF size versus eccentricity found significant differences between conditions in both intercepts (t(16) = 5.498, p = 2.431 × 10⁻5) and slopes (t(16) = 3.402, p = 0.002), reinforcing that the spatial profile of pRF sizes across eccentricity differed under DMT. The pRF model goodness-of-fit metric met the inclusion threshold (R2 > 0.09) and no significant difference in model fit between conditions was reported. Subjective measures aligned with physiological changes. The HRS perceptual dimension differed strongly between DMT and control (repeated-measures ANOVA: F(1,8) = 50.07, p = 1.04 × 10⁻4). Analysis of the visual-domain-specific HRS items (items 57–68) also showed a marked effect (F(1,9) = 46.434, p = 7.8 × 10⁻5), and a focused test on item 60 (object visual distinctiveness) yielded F(1,7) = 18.778, p = 0.003. Participant debriefing described visual phenomena consistent with the physiological findings: 4 of 11 reported pronounced tunnel vision (sharp centre, blurry surround), 7 of 11 reported increased brightness of objects, 3 of 11 noted sharper object distinctiveness, 4 of 11 reported visual distortions, and several participants reported changes in perceived dimensionality (3 reported 3D, 2 reported 2D, 4 reported multi-dimensional vision). Quality-control analyses indicated no significant differences between control and DMT sessions in head-motion parameters or quantitatively derived eye-angle metrics. The investigators excluded two sessions from pRF analysis due to excessive eye movements identified in real-time video monitoring. Sample size calculations (reported from prior data) had estimated a required within-subject group size of 10 for an effect size of 1.2 with 95% power; the study analysed data from 11 participants with the noted exclusions.
Pais and colleagues interpret the observed rapid increase in mean pRF size in parafoveal and perifoveal V1 under inhaled DMT as a plausible neural substrate for characteristic psychedelic visual distortions, particularly peripheral blurring and tunnel vision. Larger pRFs imply lower spatial resolution in affected regions, which the authors argue would produce a low-pass filtering effect in peripheral vision and a relative perceptual sharpening of central vision; this mechanism could also account for altered perceived object size (micropsia/macropsia) and curvature or tilt of nearby visual space. The results are positioned relative to earlier behavioural and clinical observations of distorted visual space under psychedelics and in clinical conditions associated with tunnel-vision-like phenomena (for example, reports in psilocybin states, autism and glaucoma). The investigators note consistency with prior work indicating a high density of 5-HT2A receptors in early visual cortex and hypothesise that serotonergic modulation — specifically activation of 5-HT2A receptors — could alter gain control of ongoing and evoked activity in visual cortex, producing the pRF changes observed. They cite the need for future studies to link these pRF changes more directly to receptor-level effects, including molecular imaging and exploration of extrastriate areas involved in disparity and depth processing. Several limitations and uncertainties are acknowledged. The extracted text indicates that session order was not counterbalanced across subjects, and although a long washout period was used this remains a design limitation. Potential nuisance factors such as brain pulsatility and optical defocus are recognised as possible sources of noise in retinotopic estimates, even if head and eye movements were monitored and found not to differ between conditions; pupil size changes early after administration are discussed as unlikely to affect pRF measures taken after the subjective peak (~35 minutes). The sample size is small and comprised experienced users, which may limit generalisability. The authors also recommend further investigation of extrastriate cortex and models proposing reduced bottom-up sensory drive, referencing animal work showing reduced surround suppression in V1 as a potentially related mechanism. In their summary the investigators conclude that inhaled DMT produces rapid, spatially selective increases in pRF size in early visual cortex that correlate with subjective reports of peripheral blur and tunnel vision, and that serotonergic modulation via 5-HT2A receptors offers a mechanistic framework for these effects.
