Conference Agenda

Altered states of consciousness (ASCs) refer to alternate patterns of experience which differ from a normal waking state, such as psilocybin or LSD intake. Psychedelic drugs, however, pose risks, including health concerns and lack of experimental control due to metabolic variability. Multi-modal Ganzfeld (MMGF) is a non-pharmacological method for inducing ASCs, by homogenizing visual and auditory input, leading to pseudo-hallucinations similar to those seen with psychedelics. While previous studies have used different light colors in MMGF, no consensus exists on which wavelength best optimizes pseudo-hallucination intensity. Our study aims to determine the most effective MMGF setup by manipulating light wavelength while maintaining exposure to unstructured white noise. Using a repeated-measures design, 33 participants will undergo 25-minute MMGF sessions with either red or green light. We will assess ASCs using the 5-Dimensional Altered States of Consciousness Rating Scale and a phenomenological interview. Additionally, cardiac, cerebral, and electrodermal activity will be recorded to examine arousal changes associated with different wavelengths. We hypothesize that red light will enhance pseudo-hallucinations more than green light, as suggested by previous studies on perception and arousal. By determining the optimal MMGF parameters, will expect to refine ASC induction methods, providing a controlled, accessible, and safe alternative for studying ASCs in experimental and clinical protocols.

Altered states of consciousness (ASCs) share a common phenomenological structure regardless of induction method, as proposed by the etiology independence hypothesis (Fort, under review; Dittrich, 1998). With increasing interest in psychedelic-assisted psychotherapy, identifying clinically relevant phenomenological dimensions and understanding the degree to which induction methods elicit them is crucial (Fort, under review; Yaden et al., 2024). This preliminary study compared ASCs induced by psilocybin (0.17 mg/kg; Mortaheb & Fort, 2024) and multi-modal Ganzfeld (MMGF) using the 5-Dimensional Altered States of Consciousness (5D-ASC) Rating Scale. No significant difference in general score was observed between psilocybin and MMGF. Dimension-level analysis revealed psilocybin produced greater scores on Oceanic Boundlessness (OBN) and Visionary Restructuralization (VRS). Factor-level analysis revealed psilocybin as significantly higher in Blissful State, Insightfulness, Audio-Visual Synesthesia, and Changed Meaning of Percepts scores. The largest effect size was observed for Changed Meaning of Percepts. No significant differences were found in Elementary or Complex Imagery scores, suggesting MMGF may be a hallucinogenic equivalent to psilocybin. OBN differences align with findings showing it mediates psilocybin's therapeutic effects on depression further supported by the strong correlation its egotropic effects have with its hyperconnected pattern (von Rotz et al., 2023; Mortaheb & Fort, 2024). Differences in VRS are more complex. At the factor level, it seems to be driven by meaning shifts and not the production of content, suggesting an important caveat for the 5D-ASC (Fort, under review). Together, these preliminary findings suggest MMGF as an alternative for studying hallucinations when access to psychedelics is limited.

Visual hallucinations can be phenomenologically divided into those of a simple or complex nature. Both types of hallucinations can be induced experimentally through light manipulations, specifically, through high-frequency, eyes-open flicker (Ganzflicker) and perceptual deprivation (Ganzfeld). We investigate the mechanisms underlying the hallucinations elicited by these methods in a series of behavioural experiments.

In Experiment 1, we show using button presses, drawings, interviews, and retrospective questionnaires of hallucinatory intensity that while simple hallucinations are more frequent during Ganzflicker and therefore primarily bottom-up driven, complex hallucinations are relatively more common during Ganzfeld and are likely driven by top-down processes acting independently of visual stimulation. Having found that simple hallucinations are primarily driven by bottom-up processes, Experiment 2 explores how a shifting weighting between prior knowledge and sensory input may predict complex hallucination formation. Specifically, we tested older individuals, who ostensibly have increased top-down influences on perception due to an accumulation of knowledge and multi-sensory loss. Contrary to our hypotheses, older adults are less prone to complex visual hallucinations than younger adults, suggesting that in our sample, top-down influence accumulated throughout age does not overtly impact the tendency to experience hallucinations with semantic meaning.

Finally, we present preliminary data from Experiment 3, in which we perturb these processes with a low (“micro”) dose of psilocybin. We expect that psilocybin will enhance both simple and complex hallucinations by stimulating cortical pattern formation via 5-HT2A receptors in early visual cortices and inducing cortical noise in higher-order and feedback regions, which may be misinterpreted as signal.

During Ganzfeld stimulation, participants are exposed to unstructured sensory stimuli, such as an edgeless visual field with uniform light and monotonous auditory input. A previous study found that alpha power was stronger during conditions of Ganzfeld stimulation and eyes-closed rest relative to eyes-open rest (Miskovic et al., 2019). This effect has been interpreted to reflect an attentional shift away from externally-generated stimuli, and towards internal mentation. However, alpha power also strengthens with sleepiness, raising the question of whether the changes in alpha power observed during Ganzfeld reflect a shift in attentional orientation or fluctuations in arousal state. In this study, we replicate a previous Ganzfeld paradigm while measuring subjective sleepiness across conditions. Participants (N = 10) underwent resting-state EEG recordings (eyes open and closed) followed by Ganzfeld stimulation, and completed the Karolinska Sleepiness Scale after each five-minute block. Our results show that alpha power changes across conditions in a manner that is similar to the previous study. Importantly, changes in alpha power co-vary with changes in subjective sleepiness across conditions. These findings suggest that arousal may contribute to the changes in alpha rhythms observed during unstructured sensory stimulation. We will examine whether these findings persist with a larger sample size.

Mental health research is increasingly recognising the well-being potential of altered states of consciousness (ASCs) mediated by psychedelics and mind-body practices such as yoga, meditation and breathwork; however, their neurophysiological mechanisms and the direct ways in which these are correlated with subjective experiences requires further research into mechanisms, modes of action and further theoretical development. Capturing the dynamic interplay between measurable changes and reported first-hand experiences remains challenging due to the typical use of methodologies and measurement techniques lacking in temporal resolution. In this talk, I will discuss my current PhD study, which seeks to combine the use of portable Electroencephalography (EEG) and a novel phenomenological reporting system, namely, Temporal Experience Tracing (TET) developed at the Cambridge Consciousness and Cognition Lab, to capture the phenomenological granularity/intensity of ASCs elicited by Kundalini Yoga, and correlate them with observable fluctuations in neuroimaging data. The study’s objective is to shed light on the neurophenomenology of ASCs, to more accurately characterise the potential triggers of non-pharmacological altered states (such as proxy-hypoxic states), and evaluate Kundalini Yoga as a potential intervention for common mental health disorders such as depression and anxiety. I will discuss the results from my first pilot study involving 32 EEG and TET datasets across 8 participants and a further full study planned for April 2025. The results from these studies will inform the development of a neurophenomenology clinical trial, and inspire further studies to infer the neural correlates of ASC subtypes, such as blissful, equanimous and mystical states.

We present a comparison of complexity measures applied to EEG data recorded during pharmacologically induced altered states of consciousness (ASCs). Our primary aim is to assess the performance of complexity measures, including Lempel-Ziv complexity (LZc), Statistical Complexity and Complexity via State-space Entropy Rate. We evaluate their sensitivity, specificity and amount of data required for reliably differentiating ASCs from baseline states. Each of these measures has theoretical advantages and limitations but have not been compared empirically.

Since Tononi & Edelman’s early popularisation of complexity approaches, a major challenge remains in formalising empirical measures that can quantify global brain states, and how these may constrain theories of consciousness (Seth & Bayne, 2022; Kirkeby-Hinrup, 2024). Psychedelics provide a powerful experimental tool to investigate correlations between phenomenology and neural complexity. Findings converge on increased neural signal diversity in the psychedelic state, often interpreted as reflecting the richness of experience (Carhart-Harris, 2018). Lempel-Ziv complexity (LZc) has demonstrated remarkable success in differentiating conscious states, and although debates continue surrounding its suitability for capturing conscious content, remains the dominant measure in psychedelic research (Casali et al., 2013; Schartner et al., 2015; Lewis-Healey et al., 2024).

We describe our approach for evaluating complexity measures across datasets, and present preliminary findings from EEG data for psilocybin (macro)dosing and LSD microdosing. Our findings will contribute to the broader discussion on how different metrics might capture dimensions of experience and highlight key methodological considerations for selection and interpretation in future studies.

Issues of pathological canalization—including depression, anxiety, addiction, and PTSD—have recently seen treatment inroads. Among the most effective treatments are psychedelic and dissociative-assisted psychotherapy, as well as TMS/ultrasound neuromodulation. In these, it is proposed that the primary mechanism for acute changes that give way to enduring changes is “temperature or entropy Mediated plasticity” (TEMP) (Carhart-Harris et al., 2023). Given this, we have developed experiments that explore the possibility of inducing TEMP without the need for direct pharmacological/neural modulation. In these, we present intensive visual, audio, and tactile stimulus in a VR context, to the end of reaching hyperplastic states that can lead to lasting positive behavioral changes.

In a pilot study (MTurk, N=78), we examined a visual stimulus based on a psychedelic aid selected for its apparent dissociative qualities.

The following study (N=53) involved intensive visual, auditory, and tactile stimulus in a VR context, measuring wellbeing, connectedness, depression, anxiety, divergent association, altered states/acute dissociation, as well as meaning and meaning-making.

Our current study (ongoing, targeting N=40), introduces breathwork as a tool for increasing arousal and setting a baseline. We are also evaluating HRV and focusing our measures on evaluating whether the subjectively altered state we induce is a practical proxy for psychedelics and/or dissociatives.

We will present the results from these studies—in short, there has been some indication that these kinds of stimuli show promise as tools for accessing the benefits of TEMP, though individual differences play a substantial role in determining the effects of our stimuli.

This work presents the design and implementation of a novel bio-responsive virtual reality (VR) system that augments breathwork to induce altered states of consciousness (ASCs). Breathwork has demonstrated efficacy in eliciting ASCs comparable to psychedelics (Bahi et al., 2024; Lewis-Healey et al., 2024). Our system synchronizes user breathing patterns with dynamic spatial transformations in VR environments—expansion during inhalation and contraction during exhalation—leveraging embodied cognition to enhance immersive experiences. Building on findings that inhalation and exhalation influence perceptions of quantities and magnitude (Belli et al., 2021). We integrate these concepts to create immersive environments using Gestalt principles to dynamically adapt the sense of space to the user's breath. First pilot data exhibits the impact of this breathwork intervention on subjective states as assessed via psychometric time series (experience graphs) and standardized questionnaire for measuring ASCs. Further research explores reverse coupling mechanisms, where spatial contraction aligns with lung expansion, aiming to evoke body transformation illusions such as the sensation of growing while surrounding space shrinks, and investigates behavioral outcomes, including body size perception. Moreover, we apply neurophysiological measures to investigate the dynamic rescaling of peripersonal space through neural adaptation mechanisms (Noel et al., 2018) that are elicited via multisensory integration mechanisms (Serino et al., 2018). The VR system will be demonstrated after the presentation, allowing attendees to experience its functionality firsthand. Further information can be found at: https://www.aliusresearch.org/viscereality.html

As a serotonin 2A receptor (5-HT2AR) agonist, psilocybin elicits altered states of consciousness, consisting of changes in perception, distortions of time, and a range of “mystical experiences”. Randomized clinical trials (RCTs) of psilocybin therapy (PT) report rapid and robust antidepressant effects in patients with depressive disorders. All trials have administered psilocybin at doses of 20 to 30 mg in conjunction with several hours of psychological support. However, the mechanisms of PT’s antidepressant effects remain unknown. Although 5-HT2A agonism (and hence psychedelic effects) may be critical, antidepressant effects may also be mediated through rapid activation of other critical 5-HT receptors, downstream effects on synaptic growth, attentive psychological support, or a combination of all these factors. This session will describe the design of two ongoing RCTs at our centre in Toronto, Canada that aim to characterize PT’s mechanism of antidepressant action. The first is a “double dummy” RCT of psilocybin combined with risperidone in 60 adults with treatment resistant depression (TRD) to determine whether psilocybin’s antidepressant effects are dependent on 5-HT2AR agonism. The study is using risperidone, a potent blocker of the 5-HT2AR, because a previous study in healthy volunteers showed that risperidone 1 mg effectively blocks psilocybin’s psychedelic effects. The second RCT will determine whether psilocybin treatment with accompanying psychotherapy is superior to psilocybin treatment with safety monitoring for 40 adults with TRD. The current trials will address key unanswered questions in the field of psychedelic research and inform future clinical trial protocols and treatment guidelines with definitive global impact.