N, N-dimethyltryptamine (DMT) is an indole alkaloid widely found in nature, best known for its presence in Ayahuasca. The psychoactive properties of this classic psychedelic are mediated primarily via the serotonergic pathway and 5-HT2A serotonin receptor agonism. It is a psychedelic tryptamine associated with intense visual hallucinatory phenomena, perceptual changes and profound spiritual experiences. Alterations in visual processing induced by psychedelics are the most marked and characteristic perceptual effects of these states). One of the first studies investigating these alterations found an enlargement of nearby visual space in psilocybin-induced states, inferred from changes of apparent frontal plane curvature and an enlargement of the handwriting area. These changes in perceived visual space may be attributed to changes in receptive field properties of cortical neurons, generating the question of whether these changes lead to local alterations in cortical Abbreviations: DMT, N-N-dimethyltryptamine; fMRI, functional magnetic resonance imaging; HPLC-DAD, high-performance liquid chromatography; HRS, Hallucinogen Rating Scale; LOD, limit of detection; LOQ, limit of quantification; MRI, magnetic resonance imaging; OLS, Ordinary Least Squares; pRF, population receptive field; ROI, region of interest; V1, primary visual cortex. magnification that would lead to perceived visual distortions in space. Other visual alterations reported in psychedelic states include blurring and difficulty in focusing) with a bias for peripheral vision and visual distortions, which might also be explained by the changes in the filtering properties of visual receptive fields. Collectively, these reports suggest that psychedelic states induce changes in visual perception in retinotopic space. These aspects raise the question of whether the physiological properties of visual cortical responses and topological changes can explain the underlying phenomenology. Early visual areas can be mapped and delineated using functional magnetic resonance imaging (fMRI) retinotopic approaches. More recently, an alternative technique to conventional retinotopy that employs a model-driven approach to estimate neuronal population receptive field (pRF) responses and visual field maps has been proposed. This model is based on earlier pioneering work that originally conceptualised the pRF approach for neuroelectrical recordings in the visual cortex. Briefly, the pRF approach estimates a neural response for each voxel that best explains the cortical visual field responses to a wide range of stimulus positions. pRF mapping analysis with fMRI provides important insights into visual cortical functional properties. pRF size properties in early visual field maps have been well characterised, increasing with visual eccentricity within a visual field map and hierarchically across visual areas. Thus, pRF sizes are smaller where visual acuity is higher: near the foveal regions of the cortex. Human studies investigating visual neuronal changes in pRF sizes are scarce. Larger pRF sizes have been reported in autism spectrum disorder (specifically in the extrastriate cortex), ageing, after cataract surgery) and in glaucoma. Interestingly, two of the aforementioned studies include clinical conditions with a visual phenomenon in common, which is also one of the visual distortions induced by psychedelics: tunnel vision. People with autism are often described as having tunnel vision, being attracted to details of a visual scene while neglecting surrounding stimuli. Regarding glaucoma, tunnel vision (loss of peripheral vision) is a symptom of the advanced stages of this disease. Therefore, investigating a fundamental biological property of the visual cortex can help explain this visual phenomenon induced by psychedelics. Less sharp vision in the periphery may lead to relatively increased perceptual sharpness in the centre. Here, we used pRF analysis to test the hypothesis that changes in population receptive field properties (pRF size) in the primary visual cortex (V1) might underlie visual distortions in psychedelic states, particularly at the visual periphery. Anatomical and functional magnetic resonance imaging (MRI) data were acquired in two different sessions, control and inhaled DMT (within-subject study, n=11). In the DMT condition, individuals inhaled approximately 50 to 70 mg of DMT (root bark of Mimosa Hostilis in freebase form) immediately before the acquisition. We found that the pRF size in V1 was significantly increased in peripheral bins of eccentricity in the active condition (inhaled DMT). This novel finding may explain perceptual visual distortions induced by psychedelics, such as field blurring and the enlargement of nearby visual space, particularly at the visual periphery (tunnel vision).
After neuropsychiatric screening and medical evaluation, 11 healthy subjects, verified normal or corrected-to-normal vision (four females, mean age ± SD = 37 ±12.4 years old), were included in this withinsubject study. All individuals were tested at the Institute of Nuclear Sciences Applied to Health (ICNAS) in two sessions on different days for the acquisition of the structural and functional MRI. Exclusion criteria included psychiatric disease (personal or family history of schizophrenia, bipolar disorder, mania or hypomania); neurological disease; cardiac or liver dysfunctions; immunosuppression; use of antihypertensive, sympathomimetic, psychoactive, antidepressant and benzodiazepine drugs; and pregnancy, suspected pregnancy or breastfeeding. Written informed consent was given for the study, which was conducted according to the Declaration of Helsinki and subsequent revisions. Ethical approval was obtained from the ethics committee of the Faculty of Medicine of the University of Coimbra. Critically, in our withinsubject study design, the same individuals were submitted to the active and control sessions within a time interval of weeks (4 to weeks), so variations such as age and visual acuity were minimised. To overcome the limitation that the active and control sessions were not counterbalanced across subjects a long washout period was introduced between sessions (see below).