Music and ritual play vital roles across traditional and modern psychedelic experiences (1). However, neuroscientific research exploring this crucial interplay remains sparse. The Santo Daime Church uses ayahuasca together with music to have deeply meaningful experiences of the divine (2) and 24 members took part in a semi-naturalistic study, allowing for the examination of associated causal changes in the brain’s functional hierarchy.

Utilising whole-brain modelling with trophic coherence hierarchy metrics, we assessed how presence of Santo Daime ceremonial music, and manipulations to this music’s familiarity and predictability, moderate ayahuasca's effects on the hierarchy. Similar methods previously showed distinct hierarchical reconfigurations following LSD with and without music (3).

Findings indicated a significant flattening of the functional hierarchy following ayahuasca ingestion, during both resting states and music listening, in line with the predictions of the REBUS (Relaxed Beliefs Under Psychedelics) model (4). Notably, these effects appear to be modulated by the familiarity and predictability of the music.

This work underscores the integral role of music in modulating psychedelic experience, highlighting how variations in music can affect the functional hierarchy, and how this may allow the study of deeply meaningful experiences.

1. Jerotic K, Vuust P, Kringelbach ML. Ann N Y Acad Sci. 2024;1531(1).

2. Hartogsohn I. Front Pharmacol. 2021;12.

3. Shinozuka K, et al. Imaging Neurosci. 2025;3.

4. Carhart-Harris RL, Friston KJ. Pharmacol Rev. 2019;71(3).

Background: Recent interest in psychedelic compounds, particularly psilocybin, has expanded beyond clinical outcomes to explore their impact on conscious experience and cognitive processes. This study aimed to delineate how psilocybin influences the neural correlates of reward-based decision-making, focusing on the intersection of altered states of consciousness and reinforcement learning.

Methods: In a within-subject design, participants received either 1.5 g of psilocybin-containing mushrooms or an inactive placebo before completing a probabilistic reward learning task. Choices yielded positive or negative outcomes at varying probabilities, permitting assessment of feedback-driven adaptations in behavior. Event-related potentials (ERPs), including Contingent Negative Variation (CNV), Stimulus-Preceding Negativity (SPN), Reward Positivity (RewP), Late-Positive Potential (LPP), Feedback-Related Negativity (FRN), Error-Related Negativity (ERN), and Feedback-P3 (FB-P3), were recorded to capture neural indices of saliency, expectancy, and affective valuation during reward processing.

Results: Preliminary findings indicate differential ERP responses between the psilocybin and placebo conditions, reflecting changes in reward dynamics. Behavioral data also show an increase in erroneous responses under psilocybin, suggesting modifications in how feedback signals are integrated and acted upon. These outcomes hint at the possibility that altered conscious states may influence attention and affective components of decision-making.

Conclusions: This work suggests that psilocybin can reshape the processes involved in reward-based learning. By examining how altered states of consciousness intersect with core neural and behavioral markers, the study contributes to a deeper understanding of the mechanisms through which psychedelic-assisted interventions might benefit mental health conditions.

While there is consistency in the EEG power and complexity findings across psychedelic studies, this is not the case for the connectivity analyses. In this study, we investigate the effect of DMT (N, N, Dimethyltryptamine), a highly potent psychedelic compound, on Weighted Symbolic Mutual Information (wSMI) (King et al., 2013). wSMI is a functional connectivity metric that reflects the integration of computational processes in the brain and is the only connectivity metric that characterises nonlinear interaction dynamics (Imperatori et al 2020). We used wSMI to construct the connectivity network, and we employed graph network measures, such as the clustering coefficient (CC), a metric of local segregation, and the characteristic path length (CPL), a metric of global network efficiency.

Thirteen participants were exposed to separate placebo and DMT sessions during which their EEG activity and real-time subjective ratings of the intensity of experience were collected (Timmermann et al., 2019). The EEG data was utilised to compute the wSMI connectivity and the subsequent graph network metrics. We observed a significant increase in the CPL and a decrease in the CC during the DMT session compared to the placebo, notably pronounced in the alpha (8-12 Hz) and beta (12-30 Hz) EEG bands, along with a positive correlation with the ratings. These findings indicate a decrease in segregation and efficiency during DMT. This results in the brain functioning more irregularly and suggests a temporary disruption of the brain networks, facilitating flexible network reconfigurations necessary for improved performance in everyday tasks, thereby promoting plasticity.

The use of low doses of psychedelic substances (microdosing) is attracting increasing interest. Proponents of microdosing claim a wide range of clinically relevant cognitive enhancement effects. However, ongoing questions around efficacy, appropriate doses, and the role of expectation have led to confusion and controversy in microdosing science. In particular, the popularity of microdosing forces us to consider whether it may be possible for psychedelics to enhance cognition in the absence of marked alterations in consciousness.

To help unpack these issues I will present results of two recent reviews summarising all empirical microdosing research until 2024, including a set of infrequently cited studies that took place prior to prohibition.

Specifically, we reviewed 55 studies published since 1955, and summarised reported effects across six categories: neurobiological, physiological, phenomenological, affective, cognitive, and mental health.

Studies showed a wide range in risk of bias, depending on design, age, and other study characteristics. Laboratory studies found changes in pain perception, time perception, conscious state, mood, and neurophysiology. Self-report studies found changes in cognitive processing and mental health.

I will discuss methodological issues in microdosing research, and in particular will evaluate recent claims that microdosing may be a placebo effect. I aim to highlight differences between acute and enduring microdosing effects, and provide some specific design suggestions to facilitate more rigorous future research.

Psychedelic-assisted psychotherapy has shown promise in alleviating treatment-resistant depression by combining psychotherapy with psychedelic substances. While prior research has elucidated the neuropharmacological and subjective aspects of psychedelic therapy, the role of interpersonal dynamics remains understudied. Building on evidence from conventional psychotherapy that underscores the therapeutic alliance as one of the most robust predictors of clinical outcomes, my PhD research aims to fill this gap by systematically assessing the co-creation of non-ordinary states of consciousness that are central to the therapeutic process in psychedelic therapy for depression. We investigate interpersonal cardiovascular synchrony as an index of the intersubjective field in ketamine- and psilocybin-assisted therapy (during pre-dosing, dosing, and integration) conducted at McGill University’s Jewish General Hospital in Montreal. Data collection for the Ketamine therapy with 24 participants across 5 sessions is ongoing, the psilocybin trial with 15 participants who receive 2 doses of psilocybin has received Health Canada approval and is currently undergoing internal ethics review. Building on temporary social neuroscience findings, we examine how interpersonal physiological synchrony, reflected in mutual information exchange between electrocardiogram signals of patients and clinicians, converge with linguistic content, and self-reported measures of therapeutic alliance and psychedelic states. By integrating phenomenological insight, neurobiological measures, and established psychotherapeutic frameworks, my research advances our understanding of the relational nature of altered states of consciousness as they come to be in psychedelic therapy. Through assessing how co-created experiences in psychedelic therapy may moderate therapeutic outcomes, we ultimately seek to inform both clinical practice and theoretical models of consciousness.

Recent advances in studying psychedelics and related compounds have deepened our understanding of their neurophysiological effects and potential for exploring consciousness. While psychedelics like LSD, entactogens such as MDMA, and amphetamines like d-amphetamine share certain pharmacological properties, they also exhibit distinct neurophysiological effects. Moreover, individual psychedelics often possess unique neuropharmacological profiles. Understanding these commonalities and differences is vital for linking subtle neurochemical variations to profound changes in conscious experience.

This study used resting-state fMRI data from two independent cohorts to compare LSD’s effects with d-amphetamine and MDMA in one cohort, and with psilocybin and mescaline in another. Data were collected during peak effects under consistent protocols.

Connectivity metrics included network integrity, segregation, seed-based connectivity, and global connectivity. Results revealed shared and distinct effects: all substances decreased integrity in the visual network, but only psychedelics significantly reduced default-mode network (DMN) integrity. Psychedelics reduced network segregation, whereas amphetamines increased segregation between networks. Regarding seed-based connectivity, LSD exhibited stronger connectivity across seed networks compared to amphetamines. Furthermore, psychedelics also showed differential modulation within seed networks. Voxel-wise comparisons further highlighted nuanced differences both within (e.g., LSD vs. psilocybin) and between pharmacological classes (e.g., psychedelics vs. amphetamines).

The findings from these unique datasets emphasize the distinct neurophysiological profiles of psychedelics and related compounds, suggesting they modulate brain connectivity in ways that are both class-dependent and compound-specific. Furthermore, they highlight the potential of these substances to probe varying alterations in consciousness.

Classical psychedelic drugs such as lysergic acid diethylamide (LSD) induce an altered state of consciousness through partial agonism at the 5-Hydroxytryptamine-2A receptor (5-HT2AR). It is debated, though, what core cognitive computational mechanism mediates these effects. We conducted a double-blind, randomized, controlled trial testing the effects of low doses of LSD (0, 10, 20 µg; LSD base-equivalent doses) on visual perceptual decision making alongside electroencephalography (EEG) measurement and high-frequency eye-tracking. With this, we seek to unravel how LSD affects the relative weighting and the interaction of feedforward and feedback signals towards constructing perception.

To that end, we used a set of tasks probing distinct effects of context on perception which provide insight into a variety of EEG signatures of feedback and feedforward signaling. Specifically, we tested the effects of LSD in three two-alternative-forced choice tasks. The first task investigated evidence accumulation in an environment subject to hidden state changes. The second task tested the adaptation of visual serial dependence to block-wise manipulation of statistical regularities in the stimulus sequences. The third task tested visual contrast surround suppression, where observers underestimate contrasts embedded in a high-contrast surround.

The results will crucially advance our understanding of the basic mechanisms by which classical psychedelics alter perception. This has implications for the role of serotonin in perception in health and illness as well as for the use of psychedelics in psychiatry.