We used the Mini-International Neuropsychiatric Interview (M.I.N. I.), a short structured diagnostic psychiatric disorders interview of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV)) and the International Classification of Diseases 10th Revision (ICD-10). The Portuguese version of the Mini-Mental State Exam (MMSE)was used for the screening of cognitive impairment, and the GRAFFARD SCALE to measure socioeconomic status).
To study the psychedelic-induced visual and behavioural manifestations, we used the Hallucinogen Rating Scale (HRS). This assessment instrument is a questionnaire consisting of items and employs a rating scale from 0 to 4. Its purpose is to evaluate the subjective effects of psychedelic substances across six categories: somaesthesia, affect, perception, cognition, volition, and intensity. The HRS was administered following the MRI procedure when psychoactive effects were absent, and participants were able and willing to talk clearly and share their experiences with the team members. A phenomenological debriefing of the experience was also conducted, emphasising its visual aspects.
An ultrasound-assisted extraction was performed on 10 mg of changa, in the presence of 100 μL of methanol (for 3 times, 15 min each). The quantification of DMT present in each changa sample was performed on high-performance liquid chromatography (HPLC-DAD). The mobile phase was composed of methanol, and 0.3 % trifluoroacetic acid in water. The elution was carried out in gradient mode from 30 % methanol (0 min) to 55 % methanol (10 min) and again to 30 % methanol (10-15 min). The time of analysis was 15 min, including the stabilisation of the RP-18 column, and the temperature was set at 25 • C. The flow rate was 1 mL/min, the injection volume was 5 µL and the detection wavelength was 278 nm. A triplicate of the sample (at a concentration of 1 mg/mL) was performed, using methanol as the solvent. DMT reference standard was prepared in a concentration range of 0.05-0.01 mg/mL. Compound identification was based on the comparison of retention time and ultraviolet (UV) spectra overlay with authentic standards. The limit of detection (LOD) and limit of quantification (LOQ) were calculated based on the standard deviation error of the response and the slope. LOD and LOQ were calculated as 3.3σ/S and 10 σ/S, respectively, where σ is the standard deviation error of the response and S is the slope of the calibration curve. The sample showed the presence of 30.92 % of DMT. The LOD and LOQ were found to be 0.868 µg/mL and 2.631 µg/mL respectively. See Figs.andfor details on calibration.
In this study, the participants were experienced in altered states of consciousness, particularly DMT inhalation (> 5 Changa sessions). Special attention was given to aspects surrounding the preparation and administration of DMT and session monitoring. Guidelines proposed by. After the initial assessment and inclusion decision, detailed information about session preparation was provided to the participants, namely evidence-based information about the substance, possible risks and benefits. A randomised placebo-controlled study in healthy participants showed that 75-79 % of DMT was rapidly cleared from plasma, with an initial half-life of 5, 6 min. Subjective effects declined in parallel with the rapid half-life within 15 min. Since it typically takes about 5 to 7 half-lives to effectively clear a substance from the organism and reduce it to less than 1 % of its original concentration, DMT substance levels are considered negligible in less than two hours. Vogt et al. () used a wash-out period between sessions of at least 1 week. Participants were instructed for a wash-out period of 2 months regarding recreative drugs and 24 h caffeine, alcohol and tobacco abstinence interval before the session day. To familiarise participants with the MRI environment and, in this way, minimise the probability of anxiety and adverse effects, didactic videos and MRI sounds audio files were pre-assigned to the participants. To promote a safe, cosy and supportive environment, a comfortable, properly decorated private room next to the MRI facilities was prepared for the day of the session. In this setting, participants engaged in their customary rituals before the sessions. These rituals encompassed various elements, such as setting an intention for the experience, engaging in meditation, chant singing, drumming, symbolic actions, and burning plants like Palo Santo and white sage. Moreover, these rituals were performed equally during both the active and control conditions, allowing the participants to prepare for the psychedelic experience in both conditions adequately. A trained team composed of a medical doctor and two psychologists assisted the participants through all stages of the protocol and session day. For the active session, participants self-administered inhaled DMT obtained by extraction from the Mimosa hostilis Benth. synonym of Mimosa tenuiflora (Willd.). The percentage of DMT present in the sample was quantified by HPLC-DAD, revealing a percentage of 30.92 %. The average load of the pipes was estimated by the participants to be approximately between 50 and 70 mg. MRI data was acquired immediately after inhalation. MRI acquisition protocol was designed according to the well-known peak effects profile for inhaled DMT (very intense but short-lived)), with resting-state, functional and anatomical acquisition blocks taking place first, and pRF mapping acquisition after the peak of the subjective effects (around 35min after inhalation). Specific instructions about eye fixation and body movements were given to the participants. Fixation and ocular movements were qualitatively accessed and constantly controlled by video-based EyeLink® software, SR Research Ltd. Sessions where participants were not able to follow central fixation instructions, assessed by real-time video monitoring, were excluded from data analysis. Sample size calculation was based on data from another case control study from our laboratory. Using G-Power for an effect size of 1.2, Power 95 % and one tailed alpha of 0.05, the required group sample size for a within-subject design and a dropout of 2 was 10.