Recent research has focused on how psychedelics and empathogens may assist in the treatment of PTSD and depression following trauma. However, there is almost no knowledge regarding how psychoactive substances may impact the processing of Traumatic Events (TE) in real time. The large-scale terror attack by Hamas on October 7th 2023 on the 4000 attendees of the Supernova music festival has provided a tragic opportunity to study these two phenomena in conjunction. The attack took place shortly after sunrise, when many attendees were under the influence of mind-altering substances, mostly MDMA, LSD, ketamine and, as well as cannabis. This is an unprecedented mass trauma event that many of its victims experienced while in altered states of consciousness. We have begun a large-scale, longitudinal study with these survivors regarding the processing of trauma under the influence of psychedelics in the peritraumatic and post traumatic periods. Our results indicate that approximately 65% of the participants were exposed to severe trauma while under the influence of mind-altering substances. In this talk I will show preliminary results from this unique cohort including clinical (n = 1400), physiological (n = 300), cognitive (n = 900) and neural (fMRI) measures (n = 140) and how they relate to trauma and psychedelic use. The data and experiences of the courageous survivors of the festival provide novel insights into how trauma processing is impacted by psychoactive substances revealing unique interactions between cognitive, pharmacological and clinical factors.

The literature on psychedelic experiences, and especially mystical psychedelic experiences, often highlights their transformative potential. Laurie Paul (2014) introduces the concept of “transformative experiences,” which bring about extreme, irreversible changes in personality, values, and worldview. Paul uses the fictional example of someone being offered to permanently turn into a vampire: becoming a vampire is a decision that cannot be made based on someone's human value system, because their future vampire self will have an entirely different set of values that they might not even ben able to comprehend yet. Similarly, having children is a leap of faith: the transformation reshapes priorities and values irreversibly.

Psychedelic experiences are often described as transformative in this sense. Edward Jacobs (2023) argues that this raises ethical questions about informed consent in psychedelic-assisted therapy, since such experiences might profoundly alter a person’s value system, rendering pre-experience consent problematic. Many people report significant life changes following psychedelic experiences. However, I will argue that most psychedelic experiences are not transformative in Paul’s sense. Transformations like becoming a vampire or a parent alter not only someone's values, priorities and beliefs, but also their environments and relationships in a way that the subject has little control over. In contrast, psychedelic experiences are often contingent on individual agency. Without conscious effort to implement changes, life often reverts to its previous state. My talk will explore this distinction, examine cases of reversed transformations, and propose relational approaches to support lasting change.

Unlike qualitative features like 'redness,' duration is likely the only common attribute between qualia and their neural substrates. Thus, time-varying experiential dimensions and neural recordings can be seen as noisy measurements of a shared dynamical system, with each point on a latent manifold representing a joint neural-phenomenological state. We propose a framework for learning these latent state-spaces to identify which phenomenological dimensions best reconstruct multivariate neural signals, and vice-versa. This approach was applied to 12 temporal experience traces (TET) dimensions related to altered states of consciousness, collected after EEG scans during DMT administration.

Grounded in Takens' theorem, we derived a latent representation of shared dynamics between neural activity and subjective experience, using delay-embedded TET dimensions as inputs to a variational auto-encoder (VAE). From this latent-space we reconstructed EEG power and connectivity measures. TET reflecting participants’ memory of the auditory component of the DMT experience produced the best out-of-sample EEG reconstructions, with a mean distance correlation of 0.64 (R² = 0.44).

We then trained a VAE to learn a six-dimensional representation from a joint delay-embedding of 12 TET and 25 EEG dimensions. The model successfully reconstructed neural and behavioral inputs, achieving R² values of 0.516 and 0.966, respectively. We present brain-states reconstructed from this model, reflecting idealized, synthetic, altered states of consciousness: specifically, temporal distortion and entity encounters.

Our findings serve as a proof of concept for the viability of jointly modeling neural and subjective time series as a single joint dynamical system.

N,N-Dimethyltryptamine (DMT), a classical serotonergic psychedelic, is known for its vivid visual and somatic effects, making it a valuable tool for studying the neuroscience of consciousness (Strassman et al., Arch Gen Psychiatry. 1994). We investigated the neurobiological mechanisms of DMT-induced altered state of consciousness by examining how the spatiotemporal dynamics of large-scale brain activity patterns evolve over time.

We analyzed functional magnetic resonance imaging (fMRI) data from seventeen healthy participants during acute DMT and placebo administration in a crossover study (Timmermann et al., PNAS. 2023). Using a turbulence framework inspired by fluid dynamics (Deco, Kringelbach, Cell Reports. 2020), we measured local synchronization of coupled brain signals. This framework describes how synchronized brain activity forms a turbulent core, facilitating information processing through rotational vortices, akin to those in fluids. Additionally, we characterized the dynamics of vortex-like rotational wave patterns, known as brain spirals. Synchronization and vorticity were examined both, in a time-averaged and a time-resolved manner, providing detailed, moment-by-moment profiles of their evolution.

Our results demonstrate that DMT reduces global brain synchronization and its variability, while simultaneously increasing global vorticity and its variance. Notably, these measures display distinct, time-resolved fluctuations that evolve dynamically and diverge significantly from those observed under placebo condition.

These findings suggest that DMT induces a state of desynchronization and heightened dynamic complexity in brain activity. Investigating these time-resolved features provides novel insights into how DMT alters brain dynamics and could serve as a tool for predicting subjective experience, advancing our understanding of consciousness under psychedelics.

Information theory provides a versatile and robust framework for investigating the intricate behaviours of complex systems, such as the human brain. Recent advancements in the field have increasingly been applied to computational neuroscience, offering researchers powerful tools to analyse how neural information flows and integrates across different regions of the brain. These methods have proven particularly promising in the study of consciousness, enabling quantitative characterization of various states of consciousness.

In this study, we leverage the framework of Integrated Information Decomposition (PhiID) to examine the effects of psychedelic drugs on the brain's informational organization. PhiID allows us to break down the system's information dynamics into fine-grained components, shedding light on how information is transferred, stored, and processed in the human brain under placebo and psychedelic conditions. By combining this approach with Vector Autoregression (VAR) models, we analyse MEG data from participants under placebo and psychedelic substances (ketamine, LSD, psilocybin). This enables us to assess differences in cortical dynamics between the two states.

Our findings reveal that, overall, there is a higher magnitude of information in placebo states, aligning with prior claims that psychedelic conditions increase neural complexity. Furthermore, psychedelic states exhibit an increase in persistent redundancy and a decrease in persistent synergy, indicating a substantial reorganisation of the brain's modes of segregation and integration.

These results offer valuable insights into the neural correlates of conscious experience, providing a deeper understanding of how information is synthesized and organized across different states of consciousness.

Introduction:

Psychedelics have gained increasing interest in scientific research due to their ability to induce profound alterations in state of consciousness. However, the research regarding the functioning of individuals who use psychedelics in naturalistic contexts remains limited. To fill this gap we explored neuropsychophysiological differences between naturalistic psychedelics users and non-users in terms of processing of self-related thoughts.

Methods:

We used behavioural testing combined with electroencephalography with source localisation. To mitigate potential confounding effects of personality traits and personal history which makes one willing to take psychedelics, we compared users to individuals who did not take psychedelics, but are intending to do so in the future. We included two datasets collected at two different laboratories.

Results:

The results from Dataset I (N = 70) suggest that during self-related thoughts psychedelics users exhibit weaker increases in alpha and beta power in comparison to non-users, primarily in brain regions linked to processing of self-related information (such as posterior cingulate cortex). Analysis of Dataset II (N = 38) did not replicate between-group effects. Analysis of combined datasets showed similar results as findings from Dataset I alone.

Conclusions:

While non-replicability restricts interpretation of our findings, our research expands the ongoing discussion on strength and duration of the psychedelic effects. Our results fit into growing skepticism about the specificity of the role of default-mode network in processes associated with psychedelics experience. By focusing on the intersection of psychedelics, brain function, and self-consciousness, our study contributes to understanding altered states of consciousness and perfectly aligns with ASSC scope.

Introduction:

Psychedelics profoundly alter consciousness, influencing perception, cognition, and mood (Wallach, 2009). Prolonged use has been linked to improved psychological well-being (Nichols, 2005; Nagai et al., 2007). 5-methoxy-N,N-dimethyltryptamine (5-MeO-DMT), a serotonin receptor agonist (Spencer et al., 1987), occurs naturally in mammals (Abiero et al., 2019), plants, and animals, most notably in the secretion of the Incilius alvarius toad (Jiang & Shen, 2015). Its ability to induce mystical experiences (Walsh, 1982) and ego dissolution (MacLean et al., 2011; Erritzoe et al., 2018) highlights its therapeutic potential.

Aims & Methods:

While current studies have mainly focused on brain dynamics or therapeutic outcomes in naturalistic settings (Uthaug et al., 2019; Blackmore et al., 2024), this study comprehensively investigates the subjective effects of 5-MeO-DMT-induced altered states of consciousness in a controlled clinical environment. By utilising sophisticated methods such as experience sampling and Microphenomenology, the study goes beyond standard Likert scale measures, which fail to capture the richness of subjective experiences.

This placebo-controlled, single-blind study assessed 32 participants over two dosing sessions separated by two weeks. Experience sampling and microphenomenological interviews were employed to explore subjective effects.

Results:

Preliminary findings reveal significant alterations in subjective states, including ego dissolution (or the sense of “self”) aligning with established psychedelic models. Analysis is currently underway.

Conclusion:

This study advances understanding of the subjective effects of 5-MeO-DMT, bridging a critical research gap and informing future investigations on its therapeutic potential. Findings contribute to expanding consciousness research and the development of psychedelic therapies.

The growing acceptance of psychedelics for therapeutic and research purposes underscores the need to optimize safety and efficacy of use. While psychedelic experiences can lead to profound therapeutic benefits, outcomes remain highly variable. The successful prediction of experience for a given individual will contribute not only to increased therapeutic efficacy but also answer more fundamental questions around the influence of “pre-state” on alterations in consciousness. This talk will explore research on “pre-state” as a predictor of psychedelic experience, drawing from our work on the Imperial Psychedelic Predictor Scale (Angyus et al., 2024) and a subsequent literature review (Murray et al., in prep) on the topic.

Our recent work has validated a short assessment of “pre-state” (set, rapport, and intention) which can significantly predict mystical, emotional breakthrough, and challenging experiences (Angyus et al, 2024). This work is an early step in a broader effort to map psychological “pre-state” qualities to the subsequent nature of a psychedelic experience and therapeutic outcomes.

One such effort is our application of this scale in a citizen science study design. We have developed a protocol with surveys and EEG recordings before, during, and after altered state experiences. By partnering with wearable EEG developer Interaxon, we have integrated this protocol directly in the Muse mobile application. This scalable approach hopes to generate the largest ever collection of psychedelic brain data. This effort will support the mission of mapping pre-state measures to acute experience and therapeutic outcomes, supporting the understanding of variables that impact changes in consciousness.

This presentation explores how psychedelic experiences function as Ontologically Diversifying Experiences (ODEs) by disrupting normative constructs of selfhood, relationships, and reality, thereby expanding cognitive, emotional, and existential repertoires. Drawing on three studies—Navigating Groundlessness (Argyri et al., 2024), which details how participants navigate destabilizing ontological shock; Learning from Boundlessness (Argyri et al., under review), which shows expanded concepts of normality and increased prosocial tendencies; and research on post-psychedelic changes in gender and sexual identity (Kruger et al., under review), which reveals shifts toward queerness and fluid understandings of identity—I demonstrate how psychedelics challenge entrenched assumptions and open avenues for psychological growth.

I further examine participant reports and quantitative data showing that these experiences can transform trauma into opportunities for growth, with social support and adaptive coping strategies mitigating the impact of ontological shock. By aligning these processes with the DE framework from socio-cognitive research (Gocłowska et al., 2018), I argue that psychedelics offer benefits beyond traditional mental health treatments by stimulating creative thinking, enhancing mental adaptability, thus catalyzing transformative insights into our self-identity, relationships, and understanding of reality.

The presentation concludes with practical implications for harm reduction and self-development, raising broader questions about psychedelics as catalysts for transforming human consciousness.

Lysergic acid diethylamide (LSD) is a serotonergic psychedelic that induces a profoundly altered conscious state. The dorsolateral prefrontal cortex (DLPFC) is a brain region that is known to be involved in mood regulation and disorders; hypofunction in the left DLPFC is associated with depression. This study investigated the role of the DLPFC in LSD’s psycho-emotional effects with functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG) data of healthy participants during the acute LSD experience.

In the fMRI data, we measured the correlation between changes in resting-state functional connectivity (RSFC) of the DLPFC and post-scan subjective ratings of positive mood, emotional arousal, and ego dissolution. Ego dissolution was significantly correlated with functional connectivity increases between the combined left & right DLPFC, thalamus, and the fusiform face area. Additionally, emotional arousal was significantly associated with increased connectivity between the right DLPFC, intraparietal sulcus, and the salience network. A confirmational “reverse” analysis, using the original RSFC clusters as input seeds substantiated the role of the right DLPFC and the aforementioned regions in both measures.

Utilizing a complimentary MEG dataset, we source-localized to the output clusters. The Granger causality analysis revealed that LSD increased information flow between two nodes of the ‘ego dissolution network’, the thalamus and the DLPFC, in the theta band, substantiating the hypothesis that disruptions in thalamic gating underlie the experience of ego dissolution.

Overall, this multimodal study elucidates a role for the DLPFC in LSD-induced states of consciousness and sheds more light on the brain basis of ego dissolution.

N,N-Dimethyltryptamine (DMT) is a potent and fast-acting psychedelic drug that induces a radical reorganisation of the contents of consciousness, comprising the dissolution of time and space and perceptual immersion into an ‘alternate reality’. While contemporary research has somewhat advanced our understanding of DMT, and psychedelics more broadly, there is little research that integrates time-resolved measures of subjective experience with temporally fine-grained brain imaging. We therefore present the current study, a repeated-measures dose-dependent study of the subjective and neural dynamics induced through DMT under naturalistic conditions. Nineteen participants received either a 20mg or 40mg dose of freebase DMT across two dosing sessions in a blinded, counterbalanced order, with blinding rates consistent across doses. Electroencephalography (EEG) data was collected, as well as time-resolved retrospective measures of subjective experience (Temporal Experience Tracing). Both doses of DMT induced rapid changes in experience dimensions. However, the 40mg dose induced significantly more extreme visual hallucinations and emotionally intense experiences. Further, we computed a variety of neural markers on the EEG data, and found that oscillatory alpha power and permutation entropy were most strongly associated with continuous subjective experience dimensions. Strikingly, lempel-ziv complexity, a previously hailed as a robust correlate of subjective experiences within the psychedelic-state, was the least

strongly associated neural marker. These findings provide an important insight into how distinct neural dynamics may contribute to this radical and intense altered state of consciousness.

5-methoxy-N,N-dimethyltryptamine (5-MeO-DMT) is a psychedelic drug known for its uniquely profound effects on subjective experience, reliably eradicating the perception of time, space, and the self. However, little is known about how this drug alters large-scale brain activity. We collected naturalistic electroencephalography (EEG) data of 29 healthy individuals before and after inhaling a high dose (12mg) of vaporised synthetic 5-MeO-DMT. We replicate work from rodents showing amplified low-frequency oscillations, but extend these findings with novel tools for characterising the organisation and dynamics of complex low-frequency spatiotemporal fields of neural activity. We find that 5-MeO-DMT radically reorganises low-frequency flows of neural activity, causing them to become incoherent, heterogeneous, viscous, fleeting, nonrecurring, and to cease their typical travelling forwards and backwards across the cortex compared to resting state. Further, we find a consequence of this reorganisation in broadband activity, which exhibits slower, more stable, low-dimensional behaviour, with increased energy barriers to rapid global shifts. These findings provide the first detailed empirical account of how 5-MeO-DMT sculpts human brain dynamics, revealing a novel set of cortical slow wave behaviours, with significant implications for extant neuroscientific models of serotonergic psychedelics.

Understanding the neural substrate of altered conscious states is an important cultural, scientific, and clinical endeavour. Although hypnosis causes strong shifts in conscious perception and cognition, it remains largely unclear how hypnosis affects information processing in cortical networks. Here we manipulated the depth of hypnotic states to study information processing between cortical regions involved in attention and awareness. We used high-density Electroencephalography (EEG) to record resting-state cortical activity from 30 hypnosis experts during two hypnotic states with different depth. Each participant entered a light and a deep hypnotic state as well as two well-matched control states. Bridging top-down and lateralisation models of hypnosis, we found that interhemispheric frontoparietal connectivity distinguished hypnosis and control conditions, while no difference was found between the two hypnotic states. Using a graph-theoretic measure, we revealed that the amount of information passing through individual nodes (measured via betweenness centrality) is reduced during hypnosis relative to control states. Finally, we found that theta power was enhanced during hypnosis. Our result contributes to the current discussion around a role for theta power in bringing about hypnotic states, as well as other altered conscious states. Overall, our findings support the notion that altered top-down control in frontoparietal regions facilitates hypnosis by integrating information between cortical hemispheres.

Our world is full of beings and the space of possible minds is vast. An individual mind has experiences. The structure of such experiences gives rise to particular states. Individuals find themselves at a particular location on the landscape of such states. Regions that are often traversed in normal life are only a spectrum of the possible conscious states that exist.

This presentation explores how we might systematically map the inner world to better navigate and expand our range of possible experiences. Different avenues of approach—meditation, psychopharmacology, and brain-machine-interfaces—can be used to explore the landscape. Each method produces examples of novel states of consciousness, helping us to build a map while simultaneously being tools to navigate it.

We can see conscious experiences as belonging to a universal set, where individuals and species each inhabit distinct subsets. A set-theoretic approach provides a formal model, where experiences are elements within larger sets of possible and lived experiences. It provides a useful framework for analysing the diversity in experiences, states, and minds that exist.

By utilising mathematical concepts, we refine our understanding of how consciousness can be structured, compared, and intentionally modified. The development of a Lexicon of Being, cataloging known and yet-to-be-discovered states, provides a means for democratising access to the varieties of consciousness.

Building maps of the inner landscape allows us to move toward a future where individuals can explore, understand, and engineer their own mental states with greater precision and intention.

The democratization of deep meditative states through neurofeedback technology represents a promising frontier in contemplative science, yet its development has been hindered by the absence of personalized, objective markers tracking real-time meditation depth. Here, we present heartbeat-evoked potentials (HEPs) as a dynamic index of meditative states, examining how neural responses to cardiac events reflect moment-to-moment shifts in phenomenological experience. We focused on Vipassana meditation—a practice defined by systematic cultivation of non-judgmental bodily awareness—as it offers a direct window into enhancing interoceptive sensibility, a faculty notably disrupted across mental health conditions that show robust improvements with meditation. Expert practitioners (n=30) underwent two 90-minute sessions separated by a minimum of one week, during which they employed a novel "spontaneous emergence" method to collect real-time measures of meditative depth while recording 64-channel EEG and ECG. Capitalizing on meditation's inherent silence, we utilized the R-peak of the ECG signal as a temporal marker to examine neural responses to these endogenous events across varying meditative depths. Our analyses revealed that HEP amplitude at electrode C3 exhibited remarkable sensitivity in distinguishing five levels of self-reported meditation depth (p<0.0001, η²=5.55), with deeper states permitting cardiac events to hold increased causal power over neural activity. The clinical relevance of these markers was further supported by HEP modulation range predicting post-session improvements in mood and decentering (both p<0.0001). This work introduces a framework for developing personalized neurofeedback protocols to facilitate meditative states, with implications for both basic research in contemplative neuroscience and clinical applications of mindfulness-based interventions.

The Transcendental Meditation (TM) technique is a simple, effortless, standardized procedure practiced for 20 min twice daily while sitting comfortably with eyes closed. TM practice enables the mind to settle down effortlessly and spontaneously to experience more refined, quiet states of the thinking process until one transcends thinking to experience the source of thought (i.e. pure awareness). Electroencephalography (EEG) studies have shown that as practitioners report quieter, more peaceful experiences, brain functional connectivity exhibits higher alpha coherence in frontal and posterior electrodes. It is an open question to what extent functional connectivity metrics that capture nonlinear interactions are modulated during TM. Here, using EEG recordings and Support Vector Machine (SVM) classification from 38 long-term meditators (mean age = 40.3 yo; 23 males; mean years of practice = 12.9) and 38 controls (non-meditators) matched by age and gender, we evaluated connectivity metrics tracking distinct large-scale neural interactions: weighted symbolic mutual information (WSMI), capturing nonlinear interactions; and weighted phase-lag index (WPLI), capturing linear interactions. WSMI and WPLI were computed during baseline periods (10 mins before and after TM) and TM period (30 mins). Interestingly, our preliminary results demonstrate that TM and baseline periods in practitioners are better decoded by WSMI than WPLI across several canonical frequency bands, showing a sharp decrease in WSMI during TM compared to baseline. Our results reveal distinct neural dynamics during TM, offering insights into refined conscious states and highlighting the value of nonlinear approaches to studying meditation.

Recent findings suggest that Emergent Phenomena (EP)—unusual mental or somatic states often labeled as spiritual, mystical, energetic, psychedelic, or magical (Sandilands & Ingram, 2024)—are far more prevalent than previously assumed. Wright et al. (2024) reported that 45% of a general population sample experienced at least one non-pharmacologically induced EP. While some individuals report transformative growth from these profound experiences, others face significant distress. As consciousness altering practices and “non-ordinary” states (e.g., meditation, psychedelics) become increasingly common in Western contexts, there is a pressing need to refine diagnostic and treatment approaches that account for both potential risks and benefits. Limited clinical recognition of EP contributes to their being overlooked or mistakenly pathologized, which may result in suboptimal treatment.

To address this gap, Emergent Phenomena Research Consortium launched the Expert Opinion Project. In the first phase, a three-round Delphi study, we are engaging researchers and clinicians with expertise in both clinical/medical and contemplative domains to identify key assessment criteria, treatment strategies, and systemic barriers to effective care of EP. This work aims to develop consensus guidelines for recognizing and supporting EP in clinical contexts. Future phases will expand participation to include cultural leaders, integration specialists, and individuals with lived experience, ensuring guidelines balance clinical rigor with cultural sensitivity and practical utility.

We propose that physiological childbirth, particularly in unmedicated contexts, represents a spontaneous altered state of consciousness. Our theoretical framework asserts that the hormonal cascades and intense somatic engagement during natural labour produce transient modifications in perception, affect, and self-awareness that overlap with those experience during recognised altered states. This model builds on previous case observations that document measurable shifts in experiential dimensions during childbirth.