MRI data were acquired using a 3T imaging system (MAGNETOM Prisma, Siemens Medical Solutions) using a 64-channel head-coil. We acquired T1-weighted anatomical MRI data at 1 × 1 × 1mm 3 spatial resolution, Repetition Time 2530ms, Echo Time 3.5ms and Flip Angle 7deg; and two T2*-weighted functional 2D echo planar images at 2.0 × Fig.. DMT calibration curve. The solid line represents the best-fitting linear regression. As described by the equation y = b1x + b0; where y is the area, x is the concentration, and b1 and b0 are the slope and intercept, respectively. Measure of the goodness of fit of a model (R 2 ).
2.0 × 2.0mm 3 spatial resolution, with 29 slices. For this case, the Flip Angle was set to 90deg, the Repetition Time was set to 2000ms, and the Echo Time was set to 30ms. Continuous monitoring of binocular fixation and ocular movements was done using EyeLink® software and hardware inside the MRI scan during all acquisition in both conditions, DMT and control (Eyelink 1000 plus -SR Research, version 5.03 sample rate 500Hz, calibration with 9 points). Two sessions (one from the control and the other from the DMT condition) were excluded from pRF mapping analysis due to eye movements observed during block acquisition. Still, the polar angle maps of these sessions were generated to confirm their visual exclusion. As expected, the maps were noisy and were excluded from the analysis. For details, see Figs. S3 and S4 of the supplementary material.
A 40-inch LCD monitor (Inroom Viewing Device; Nordic-NeuroLab, Bergen, Norway) with 1920 × 1080 pixel resolution was positioned at the end of the scanner bore and used to display the visual stimuli. Participants viewed the visual stimuli through a mirror attached to a head coil. The display (87.8 × 48.5 cm) and the viewing distance of ~168cm covered a visual angle of 27.361 • × 15.316 • . Visual bars moving stimuli followed the protocols implemented in a previous work of our institute. The visual stimuli consisted of two perpendicular bars moving simultaneously in opposite directions, exposing a high-contrast randomly chosen RGB-coloured checkerboard pattern. In the first half of the stimulus run, the horizontal bars move vertically downwards across the display in 4 cycles of 18 equally spaced steps (2 seconds duration). At the same time, the vertical bars moved horizontally leftwards across the display in 3 cycles of 24 equally spaced steps (2 seconds duration). After that, a blank period of mean luminance (zero-contrast) takes place (30 seconds duration) to allow pRF models to determine the baseline fMRI response amplitude. Finally, in the second half of the stimulus run, horizontal and vertical bars performed the same cycles in opposite directions. Two runs of this visual stimuli were acquired (~5min duration each). Participants were instructed to maintain their attention and fixation on a central dot that changed colours at random intervals between red and green while ignoring the surroundings (moving bars). As already mentioned, during all the sessions, a video-based eye tracker (EyeLink) was used by a team researcher to monitor the eye movement, central fixation and attention of the subject (see Fig.of the supplementary material).
Anatomical and functional MRI data processing was performed with Brain Voyager 22.0 (Brain Innovation, Maastricht, The Netherlands). Brain Voyager 22.0 was also used to estimate the pRF model that most closely fitted each voxel response measure and to define the V1 visual field map. Mean pRF sizes were then calculated using MATLAB (MAT-LAB and Statistics Toolbox Release 2021a, The MathWorks, Inc., Natick, Massachusetts, United States) and NeuroElf (BVQXTools). Statistical analyses were performed in IBM SPSS Statistics for Windows, version 25 (IBM Corp., Armonk, N.Y., USA) and Microsoft Excel.