We conducted a large-scale survey of over 10,000 participants drawn from the Great British Intelligence Test cohort to evaluate this framework empirically. The survey on-goingly collects retrospective self-reports of childbirth experiences using validated measures to assess temporal perception, sensory modulation, and affective response. We then apply factor analysis and multivariate regression to systematically map the phenomenological dimensions and identify latent constructs associated with altered states of consciousness, then relate them to childbirth experiences.

This investigation refines conceptual models of spontaneous altered states and holds implications for both theoretical understanding and clinical practice in obstetric care, but also for the science of consciousness. Detailed results and further analyses will be presented to stimulate discussion on the integration of neurobiological and phenomenological perspectives related to childbirth in the study of consciousness.

We are entering a third wave of meditation research focused on advanced meditation, including states and stages of meditation practice that unfold with time and mastery. Nirodha samapatti (NS), sometimes called extended cessation in the scientific literature, is a meditative endpoint described in Theravada Buddhist meditation that corresponds to complete cessation of perception and feeling. In other words, meditators report periods of unconsciousness during NS. Here, we aimed to determine the neural correlates of NS, including their relation to biomarkers of unconsciousness. For example, EEG-based Lempel-Ziv complexity (LZc) has been shown to consistently decrease during anesthesia and disorders of consciousness. Using the first-ever EEG and MEG data from NS meditators, we measured differences in whole-brain, source localized, normalized, broadband LZc between NS and two non-meditative, cognitively engaging control tasks. No significant differences in normalized, broadband LZc were identified between NS and the control tasks. These results suggest that NS is a unique brain state from other interventions and conditions that lead to unconsciousness. In the future, we intend to analyze frequency-resolved LZc, as well as other neural correlates of consciousness, including hierarchical brain function and information integration.

Introduction:

Virtual Reality Hypnosis (VRH), combining virtual reality (VR) and hypnosis, is a promising tool for alleviating pain and anxiety by fostering dissociation and absorption. Proponents of VRH hypothesize that VR enhances the hypnotic experience for people with a low hypnotic suggestibility (HS). While HS influences hypnotic effectiveness, it remains unclear whether VRH benefits low HS individuals. This study investigates whether VRH enhances hypnosis for individuals with low HS using the Elkins Hypnotizability Scale1.

Methods:

Sixty-six participants (41±14 years; 38 women) underwent two randomized conditions: VRH (Aqua©, Sedakit HypnoVR, underwater whale-following simulation) and traditional hypnosis (prerecorded audio with black cross fixation in the VR headset). HS was assessed beforehand, and participants were categorized as low, medium or high (scores: 0–3, 4–7, 8–12 respectively). After each condition, dissociation, absorption, and anxiety were rated using numerical scales (0–10), and time perception was evaluated with open-ended responses. A repeated measures ANOVA examined effects of group (low, medium, high), condition (VRH, traditional hypnosis), and their interaction.

Results:

Among participants, 29 had low, 21 medium, and 16 high HS. No significant differences were found across groups, conditions, or their interaction for any measured variable.

Conclusion:

Findings do not support the hypothesis that VR enhances hypnosis for low HS individuals. Nor it does for high and medium HS participants. These results are critical given the growing use of VRH in clinical settings. Further research is necessary to refine this innovative tool and better understand its clinical potential.

1. Kekecs Z., et al. IntJClinExpHypn. 2016.

Embodied self-specific processing is linked to both mortality awareness and Buddhist-inspired meditation practices. Human predictive capacity underlies its adaptive strength but also the potential for existential terror. Grounded in the predictive processing framework of brain function, we recently showed using a magnetoencephalogram visual mismatch-response (vMMR) paradigm that prediction-based self-specific neural mechanisms shield the self from existential threat–at the level of perception–by attributing death to the ‘other’ (non-self). Given the mechanistic role of self-other distinction in death-denial, and the Buddhist soteriological aim of its dissolution, we tested whether meditation practice was associated with a reduction in the brain’s defensiveness towards mortality. Thirty-eight meditators pooled from a previous project investigating self-boundary dissolution neurophenomenology underwent the vMMR task. The pre-registered results indicated vMMR responses to the coupling of death and self-stimuli in a manner indicating acceptance rather than denial, corresponding to increased self-reported well-being. Furthermore, death acceptance degree was associated with the phenomenological affective valence dimension (blissful to terrible), but not dissolution depth dimensions (degree of agency/1st-person perspective/attentional disposition), or previous meditative experience. These findings support the hypothesis that deeply seated defenses underlying the human tendency to avoid death are amenable to mental training. Furthermore, the results underscore, in line with other studies, the importance of affective valence during meditation, in particularly in non-dual awareness practices (as well as psychedelic interventions) which can induce radical disruptions to self-consciousness.

This review integrates anthropological and neuroscientific perspectives to elucidate music-induced trance processes and their implications for non-ordinary states of consciousness (NSCs), with a focus on unifying psychobiological frameworks. The objectives are to deepen the understanding of trance phenomena, delineate links between various trance expressions, and identify gaps to guide future neuroscientific research. Drawing on anthropological insights, we examine trance processes focusing on their phenomenology, benefits, and the central role of music in their expression. Examples such as shamanism in non-Western cultures and rave parties in Western societies reveal shared cultural narratives and universal features and functions of music—such as rhythmic structuring and synchronization during rituals and entertainment events—that facilitate trance states by guiding their induction and maintenance. On the neuroscientific front, we review studies employing imaging techniques to map brain activity and connectivity during music-induced trance states. Findings reveal shared neurophysiological mechanisms across NSCs, including shifts in neural activity from executive networks to limbic and default mode networks, as well as heightened synchronization in low-frequency brainwaves—particularly delta and theta bands. These neural signatures underline the dynamic interplay between cognitive, emotional, and sensory systems during trance, providing insights into its universal neural basis. Our interdisciplinary synthesis highlights the universality of trance phenomena and underscores their cultural and biological roots. Future research should examine individual suggestibility and the therapeutic potential of music-induced trance in modern clinical applications.

While the beneficial impacts of meditation are increasingly acknowledged, its underlying neural mechanisms remain poorly understood. Here, we employed magnetoencephalography (MEG) to assess changes in brain dynamics in expert Buddhist monks during resting state (RS), and two established meditation methods known as Samatha and Vipassana, which employ focused attention and open monitoring techniques. We leveraged the high spatio-temporal resolution of MEG to compute a wide range of features with a focus on neural complexity and criticality metrics. More specifically, the study assessed spectral power density (PSD) in multiple frequency bands, the 1/f exponent, the long-range temporal correlations (LRTC) on oscillatory envelopes via Detrended Fluctuation Analysis (DFA), Lempel-Ziv complexity (LZc), Spectral Entropy (SpecEn), Higuchi's fractal dimension (HFD) and the distance to criticality coefficient (DCC).

Our findings indicate increased levels of neural signal complexity during both meditation practices compared to the resting state, alongside widespread reductions in gamma-band LRTC and 1/f slope. In addition, a feature importance analysis based on a Random Forest classifier, revealed that the features with the highest discriminative power between the meditative states and resting-state were the DFA exponent. Importantly, the DCC analysis revealed a separation between Samatha and Vipassana, suggesting that their distinct phenomenological properties are mediated by specific computational characteristics of their dynamic states. Overall, these results suggest that different meditation techniques engage unique neural circuits and processes, enhancing our understanding of how meditation affects brain function through the lens of complexity and critical dynamics.

In the wake of the active inference framework, two popular theories of consciousness highlight the relevance of insular cortex for interoceptive self-modeling (Fermin et al., 2022; Seth, 2021) and subcortical brain regions for qualitative experience (Solms, 2021). Both provide a compelling ecological argument for integrated conscious experience, i.e., self-organization of complex organisms with optimization goals that are usually parallel, multifaceted, and hard to reconcile. Computationally, both theories require some kind of self-model as basis, which could contradict credible reports of minimal phenomenal experience (MPE) (Metzinger, 2024) that argues for conscious states without an experience of selfhood. However, duality of conscious experience could be explained by a neuroscientific theory of two distinct brain networks emerging from different neurodevelopmental pathways, leading to marked differences in cytoarchitecture and function (Luu et al., 2024; Sanides, 1962). System-A, originating from an olfactory system and amygdala-centered expansion gradient towards ventral cortex, could enable interoceptive self-modeling for habitual interactions with the body and the world. System-H, hippocampus-centered towards dorsal cortex, could enable less egocentric forms of cognition and experience. In this sense, MPE could be a less salient/habitual form of experience, when neither interoceptive/exteroceptive prediction errors nor spontaneous episodic memory reactivation provide self-referential engagement of System-A. System-H, if not operating on content, could give rise to self-less experience of the world as such.

Computational theories of affective valence employing the active inference framework link valence to uncertainty reduction and subjective fitness. However, such a view cannot easily account for the conscious phenomenology of deep meditative experience, where valence is modulated by the degree of contraction, tensing or constriction present in subjective experience. I examine tensions between these two perspectives, aiming for an integrated active inference account of valence.

Drawing on work on meditative deconstruction, cognitive effort and traditional Buddhist perspectives, I argue that while valence is superficially tied to allostasis (the process of achieving stability through change), this relationship is mutable via the release of contraction during deconstruction. Synthesizing meditative and uncertainty-reduction perspectives, I show how the release of contraction results in more positive valence, irrespective of subjective fitness, and correlates with decreased hierarchical depth and complexity of an agent’s generative model of the world.

This more comprehensive computational understanding of valence has important implications for the design of interventions targeting the reduction of suffering and improved well-being.

Mindfulness practice has been associated with numerous cognitive benefits, yet its specific impact on automatic mental processes remains understudied. This research investigates how focused attention (FA) mindfulness influences automatic semantic processing through a series of modified Stroop tasks. In a between-groups design, participants were randomly assigned to either brief FA meditation or mind-wandering control conditions. Results demonstrated that FA meditation significantly reduced Stroop congruency facilitation effects specifically in the classic color-word task, but not in control tasks involving non-semantic visual features. This selective reduction suggests that FA meditation temporarily inhibits automatic semantic processing rather than enhancing general attentional control or proactive maintenance. Notably, these effects emerged in meditation novices following brief (10-minute) guided FA sessions with 2-minute maintenance periods, indicating that even short-term mindfulness practice can modulate automatic cognitive processes. These findings provide empirical support for mindfulness' role in reducing automatic thought processes and contribute to our understanding of its underlying cognitive mechanisms. The study offers insights into how brief mindfulness interventions might help individuals better manage intrusive thoughts and automatic cognitive processes in daily life.

Background

The modulation of the sense of self is a central part in many contemplative traditions. It has been described phenomenologically as involving a profound alteration of the distinctions of self and world, which has been termed as ‘self-boundary dissolution’. We used neurofeedback to explore (A) the neurophenomenology of self-boundary dissolution by creating a closed loop between brain activity and phenomenology, and (B) how neurofeedback can support this state.