After brain extraction, the intensity inhomogeneities of anatomical data were corrected by estimating bias fields. This correction was performed mainly to avoid negative effects on the outcome of coregistration and segmentation procedures. Later, the volumes were aligned to the Anterior Commissure (AC) -Posterior Commissure (PC) plane and transformed into Talairach coordinates. Thereon, the cerebral cortex was segmented into the cerebral fluid, white matter and grey matter using automatic segmentation tools to create the 3D inflated meshes of each hemisphere for better visualisation of the fMRI data. All 3D inflated meshes of each hemisphere were visually inspected, whilst manual corrections were performed in the occipital cortex. These manual corrections included the remotion of the topological errors that often result from connections across opposite banks of sulci or holes within sulcal banks by adding or deleting voxels. For this purpose, the topological errors were visually identified in the 3D inflated mesh and linked to the white matter and grey matter segmentation results to localise the source of the error. Finally, voxels were added or deleted to correct the topological errors in the source of the error. Regarding functional data, the first 6 timeframes of each run were discarded due to start-up magnetisation transients. In addition, functional MRI data preprocessing consisted of mean intensity adjustment, slice scan time correction, 3D motion correction and temporal filtering (high pass), with an intra-session alignment to the first volume of the first functional acquisition. Finally, anatomical and functional scans were co-registered. This coregistration involved keeping the anatomical data fixed while scaling, translating, and rotating (9 parameters) the functional data to align it into the anatomical space.
Data on quality control of head motion was first assessed using standard criteria for detecting and rejecting head motion, which was described by 6 parameters: three translation (displacement) parameters and three rotation parameters. See Tableof the supplementary material for the individual values of these 6 parameters. In particular, we used a head motion cut off ≤ 3 mm with no notable variation in translation and rotation axis between the two visits as inclusion criteria. Second, we tested these 6 head motion parameters between conditions (control and DMT) using the paired sample T-test. No significant differences between conditions (Control and DMT) for any parameter of head motion were observed (see Tablefor details). Examples of temporal variations in head motion parameters in individual participants are shown in Fig.of the supplementary material. Data on quality control of eye motion were first visually assessed by the research team using a video-based eye tracker (EyeLink) to monitor subjects's eye movements. Second, quantitatively, we also calculated the individual values of the angle of the eyes during the visual task (See Tableof the supplementary material for the individual values) through a freely available method implemented via a MATLAB toolbox capable of extracting eye positioning vector and angle. The full method is described in. After defining a region of interest (ROI) that covered both eyes, the method applied an algorithm to segment the eye bulbs based on their intensity values and extracted a position vector and its variation over time. We then centred this data by subtracting the overall mean gaze position across all data points. Finally, we tested the parameters obtained of the angle of the eyes between conditions (control and DMT) using the paired sample T-test. No significant differences between conditions (Control
Paired sample T-test results comparing the maximum minus the minimum (Max-Min) of 6 head motion parameters between conditions (control and DMT). dx, dy and dz are the translation (displacement) parameters in millimetres (mm) of the axes x, y and z, respectively, and rx, ry, and rz are the rotation parameters in degrees (deg) of the axes x, y and z, respectively. df = degrees of freedom; Sig. = significance value; t = T-test statistics.
The pRF model was estimated from BOLD responses to the simultaneous bars moving stimuli by using a model-driven approach developed in 2008 by. The methods of BrainVoyager QX are described in, which evaluate population receptive field estimation frameworks in terms of robustness and reproducibility. First, a sequence of 174 binary stimulus images with a resolution of 300 × 300 pixels squared was generated, which contained detailed information about the sequence of visual field positions covered by the bars of the stimulus. In the occipital lobe, for a large set of combinations of pRF positions and sizes, we calculated the proportion of the Gaussian (two-dimensional circular) pRF model that overlapped the binary stimulus across time. This proportion determined the predicted neural response amplitude for each candidate pRF and stimulus. Next, each candidate neural response time course was convolved with a canonical two-gamma BOLD hemodynamic response function (HRF) to predict the BOLD response time course. The pRF method returned six statistical maps related to the model fitting procedure: R (squared root of the explained variance), the size (standard deviation or σ), the x position, the y position, the eccentricity and the polar angle for each selected hemisphere of each participant. The explained variance (R 2 ) was used as a quantitative measure of the goodness-of-fit of the model to the BOLD time series data. Therefore, and as a criterion, only voxels with R 2 > 0.09 (R > 0.3) were included in the analysis. Using the paired sample T-test, we also tested the goodness-offit model and found no differences between control and DMT conditions (see Tablefor details). The polar angle allows the definition of the boundaries of the visual areas. Thus, using the coloured polar angle map obtained from the pRF model, the visual area V1 was manually defined for each hemisphere and for each condition (control and DMT) over the inflated meshes using BrainVoyager's drawing tools (see Fig.). The V1 area of each hemisphere, for each participant and for each condition (control and DMT), were then used as masks for the analysis of the mean pRF sizes later on. In Fig., we present an example of the A) sum of the projected Gaussian pRF intensity values (centred in a voxel of interest in V1), B) squared root of the explained variance maps and C) BOLD timeseries (dashed line) of the same voxel of interest and the pRF prediction model (solid line). The same voxel (Talairach coordinates, x = 16; y = -97; z = 7) was used as voxel of interest in V1 for the control and DMT condition of the same subject. Finally, in Fig., we also present an example of the A) sum of the projected Gaussian pRF intensity values (centred in a more peripheral voxel of interest in V1,) and B) the eccentricity maps. In this case, we choose to show two voxels of interest in V1 with approximately the same eccentricity value (≈ 6 deg) for the control and DMT conditions of the same subject.
To obtain the mean pRF size along eccentricity for each participant, we used the same methodology as described in previous works of the institute. In our case, eccentricity was divided into four bins using non-overlapping 2 • intervals starting from 1 • of eccentricity, covering the rangesand]. This approach ensures that each bin reflects a distinct part of the visual field free from the influence of adjacent bins, allowing for a precise analysis of visual processing across different eccentricities. After that, to investigate if populations of visual neurons do change along eccentricity in response to inhaled DMT, we calculated the mean pRF sizes at central points within each bin, specifically at 1.25 • , 3.75 • , 6.25 • , and 8.75 • of eccentricity.
We performed within subject's comparisons using Two-Way Repeated Measure ANOVA [factors: Group (2), Control and DMT, x Bins of eccentricity (4), 1.25 • , 3.75 • , 6.25 • and 8.75 • , followed by post hoc Paired Sample T-Tests (two-tailed). We also used Ordinary Least Squares (OLS) linear regression analysis to investigate the relationship between mean pRF sizes among control and DMT conditions along eccentricity.
Since the sphericity assumption was not violated and the distribution of mean pRF sizes in each bin of eccentricity in the related groups was found to be approximately normal distributed, Two-Way ANOVA was performed to determine if populations of visual neurons do change along eccentricity in response to inhaled DMT. This was indeed the case. Bins of eccentricity (F (3,24) = 20.457, p-value = 8.448x10 -7 ) and Group (F (1,8) = 21.702, p-value = 0.002), with Bonferroni confidence interval adjustment (see Tableof supplementary material for details). Post hoc Paired Sample T-tests were used to determine which pairs of bins of eccentricity were associated with significant differences. We found statistically significant differences along peripheral bins of eccentricity, 3.75 • , 6.25 • and 8.75 • (for details, see Table). Comparisons between changes in the group's mean pRF size across visual field eccentricity were further replicated using OLS linear regression analysis (Fig.). We found significant differences between the intercepts (t-value (16) = 5.498, p-value = 2.431x10 -5 ) and slopes (t-value (16) = 3.402, p-value = 0.002) of groups' linear regressions. Additionally, as expected, our pRF modelling results show increases in pRF size as a function of eccentricity. Individual values of the mean pRF sizes in