Methods

We provided experienced meditators (n=18) with real-time 64-channel EEG auditory neurofeedback of source localized brain activity. Each subject completed three feedback conditions: Lempel-Ziv complexity in the posterior medial cortex, theta oscillatory power in the occipital cortex, and a sham feedback. In the first part, participants were blinded to which signal direction corresponded to a deeper self-boundary dissolution, and the direction was randomly flipped across 7 repeated trials. Participants were asked to identify the direction by comparing their meditation experience to the feedback signal. We predict identification of the correct direction with above-chance accuracy. In the second part, participants were asked to regulate the feedback in the direction which corresponded to self-boundary dissolution. We predict that the feedback in the experimental conditions would support the dissolution more than in the sham condition. Phenomenological interviews were conducted after each part to explore participants’ experience and strategies.

Results

Data collection is ongoing and will be completed in April. In a pilot study, participants recognized the signal direction with significantly above-chance accuracy and were able to regulate the signal in both experimental conditions.

Hypnosis is a state of “focused attention and reduced peripheral awareness characterized by an enhanced capacity for response to suggestion” (Elkins et al., 2015). Hypnotic verbal suggestions effectively alter phenomenological experiences, including pain. However, how the brain encodes hypnotic verbal suggestions to change pain perception remains unclear.

We recorded fMRI data from 23 participants exposed to four types of hypnotic verbal suggestions: hyperalgesia, analgesia, and two matched neutral conditions. After each suggestion (~2 min.), cutaneous electric stimulations were administered, and subjective pain ratings were collected. Leveraging the time-locked nature of this paradigm, we applied intersubject correlation (ISC) analysis to assess shared neural fluctuations during suggestions. Hypnosis-relevant traits (e.g., hypnotic susceptibility) and pain modulation were examined as between-person factors.

Pain ratings confirmed the expected modulatory effects of suggestions (Hyper > Neutral: t(22) = 3.43, p = .003; Analgesia > Neutral: t(22) = -2.48, p = .02). Significant ISC was observed for pain modulation conditions (combined hyperalgesia and analgesia) in regions associated with language comprehension and multimodal processing (peak r = 0.14, FDR < .05). The analgesia condition induced greater ISC than hyperalgesia in the superior and middle temporal gyri (peak r difference = 0.13, FDR < .05), with synchrony increasing alongside hypnotic susceptibility.

This study is the first to investigate hypnotic suggestion encoding for pain modulation using a model-free approach. Our findings suggest that hypnotic analgesia engages more consistent neural dynamics across individuals than hyperalgesia, highlighting how linguistic content structures conscious pain experiences and offering insights into the neural encoding of suggestion-driven modulation.

We present a neurophenomenological case study investigating distinct neural connectivity regimes during an advanced concentrative absorption meditation called jhana (ACAM-J), characterized by highly-stable attention and mental absorption. Using EEG recordings and phenomenological ratings (29 sessions) from a meditator with +20,000 hours of practice, we evaluated connectivity metrics tracking distinct large-scale neural interactions: nonlinear (WSMI and Directed Information), capturing non-oscillatory dynamics; and linear (WPLI) connectivity metrics, capturing oscillatory synchrony. Results demonstrate ACAM-J is better distinguished by non-oscillatory compared to oscillatory dynamics across multiple frequency ranges. Furthermore, combining attention-related phenomenological ratings with WSMI improves Bayesian decoding of ACAM-J compared to neural metrics alone. Crucially, deeper ACAM-J indicate an equalization of feedback and feedforward processes, suggesting a balance of internally and externally-driven information processing. Our results reveal distinct neural dynamics during ACAM-J, offering insights into refined conscious states and highlighting the value of nonlinear neurophenomenological approaches to studying attentional states.

Meditation has been shown to induce states of consciousness that are free from various forms of suffering, such as physical pain and mental anxiety. However, its broad effectiveness raises the question: what kind of suffering or well-being does meditation actually address? We explore this question systematically, proposing that meditation's effectiveness stems not from addressing specific forms of suffering but from dismantling the underlying mechanisms that create the conditions for suffering to arise in the first place.

In the talk, first, I will frame the notion of suffering from three frameworks: well-being (hedonia/eudaimonia), contemporary mindfulness, and traditional Buddhist practice. While these approaches may lead to phenomenologically similar states free from suffering, they reinforce different mental tendencies, thereby affecting how conscious experience unfolds over time. Second, I will highlight this difference using the action-perception cycle as the analytical tool. Particularly, I will contrast the “presence” of suffering and its “perpetuation” (probability of suffering arising). Third, I will propose four stages through which meditation reduces perpetuation by progressively revealing that suffering, typically perceived as a passive experience, is in fact a specific pattern of action we are engaged in (i.e., enacted).

Our framework provides precise conceptual tools to computationally model how meditation alters the action-perception cycle, making it possible to formalize traditionally subjective concepts like non-reactivity, ignorance, and suffering. This has significant implications for both clinical applications and consciousness research.

Historically, clinicians and psychologists of religion have sought to distinguish spiritual from psychopathological experiences under the heading of ‘differential diagnosis.’ While distress and impairment are often proposed as criteria for pathological interpretations, recently researchers have begun to document how some distressing and impairing meditation-related experiences are appraised as normative within contemporary contemplative traditions. In light of this, how can clinicians striving to be culturally sensitive make sense of these meditation-related difficulties without doing harm? What kind of etic frameworks can explain how sometimes things get worse before they get better, whereas other times things get worse before deteriorating further? And how might we begin to understand some of the causal factors that result in one trajectory over the other? This talk will explore these questions in relation to Jewish, Christian, and Islamic contemplative traditions and will seek to address them based on novel approaches to cognitive and clinical science.

Recent research in meditation science has expanded beyond the therapeutic applications of Buddhist and other Asian contemplative practices to explore advanced meditative states, including energy-like somatic experiences (ELSEs). While ELSEs have been studied in Buddhist and Yogic contexts, little is known about their occurrence in contemporary Jewish, Christian, and Islamic meditative traditions. This study investigated ELSEs through qualitative interviews with over 60 practitioners and meditation teachers from Abrahamic traditions in the West. It found a range of co-occuring phenomenology across somatic, affective, and perceptual domains as well as a range of religious and scientific interpretions of them. While there were many similarities with ELSEs documented in western Buddhist traditions in terms of the phenomenological features and the appraisals used to make sense of such experiences, there were also notable differences in both domains. Taken together, this study highlights the interplay of phenomenology, physiology, cultural context, and meaning-making in the generation and shaping of non-ordinary meditative states of consciousness.

Meditation practices are often engaged with an altered sense of self. Self-compassion, for example, relies on processes like perspective-taking and imagining empathetic actions toward oneself to alleviate suffering. It is unclear whether and how self-identification, based on multisensory own-body signals, can contribute to self-compassion meditation. We thus designed a self-compassion meditation practice in virtual reality (VR) that integrates a multisensory stimulation that promotes self-identification for a 3D avatar in an audio-guided meditation. Participants carried out this VR-based avatar self-compassion condition (ASC), which was compared with a standard self-compassion condition (SSC). Pre- and post-meditation assessments include scores of the State Self-Compassion Scale (S-SCS), Self-Other Four Immeasurables (SOFI) Scale, and an Own Body Transformation (OBT) task to measure perspective-taking ability. Participants demonstrated significantly higher compassion ratings compared to baseline measures, exhibiting higher S-SCS scores, increased positive qualities on the SOFI scale, and reduced negative valence in both meditation conditions. Perspective-taking ability, measured in the OBT task, was more strongly facilitated in the ASC. This study shows that integrating VR and an avatar with multisensory stimulation into self-compassion practice (ASC) is feasible, with both ASC and SSC meditations improved self-compassion ratings comparable to the baseline, and that ASC meditation enhances perspective-taking more than SSC. We argue that multisensory perceptual processes in VR can support self-related processes of perspective-taking and empathy when cultivating self-compassion.

Across cultures and times, people report experiences of contact with religious, spiritual, or other special beings. While spirits or gods could be understood as products of human narrativity, recent research based on neurocomputational predictive processing theory shows that feelings of presence (FoP) of special beings can be experimentally induced in the lab. However, what specific factors are influencing such subjectively real experiences?

In designing this registered report, we plan to test the following factors: sensory uncertainty (i.e., sensory deprivation), cultural learning (Catholic Christianity, providing prior expectations potentially overriding the actual sensory experience), and suggestive stimulus (“God helmet” - a sham device allegedly inducing FoPs by manipulating brain’s electromagnetic field). We plan to sample 119 participants for a medium effect size (f2 = 0.15). We will use a within-subject experimental design for manipulating sensory deprivation under the stable God helmet condition; participants will be recruited from three different populations - formally religious, religious with practice unrelated to feelings of presence, and religious practising specifically sensed presence feelings.

We expect that under a sham God helmet: participants trained in religious FoP practices will experience more FoPs; participants will experience more FoPs under sensory deprivation than without (incl. interaction with religious practice); participants trained in FoP-oriented practice will report more intense FoPs.

Examining the effects of uncertainty, cultural learning, and suggestive context on differences in subjective sensory experience can impact disciplines across fields - from neurology and psychology to human ethology. Thus, we would be happy to hear comments on and critiques of our design.

While we intuitively trust our senses to reflect reality, experiences like illusions and hallucinations show perception isn’t necessarily tied to the external world. This disconnect highlights the need for a metacognitive Sense of Reality (SoR) mechanism that continuously evaluates whether our perceptual experiences accurately reflect the world around us. While SoR is essential for navigating the world, the lack of ecologically valid paradigms has left its electrophysiological mechanisms largely unknown.

Here, we developed a novel, highly realistic VR-EEG paradigm to investigate this mechanism by presenting virtual hallucinations - designed to evoke a feeling of strangeness - across three domains: Self (height changes), Laws of Nature (gravity alterations), and Perception (color shifts). Twenty-three participants first performed a 2AFC staircase procedure to determine individual psychophysical thresholds for seven parametrically controlled hallucination conditions. Importantly, a baseline condition with a realistic change was included to dissociate SoR from mere change detection. During a subsequent EEG study, participants provided subjective “realness” ratings for the virtual hallucination and realistic conditions and identified the objective manipulation category.

Preliminary findings demonstrated that participants rated virtual hallucinations as significantly less real than realistic changes. Multivariate pattern analysis (MVPA) using linear discriminant classification revealed distinct spatiotemporal patterns that allowed successful decoding of trials with virtual hallucinations from realistic trials, starting ~350ms post-manipulation, across all domains.

These initial findings offer new insight into the electrophysiological markers of SoR and validate a novel VR-EEG method for its study. Ongoing work includes MVPA of subjective 'realness’, pupillometry, and a replication study.

Understanding how individuals subjectively experience the world is crucial for mental health research, yet the best means to elicit meaningful responses remains elusive. Here, we introduce a computational approach to map and compare subjective reports in the embedding space of Large Language Models. We use this method to examine how subjective reports of a series of visual stimuli relate to transdiagnostic psychiatric symptom dimensions (Wise et al., 2023). More specifically, we compare the predictions from 3 different sets of stimuli: the International Affective Picture System (IAPS), the Thematic Apperception Test (TAT), and the Rorschach Inkblot Test (RIT) using verbal responses from 210 participants to 30 visual stimuli. Embedded reports to IAPS images significantly predicted compulsive behaviors and intrusive thoughts (AUC = .83, p = .0024, corr.) and TAT images best classified mood and impulsivity symptoms (AUC = .84, p = .0017, corr.). Interestingly, RIT reports did not lead to any significant prediction, possibly due to excessive individual alignment with group trends, as indexed by a higher ratio of inter-to-intra-individual variability across images (cosine similarity, ANOVA F(2,207) = 3560.1, Rorschach > IAPS > TAT, all p < 0.001, corr.). Our results align with and extend recent findings that natural language descriptions of mental states strongly correlate with psychometric scales (Kjell et al., 2021). Thus, leveraging more predictive report elicitation methods could ultimately help identify new therapeutic targets for clinical interventions. In sum, adopting a computational psychiatric perspective, we presented methods to reveal the representational trends and distortions that color subjective experience and influence mental health.