degrees across visual field eccentricity of V1 are presented in Tableand displayed in Fig.of the supplementary material. The perceptual domain of the HRS, which is used to evaluate the acute effects of hallucinogenic drugs on perception, showed significant within-group differences [DMT vs Control, One-Way ANOVA repeated measures: F (1,8)=50.07, p-value = 1.04 x10 -4 (supplementary material, Fig.and Tablefor ANOVA summary statistics), with Bonferroni adjustment for multiple comparisons (7 subscales) at p-value < 0.007]. This scale was originally designed to measure the whole constellation of psychedelic-induced subjective effects. As such, HRS Perceptual dimension comprises not only visual but also auditory, gustatory, and olfactory experiences. Therefore, we refined the analysis testing for a selected cluster of HRS visual-domain-specific items between conditions (HRS items 57-68). These items included questions regarding changes in object brightness, visual distinctiveness, vibration of visual field, overlaid visual pattern and dimensionality (see supplementary material, Tablefor HRS questions illustration, depicted from the original scale. Results supported the previous effect [DMT vs Control, One-Way ANOVA repeated measures: F (1, 9)=46.434, p-value = 7.8x10 -5 (supplementary material, Fig.and Tablefor ANOVA summary statistics and Fig.and Tablefor participants individual data across conditions). Item 60 specifically addresses the object's visual distinctiveness, so we have also tested for this difference. Once more, results confirmed a significant main effect between conditions [DMT vs Control, One-Way ANOVA repeated measures: F (1,7)=18.778, p-value = 0.003], underlying the distinct perceptual experience induced by inhaled DMT (Supplementary material, Fig.and Tablefor summary statistics and Tablefor individual data overview). This distinct visual phenomenology between the DMT and control conditions was also suggested by participants' qualitative report of their experience (debriefing) shared with the team after the effects subsided. Focusing on the main phenomena of interest, visual perceptual reported changes can be summarised as: "tunnel vision" (4 out 11 participants reported tunnel vision as very sharp at the centre and blurry in the surround); "brightness alterations" (7 out reported objects to be brighter); "distinctiveness alterations" (3 out of reported objects to be sharper); "dimensionality changes" (3 participants reported objects to be 3D. Finally, 2 participants reported 2D dimensionality and 4 participants reported multi-dimensional vision) and visual distortions (4 out 11 participants).
The neural mechanisms underlying the visual effects of tryptamine psychedelics remain unknown. In this work, we propose to investigate a fundamental biological property of the visual cortex to help explain the visual phenomena induced by psychedelics. We explicitly tested the hypothesis that changes in population receptive field properties may explain changes in perceived visual space induced by DMT. Our within-subject design showed that mean pRF size in V1 was statistically significantly higher for parafoveal and perifoveal bins of eccentricity (>3deg of eccentricity) in the active condition (inhaled DMT). This result suggests a fast visual cortical alteration due to inhaled DMT in V1 regions surrounding the fovea. Increases in pRFs size in more parafoveal and perifoveal regions lead to peripheral blur because pRF size is inversely related to resolution (acuity). Accordingly, most of our subjects reported peripheral blurring. Moreover, the higher peripheral blur may lead to a relatively perceived enhanced sharpness in centre vision, which can explain the tunnel vision phenomena. It was previously reported that changes in pRF size also lead to predictable changes in visual magnification, which can be related to the perceptual distortions experienced by DMT, as instantiated in our hypothesis. These results indicate that neuronal representations in the parafoveal and perifoveal space become a low pass filter of the visual world, explaining the central versus peripheral distortions, difficulty in focusing, and perceptual blur. Our findings are also consistent with the altered perception of visual objects as decreased or increased in their perceived versus actual size or with modified angles. To document perceptual effects, we used the HRS, a well-established scale to evaluate the acute effects of hallucinogenic drugs, which was developed to study the effects of DMT. Overall, the results suggest an augmented mismatch between subjective visual experiences and visual percept accuracy, corroborating the results of mean pRF size changes induced by DMT -peripheral blur contrasting with central sharpness leasing tunnel vision. Changes in object size (micropsia and macropsia)might also be related to changes in pRF size and are consistent with the visual distortions reported by our participants, particularly in parafoveal and perifoveal space. Our findings provide a potential explanation for reported alterations in visual processing induced by psychedelics, such as enlargement of nearby visual space, blurring and difficulty in focusingand visual distortions (David E Nichols 2016; Schartner and Timmermann 2020). Our results suggest that in psychedelic states, the subject's visual states change at a very early level, particularly in which concerns processing peripheral visual information. Given the