Hallucinations are commonly understood as purely ‘inner’ phenomena arising from cortical misfirings that simulate perception without external stimuli. Challenging this neurocentric, perception-based model, I propose an 'embodied imagination' model, integrating an imagination-based account of hallucination with an embodied approach to imagination.

I argue that hallucination is a form of involuntary sensory imagination, akin to earworms, shaped by bodily and environmental structures, not just neural activity. This builds on enactivist accounts of imagination as a participatory, embodied activity (e.g., pretend play, tool-making) (Gallagher, 2017).

The imagination-based view treats hallucinations as degenerate imagination requiring the absence of external objects (Allen, 2015). However, this view struggles with ‘veridical hallucinations’, where one hallucinates something that happens to be present, and explaining how absent entities shape hallucinatory phenomenology.

The embodied imagination view model resolves these issues by emphasizing a bodily constraint on attention – one cannot simultaneously 'attend' to an imagined object and a perceived object in the same location (Sartre, 1940/2010) – and showing that hallucinations are shaped by imaginative engagement with environmental affordances, not merely the absence of stimuli.

This model applies both real and hypothetical cases. In schizophrenia, it aligns with phenomenological psychopathology, viewing hallucinations as ‘localized’ symptoms of ‘global’ disruptions in the experiential structure. Concerning hypothetical, 'subjectively indistinguishable' hallucinations, it challenges the idea that brain states alone suffice to generate hallucinations, arguing that the absence of external objects is also necessary.

By situating hallucination within embodied cognition, this model offers a compelling alternative to internalist views, reshaping broader debates on the perception-imagination distinction.

Introduction

Altered states of consciousness (ASC) feature complex interactions between low-level perceptual changes and global-level alterations (e.g., in sense of self and temporality). The role of sensory alterations – such as psychedelic visual hallucinations – in driving broader ASC-specific effects remains unclear. While the impact of unusual sensory conditions on consciousness has been studied using intense-rhythmic (stroboscopic) or unstructured (Ganzfeld) visual stimulation, less is known about more complex stimuli. Here we examine how machine-learning simulated hallucinations – in the form of the Deep-Dream VR platform – affect subjective experience and peripheral physiology.

Methods

In a series of exploratory studies, we manipulate presentation mode (stereoscopic vs. monoscopic; Pre-study), type and strength of Deep-Dream hallucinations (Study 1), comparing them against non-psychedelic Style-Transfer alterations (Study 2), and contrasting passive exposure with gaze-contingent hallucination propagation (Study 3). As primary outcome, we assess ASC-specific alterations using a novel questionnaire (excluding sensory-related items), complemented by free reports and measurements of cardiac activity, eye movements, and pupillometry.

Results

Pre-study data (n = 14) indicated greater ASC-specific alterations during simulated hallucinations compared to normal (unaltered) videos, with a trend towards enhanced effectiveness of stereoscopic versus previously used monoscopic hallucinations. The analyses from ongoing studies will compare questionnaire responses, thematically analyzed free reports, and physiological measures across parametrically varied conditions, relating response patterns to established markers of psychedelic states.

Conclusion

The Deep-Dream VR platform constitutes a powerful tool for studying sensory-driven alterations in consciousness, developed here for enhanced immersiveness, parametric diversity of simulated hallucinations, and real-time dependence on gaze dynamics.

The ability to distinguish real from unreal experiences, known as the Sense of Reality (SoR), is a core feature of normal perception and cognition. Crucially its disruption is a defining symptom of psychiatric disorders such as psychosis and derealization-depersonalization syndrome. Recent work using virtual hallucinations in immersive VR has demonstrated evidence that SoR judgments vary systematically with both the domain and magnitude of perceptual alterations, suggesting a neural mechanism monitoring perception for veridicality. Here, we present an fMRI-adapted paradigm designed to investigate the neural basis of SoR across multiple perceptual domains and to identify cortical regions implicated in perceptual reality monitoring. Fifteen participants completed a psychophysically controlled task while viewing a realistic virtual environment, first establishing individual perceptual thresholds for seven alteration conditions. These conditions, inspired by symptoms of altered perception from psychiatric and psychedelic states, represented three domains of altered perception and a veridical baseline condition: Nature (changes in gravity), Self (changes in first person perspective), and Perception (changes in color). Subsequently, stimuli were presented in the scanner in random order at different magnitudes based on the estimated psychophysical thresholds, while participants provided realness judgments under immersive conditions. Based on previous findings, we hypothesized differential engagement of sensory, association, and prefrontal areas and searched for domain-specific neural patterns. Detection of such mechanisms may advance our understanding of SoR and its relevance to both neurotypical perception and clinical disorders.

Hallucinations can be understood as resulting from alterations in the neural processes that construct conscious experiences out of sensory input. Research into conscious perception could therefore benefit from more refined means of characterising the phenomenology of hallucinations. Here, we introduce and validate the 6-Dimensional Visual Hallucination Questionnaire (6D-VHQ) to measure how the perceptual contents of visual hallucinations may vary across different hallucinatory experiences. While previously validated questionnaires on visual hallucinations are used within research on altered states of consciousness and mental health, these primarily assess whether certain types of hallucinations co-occur with other non-hallucinatory experiences such as ego dissolution or delusions. In contrast, the 6D-VHQ focuses specifically on the contents of visual hallucinations, capturing key attributes including vividness, level of detail, and complexity. We present results showing that the 6D-VHQ has high internal validity when applied to hallucinations induced by stroboscopic light, and high face validity when used with control image stimuli. Additionally, when used in the context of stroboscopic light, the 6D-VHQ reveals how changes in stroboscopic frequency influences the reported contents of hallucinations. Altogether, the 6D-VHQ provides a new method for assessing the perceptual contents of visual hallucinations. It provides researchers investigating visual hallucinations, altered states of consciousness, and phenomenology with a robust tool that can be applied across diverse hallucinatory contexts and aetiologies.

Hallucinations are a common non-motor symptom in Parkinson’s disease (PD). They range from minor, including passage and presence hallucinations and visual illusions, to well-structured hallucinations such as visual, auditory and tactile hallucinations, and are linked to a more rapid cognitive decline (Bejr‐Kasem et al., 2021, Anang et al., 2014). To enable the study of hallucinations in controlled laboratory settings, our group has developed a robotic system that, via sensorimotor conflict, induces presence hallucinations-like symptoms, or in other words the sensation of someone being nearby (Blanke et al., 2014). Notably, PD patients who experience hallucinations as one of their symptoms exhibit heightened sensitivity to this sensorimotor conflict (Bernasconi et al., 2021). Our current study integrates the behavioural data from the robot experiment with neuropsychological assessments and resting state functional magnetic resonance imaging (rs-fMRI), to investigate neural correlates of cognitive impairment in PD patients with symptomatic hallucinations. Our dataset includes a total of 53 patients categorised into three groups: no hallucinations (n = 19), minor hallucinations (n = 18), and a combination of minor and well-structured hallucinations (n = 16). Our preliminary analysis employing partial least square correlation suggests that mainly the functional connectivity between subcortical areas and the visual network is associated with both the sensitivity to the robot task and frontal subcortical scores of the PD - Cognitive Rating Scale in the three groups of PD patients. Ongoing analyses will further explore this relationship to advance our understanding of neural mechanisms underlying cognitive decline in PD patients with symptomatic hallucinations.

According to the Bayesian predictive processing framework, conscious perception arises from the brain’s probabilistic integration of sensory evidence (likelihood) and prior knowledge (priors). This process relies on dynamically weighting these inputs to predict and update perceptions in response to prediction errors. Aberrant weighting of priors has been linked to hallucinations, a key symptom of psychotic disorders. However, there is conflicting evidence regarding the question whether hallucinations result from abnormally weak or strong priors. A hierarchical model posits that weak priors at lower levels of the predictive processing hierarchy may be compensated by stronger priors at higher levels, but empirical validation is lacking. In a large online study, we test this hierarchical model using two perceptual decision-making experiments in healthy participants with varying hallucination proneness. In the first experiment, participants perform a noisy-image discrimination task, designed to assess reliance on low-level priors by measuring choice-history biases. We hypothesize reduced choice-history biases in individuals with higher hallucination proneness, reflecting weak low-level priors. The second experiment probes the influence of high-level priors by measuring the effect of learning on the recognition performance for degraded images. We predict enhanced recognition performance for degraded stimuli after learning in individuals with higher hallucination proneness, indicating greater reliance on high-level priors. Importantly, we expect an inverse correlation between both effects across individuals. Preliminary results of this study will be presented. This research aims to reconcile opposing findings on prior weighting in hallucinations and will crucially refine mechanistic models of conscious perception in psychosis.

We effortlessly assign meaning to what we see and experience, yet understanding how meaning becomes distorted by mental health conditions remains a significant challenge. Brain imaging studies reveal that meaning is organized within a complex semantic “map” that tiles the cerebral cortex. Encoding models based on deep learning (DL) embeddings (such as CLIP) can model “generic” semantic information based on the visual properties of images and could potentially capture how an individual’s interpretation departs from what is expected. For instance, individuals with strong delusional tendencies may attribute false meanings to everyday situations, which may be revealed by decreased modeling performance in relevant brain regions.This project examines how delusion severity impacts DL models' ability to predict neural activity in individuals with schizophrenia spectrum disorders. Using data from the PRISME database, we analyzed functional MRI (fMRI) recordings from 13 participants with a history of psychosis. Each participant underwent 10–12 scanning sessions over 12 months while viewing objects from the THINGS image database. We trained encoding models to predict brain activity based on the embeddings of the CLIP model. Preliminary analyses show an inverse relation between delusion severity and the modeling performance (i.e., correlation between predicted and real brain activity) in regions such as the superior parietal lobule, superior frontal gyrus, precuneus, and central sulcus. These findings suggest that distortions in these regions may contribute to altered conscious experience in delusions. By leveraging the power DL-based encoding models, this research advances our understanding of how delusion may shape conscious perception.

Introduction: Presence hallucinations (PH), non-veridical sensations of someone being around you, are predictive of cognitive decline, dementia and earlier home placement in Parkinson’s disease (PD). Despite PH’s phenomenological richness, outcomes related to its perceived identity have not been explored. In this study we investigated the mechanisms underlying PH with a perceived identity in PD.

Methods: From a sample of 77 PD patients, we selected 27 patients reporting PH without dementia, and group them in PH with identity (NiPH=12) and without identity (NuPH=15), compared their disease progression stage, cognitive functions (PD-CRS), robot-induced PH sensitivity in a sensorimotor task and brain’s resting-state functional connectivity (rsfMRI: NiPH=10, NuPH=12) within the sensorimotor PH-network (Bernasconi, Blondiaux et al., 2021).