disparity representations in the early visual cortex, this is consistent with observed reports of altered 2D/3D vision. The fact that some subjects reported changes in 3D vision in regions involved in disparity perception suggests that one should focus on differences in extrastriatal cortical activation in the future. These changes are consistent with perceptual misjudgments regarding visual depth and the visual vertical or horizontal plane in tilted body positions due to the emergence of spatial perceptual distortions. Basically, the perceived geometry of nearby visual space is distorted. The apparent fronto-parallel plane may show changes in perceived curvature and tilt. In addition, such aberrations can lead to transformations of perceptual constancies. The psychoactive properties of DMT are mediated primarily via the serotonergic pathway and 5-HT2A serotonin receptor agonism. As such, our findings confirm the suggestion by de) that robust visions induced by Ayahuasca (orally ingested and which contains DMT) may be initiated in the primary visual cortex, which is consistent with the high density of 5HT2A receptors in this region. Another study using Ayahuasca found that the activation of the 5-HT2A serotonin receptor plays a key role in the neurophysiology of visual effects of Ayahuasca in humans. Future studies should address how these early sensory changes relate to the action of DMT in 5-HT2A receptors in the visual cortex. It is known that activation of these receptors is quite specific regarding DMT, in comparison with effects on dopaminergic or adrenergic receptors of other psychedelics). It has recently been shown that gain control of ongoing and evoked activity in the visual cortex can be controlled by serotonergic input, which would provide a mechanistic framework for the effects we observed here and their relation to visual phenomenology observed upon psychedelic intake. Molecular imaging to investigate long-term synaptic plasticity will be instrumental in this aspect. Accordingly, our data are consistent with the notion that mammalian V1 expresses a high density of 5-HT2A receptors (Watakabe) and that activation of these receptors modifies early visual potentials. Future studies should address additional effects in the extrastriate cortex. Finally, given the relevance of explaining away other causes of changes in population receptive sizes, such as head and eye movements, we carefully monitored eye and head movements in our experimentsand provided quantitative inferential analysis of both, which revealed no differences across conditions. Still, we are aware that other nuisance factors, such as brain pulsatility and optical defocus, may add noise to the visual field position estimates, although they remain similar at within-subject level. Regarding the influence of different attentional statesfixation and attention were controlled by eye-movement real-time video monitoring. Changes in pupil size, might be present at the very beginning of the acquisition, but are unlikely at the time we measured pRFs -after the peak of the subjective effects, around 35 min after inhalation. We had only one control and one DMT session that had to be excluded. Moreover, we did not find any systematic change that would be predicted by such drifts. We provide a formal analysis showing that eye and head motion parameters are not related to the observed changes. In sum, the factors mainly affected even within subjects, e.g. eye and head movements could be taken into account. Instead, we found peripheral changes consistent with tunnel vision effects and contraction in visual space. Future studies should address models of hallucinations that propose reduced bottom-up sensory drive since it seems to be a key factor leading to altered perception, as suggested by a recent animal study. This study revealed a net reduction in surround suppression in primary visual cortex. In summary, using a within-subject design, we found that pRF properties of visual neurons show rapid changes in response to inhaled DMT accompanied by significant perceptual changes. We found a centre versus peripheral difference in pRF sizes induced by DMT. This finding was linked to several of the perceptual effects that were reported by our participants using the HRS. Namely, tunnel vision, expressed as centre sharpness versus peripheral blur, and distorted peripheral representations, which also matched the subjective reports of the participants (blurred peripheral versus sharp central vision). Our findings are also consistent with the hypothesis that the activation of the 5-HT2A serotonin receptor plays a key role in the neurophysiology of the visual effects induced by psychedelics.
Create a free account to open full-text PDFs.
Kometer, M., Vollenweider, F. X. · Behavioral Neurobiology of Psychedelic Drugs (2016)
Lawrence, D. W., Carhart-Harris, R. L., Griffiths, R. R. et al. · Research Square (2022)
Nichols, D. E. · Pharmacological Reviews (2016)
Phelps, J. · Journal of Humanistic Psychology (2017)
Rickli, A., Moning, O. D., Hoener, M. C. et al. · European Neuropsychopharmacology (2016)
Strassman, R. J., Qualls, C. R., Uhlenhuth, E. H. et al. · JAMA Psychiatry (1994)
Valle, M., Maqueda, A. E., Rabella, M. et al. · European Neuropsychopharmacology (2016)
Vogt, S. B., Ley, L., Erne, L. et al. · Translational Psychiatry (2023)
Papers in Blossom that reference this study
Lima, G., Soares, C., Teixeira, M. et al. · Scientific Reports (2025)