Results: iPH patients have a more advanced stage of PD (Hoehn Yahr: p=0.036), report more daily-life motor impairments (p=0.028) and more visual illusions (p=0.043), while they did not differ in their robot-induced PH sensitivity (p=0.835). Performing ROI-to-ROI analysis of PH network, we found that iPH vs. uPH patients have stronger functional dysconnectivity between inferior frontal gyrus and posterior superior temporal sulcus (p-FDR=0.020), selective on the right side. Furthermore, the degree of dysconnectivity is correlated with patients’ stronger (posterior) cognitive impairment (p=0.037).

Conclusion: Our findings show that the iPH group differs on behavioural and neural levels, compatible with a more advanced form of PD in iPH vs. uPH patients. We suggest that PH phenomenology should thus be more widely investigated in fundamental research and as a potential clinical marker for more severe forms of the disease.

Visual mental images vary in their subjective vividness, typically measured on a unidimensional scale. But how different visual qualities influence imagery vividness is poorly understood. Here, we present results from a novel method for reconstructing visual properties of mental images across varying vividness ratings. On each trial, subjects saw a line drawing of an object and were instructed to vividly imagine it in full detail. Then, they rated the vividness of their mental image on a 1-to-5 scale. Finally, a Voronoi tessellation pattern—a tiling of differently-colored shapes—appeared on the screen with three sliders controlling its sharpness, opacity, and saturation. Subjects abstracted the qualities of their mental image and applied them to the Voronoi pattern by adjusting the sliders. A mixed-effects linear model with high goodness of fit revealed that all three dimensions significantly predicted subjects’ vividness ratings. At the group level, sharpness was the strongest predictor (followed by opacity and then saturation), explaining most of the random effects variance. Sharpness was also the top predictor for the greatest number of subjects. While our results showed individual differences in what visual features influence vividness ratings, sharpness’ primacy can help guide further research into the visual properties and neural basis of mental imagery. Additionally, the modeling results were orthogonal to subjects’ VVIQ scores, suggesting that the pattern of reliance on these dimensions was independent of overall imagery capacity. In conclusion, despite the subjective nature of mental imagery, our method robustly reconstructs vividness ratings by quantifying the contributions of different visual features.

Aphantasia, the inability to generate voluntary visual imagery, is typically identified using the Vividness Of Visual Imagery Questionnaire (VVIQ) (Marks, 1973). Does this VVIQ-based classification really capture true mind blindness? To investigate this question, we recruited a massive cohort of 1,295 Prolific participants (aged 18–45) who completed 120 trials of an imagery task and subsequently completed the VVIQ. In each trial, two images from the 73,000-image Natural Scene Dataset (Allen et al., 2022) were presented. Participants then imagined the cued stimulus for 3 seconds and rated its vividness on a continuous scale (0–100). Our results indicate that 40 participants showed low imagery in the task (- 2 SD) while 37 showed low imagery on the VVIQ. Interestingly, only 15 participants scored below 2SD on both measures. In fact, within the low-imagery participants, the correlation between performances on the task and those on the questionnaire was not significant (p = 0.07), indicating a weak convergence between the two measures. Furthermore, we identified 9 participants with aphantasia using a strict VVIQ cutoff (minimal score) and even those participants showed residual imagery in the task (M = 24, SD = 34) with some reporting vividness reaching as high as 86 on the scale. As such, defining aphantasia using the VVIQ may misclassify individuals, as a lack of imagery on the questionnaire does not necessarily translate into real-time vividness reports. These results raise the question of which method should be used to identify imagery extremes, as both seem to exhibit little convergence in low-imagery participants.

Subjective reports on the vividness of visual imagery varies widely across individuals. The neural basis of this variability remains largely unknown. Using recently developed computational techniques, we found that human brain activity (7T fMRI) during mental imagery varies across fewer coding dimensions than activity during vision, and that the number of coding dimensions for mental imagery varies considerably across the eight subjects in our experiment. Previous fMRI research (Cui et al. 2007; Lee et al. 2012) has pointed to inter-individual differences in primary visual cortex (V1) as a potential source of variability in mental imagery vividness. Therefore, we hypothesized that variation in the number of mental imagery coding dimensions in V1 across subjects might explain variation in subjective reports of mental imagery vividness. Indeed, we found that individual differences in imagery vividness ratings obtained prior to scanning correlated with the number of imagery dimensions in V1 (r=0.71, 95% bootstrap CI: 0.15-0.98 ). No association was observed between subjects’ imagery dimensionality in other visually responsive ROIs and their vividness reports. Since each coding dimension can encode a distinct visual feature (for example, color, animacy, location of stimulus along a particular axis), a V1 with few (many) coding dimensions will represent few (many) distinct features. Our results suggest that individuals who experience vivid mental imagery have V1s that encode more imagined features than individuals who experience weak mental imagery.

To have any practical use for the study of consciousness or for downstream applications, vision decoding models that are trained to reconstruct seen images from human brain activity must be able to generalize to internally generated visual representations, i.e., mental images. In an analysis of the recently released NSD-Imagery dataset, we demonstrated that while some modern vision decoders can perform quite well on mental image reconstruction, some fail, and that state-of-the-art (SOTA) performance on seen image reconstruction is no guarantee of SOTA performance on mental image reconstruction. Motivated by these findings, we developed MIRAGE, a method explicitly designed to train on vision datasets and cross-decode mental images from brain activity. MIRAGE employs a simple and robust ridge regression backbone, maps to multi-modal text and image features, and adopts multi-modal conditioning and small image embeddings as input to the Stable Cascade diffusion model. We conduct extensive evaluations with human raters and image feature metrics, establishing SOTA results for mental image reconstruction on the NSDImagery benchmark. Our work indicates that–given the right architecture–existing large-scale datasets using external stimuli are viable training data for decoding mental images, and warrant optimism about the future success and utility of mental image reconstruction for externalizing private conscious states, providing a novel and valuable tool for the study of consciousness.

Variations in mental imagery ability have been linked to emotional, cognitive, and personality traits, yet a comprehensive framework explaining their connection to mental health has been lacking. This study proposes the concept of an "inwardly focused cognitive style," which integrates traits related to mental imagery vividness, interoceptive awareness, and personality factors. Through Pearson correlation analyses, canonical correlation analysis, and structural equation modelling, we identified a latent inward trait (LIT) capturing key cognitive and personality dimensions based on cross-validated data from 591 participants across two independent sites with MRI data.

Results showed that the LIT, characterized by heightened interoceptive awareness, mindfulness, and personality traits such as openness, conscientiousness, and extraversion, plays a dual role in linking mental imagery vividness (measured by the VVIQ) to mental health outcomes. Specifically, LIT is associated with enhanced emotional awareness and reduced alexithymia, which mediate the beneficial effects of vivid imagery on mental health. Our results also highlight potential vulnerabilities, as higher LIT also related higher frequency of involuntary autobiographical memories, which can exacerbate symptoms of anxiety and depression.

Additionally, LIT mediated the relationship between VVIQ and graph-theoretical measures of local efficiency in resting-state functional connectivity networks implicated in imagery processing, suggesting a neural basis for the cognitive style.

Our study provides a framework for understanding how an inwardly focused cognitive style link mental imagery and mental health. Our findings challenge the traditional view of imagery as a primarily sensory process by highlighting its connection to personality traits and emphasizing the functional aspects of the phenomenon.

The study of mental imagery extremes—aphantasia (lack of mental imagery) and hyperphantasia (exceptionally vivid imagery)—provides an opportunity to revisit key assumptions in theories of mind. These variations highlight inner diversity and challenge the common assumption that sensory-based mental representations (in the form of imagery) are a universal feature of cognition. Many cognitive models, particularly simulation theories, implicitly rely on this assumption, raising questions about how cognition functions in individuals who do not experience mental imagery in the typical way.

This talk examines how mental imagery diversity prompts a re-evaluation of simulation-based theories, using simulation theories of dreaming as a case study. These theories propose that dreaming serves as an “offline simulation” of experience, a conscious virtual simulation of a rich phenomenal world. If this is the case, how do aphantasics dream? Research suggests that some report conceptual, non-visual dreams, while others dream vividly but recall dreams less frequently. These findings raise broader questions about the role of sensory simulation in dreaming and, by extension, its necessity in waking cognition. Additionally, studying the dream experiences of aphantasics may provide new insights into the ongoing debate in philosophy and cognitive science about the nature of aphantasia—specifically, whether it involves unconscious imagery, a selective deficit in voluntary visualization, or a complete absence of mental imagery altogether.

More broadly, this talk suggests that studying cognitive diversity can refine our understanding of the mind, encouraging a more flexible approach to cognition that accounts for different ways of processing and representing experience.

The current study aligns with a highly active research field investigating the neural mechanisms underlying the contents of consciousness. This study aimed to explore the link between visual external stimuli and the corresponding content of consciousness by manipulating the latter through a perceptual phenomenon known as face pareidolia (i.e., the tendency to perceive faces in inanimate objects).

A group of healthy participants was recruited. Three sets of target stimuli were presented: faces, houses, and face-like houses. Participants were to be focused on the presentation of the stimuli while maintaining central fixation. Catch trials (i.e., target stimuli with an additional red central dot) were included. Participants were to respond as fast as possible to catch stimuli only. EEG signal was recorded throughout the experiment. The N170 component was taken into consideration, representing the most commonly used Event-related potential (ERP) marker for face processing.

ERP results revealed a N170 modulation across experimental conditions: the amplitude was larger in response to faces compared to both face-like houses and houses. In turn, N170 was found to be larger for face-like houses than for houses. Therefore, the N170 response to face-like houses suggests these stimuli are processed like real faces.

Our findings show that, when face perception neural processes are engaged, the experience of a face emerges regardless of the nature of external stimulation, thus demonstrating the constructive nature of perceptual awareness.

Through specific experimental design and stimuli, our study enriches a research domain that is extremely crucial for everyday human life.

Aphantasia is defined as the severe diminution or complete absence of mental imagery, most commonly the inability to visualize (Zeman et al. 2025), and affects around 3.9 percent of people (Dance et al. 2022). While the cognitive and neural profiles of aphantasia are currently hotly debated, a growing body of behavioral evidence shows that people with aphantasia can perform a range of tasks that were previously assumed to rely on visual imagery (e.g., mental rotation [Pounder et al. 2022], visual memory tasks [Bainbridge et al. 2021], etc.). This poses a puzzle: How can people who lack visual imagery solve visual imagery tasks? Recently, it has been proposed that the puzzle might be solved by positing that people with aphantasia have unconscious imagery, meaning that despite not reporting mental imagery experience, they still possess the underlying neural correlate (Nanay, 2021; Michel et al. 2025). However, we argue that emerging neuroscientific findings consistently point against unconscious imagery in aphantasia. Specifically, while mental imagery has been shown to rely on perception-like shared representations (e.g., Naseralis et al. 2015), activity during imagery attempts in aphantasic subjects, despite occurring in perceptual areas, does not rely on shared representations, as evidenced by a lack of cross-decodability and representational overlap between imagery and perception conditions (Chang et al. 2025; Liu et al. 2025). We invoke cognitive plasticity and neural reuse (Anderson, 2010, 2014) to suggest that these findings indicate that people with aphantasia may utilize the same neural substrate (e.g., V1) for distinct cognitive and representational strategies.