Title: From an implicit sense of self to an explicit self-consciousness

Author: Shogo Tanaka

Abstract: In the past two decades, an abundance of research has explicated the implicit sense of self in terms of embodiment at the intersection of cognitive science and phenomenology. The implicit sense of self, also referred to as the minimal self, is considered to be embedded within direct embodied experiences. As every direct experience always involves an implicit sense that “this is my experience,” we can consider it the fundamental ground of the phenomenal self. The minimal self has been investigated in two ways, that is, as a sense of ownership and as a sense of agency, using experimental paradigms such as the rubber hand illusion. What I would like to argue in this presentation is how we can expand our research from an implicit sense of self to an explicit self-consciousness while keeping its focus on embodiment. An explicit self-consciousness is different from an implicit sense of self in terms of the structure that involves reflective consciousness represented as “I think of me.” Although this type of consciousness may seem abstract and disembodied on the surface, Maurice Merleau-Ponty (2012) pointed out that the experience of double sensation induced by self-touching is “a sort of reflection.” (p. 95).  Taking his words as a clue, I will explicate the structure of reflective consciousness as an embodied experience in which one perceives one’s own body. It is possible to find such structure not only in tactile sense but also in auditory sense.  As one’s body appears as a subject of perception as well as an object of perception, the body itself is open to sensory experiences of self-reflection that serve as a prototype of “I think of me.” The necessary components of “I” and “me” that constitute an explicit self-consciousness has its basis in bodily experiences.

Title: Is this my rubber hand? - A critical reflection on body ownership illusion studies

Author: Andreas Kalckert

Abstract: The introduction of the rubber hand illusion (RHI) has significantly impacted the field of cognitive neuroscience, offering a novel way to investigate the processes underlying body ownership and self-awareness. Originating over 25 years ago, the RHI has sparked considerable interest in the study of embodiment, leading to a proliferation of research paradigms aimed at understanding these complex phenomena. However, despite extensive investigation, several aspects of body ownership illusions remain elusive, prompting recent critiques of experimental paradigms. Critics have questioned the authenticity of these effects, stirring debate on their perceptual underpinnings and contributions to our comprehension of body ownership.

Notwithstanding these criticisms, the notion of multisensory integration remains a compelling explanatory framework for these experiments. This paper seeks to defend this perspective while highlighting areas of the illusion that warrant further exploration. Specifically, it aims to address the illusion's temporal dynamics and the subjective nature of the experience, areas that have received insufficient attention in past research. Moreover, the heterogeneity in methodological approaches and definitions poses a significant challenge, reflecting in the varied conduct of experiments. This paper will present new data and critical reflections that offer a more nuanced understanding of RHI experiments, focusing on the onset of the illusion, the use and interpretation of questionnaire data, and the variability in individual experiences.

By advocating for a more uniform approach and consensus on key methodological aspects, this presentation aims to address current challenges in the field and, consequently, enrich our understanding of embodiment from an experimental viewpoint.

Title: It should have been me: from owning to willing

Author: Lorenzo Pia

Abstract: It is largely accepted that the so-called “Sense Of Agency” SOA (i.e., “this action is due to my own will”) comes from the interplay among a variety of internal, motor-related, efferent signals (e.g., motor intentions, planning, sensorimotor predictions, and so on). Notwithstanding, new evidence supports the idea that those body-related (purely afferent) signals subserving Body Ownership BO (i.e., “this body is mine”) act per se on human conscious awareness of willed actions.

I will discuss data from both stroke patients and intact brain functioning supporting such a view. Subsequently, I will suggest that, whenever the context requests, BO can directly act on agency attribution (i.e., regardless of efferent signals). This claim supports the notion that a coherent subjective experience of willed actions (i.e., “this willed action is being achieved by my own body”) needs both awareness of being an agent and of owning the body.

Title: Developing awareness: A visceral afferent training perspective on the origins of consciousness

Author: Andrew W. Corcoran

Abstract: Embodied and enactive perspectives on consciousness are concerned with the question of how conscious experience is constituted through the agent’s bodily composition and interaction with the world. Although such approaches are often characterised by an emphasis on temporal dynamics, surprisingly little attention has been paid to the way developmental processes may shape or constrain consciousness. In this presentation, I will argue that fundamental elements of conscious experience are rooted in early developmental trajectories established during gestation. Specifically, I will outline a recently-proposed account of the way neural development is structured by spontaneous rhythmic dynamics arising from visceral systems within the foetal body, which serve to attune the cognitive system towards particular patterns of sensory input. I will then develop this ‘visceral afferent training hypothesis’ to show how the earliest actions of the nascent nervous system give rise to self-specifying information delineating controllable bodily processes from external perturbations. By virtue of asserting control over its proximal (bodily) environment, the developing brain begins to establish its self as both a subject and an agent of experience.

Title: A neurophenomenological investigation of the Rubber Hand Illusion: Combining elicitation interviews and EEG

Authors: Camille Lépingle, Miyuki Azuma, Mika Ishizu, Momoka Kimuro, Fugo Suzuki, Yves Rossetti, Jean-Michel Roy, Sotaro Shimada

Abstract: Since the seminal work of Botvinick and Cohen (1998), the Rubber Hand Illusion (RHI) has been the primary experimental paradigm used to study the sense of ownership and the neurocognitive processes involved. However, although the sense of body ownership is considered one of the constitutive elements of minimal self-awareness, few RHI studies pay close attention to its phenomenological dimension, and brain activity analyses are generally driven by predetermined experimental conditions. Therefore, the present study proposes a neurophenomenological investigation of the sense of body ownership based on first-person data collected through on-line elicitation interviews to guide the analysis of brain activity recorded under two experimental conditions. Sixteen healthy Japanese students (15 right-handed, 8 females, age 21.25±0.56 years) were asked to describe their experience while undergoing synchronous or asynchronous visuo-tactile stimulations. Electroencephalography (EEG) was recorded using 30 electrodes according to the extended international 10-20 system. The phenomenological analysis of the first-person data highlighted the existence of various successive phases in participants' lived experience as well as the occurrence of illusions both in the synchronous and asynchronous conditions, thus finding new evidence to support the initial results of a previous phenomenological study by Valenzuela Moguillansky et al. (2013). Our analysis lead to distinguish between six subphases before reaching the state of “Illusion” (phase 2) or “Absence of illusion” (phase 2’) and to isolate three main phenomena that can be classified as “illusions”. The phases identified were subsequently used to cluster the collected EEG data, thus, offering a new perspective on the neurocognitive mechanisms involved in the sense of ownership, closely tied to lived experience.

Title: A mathematical perspective on the role of embodiment in predictive coding frameworks

Author: Manuel Baltieri

Abstract: In the past few decades, Bayesian frameworks such predictive coding (processing) and active inference have emerged as a dominant paradigm for the study of cognition and consciousness, often claiming to fully embrace crucial aspects of embodied theories of the mind. In this work we look at some of the mathematical assumptions underlying these Bayesian frameworks and question the extent to which they can be said to conform to principles from embodied cognitive science.

To do so, we will use the fact that in recent years, parts of these frameworks have been recast in more abstract mathematical terms, providing principled accounts of Bayesian inference in the brain using constructions from (applied) category theory (St Clere Smithe, 2023), and an equational language in graphical form, string diagrams, to streamline and generalise current theories to a larger class of (dynamical) systems (Tull et al., 2023). 

We will then follow the tracks of St Clere Smithe (2023) and Tull et al. (2023), and further propose a characterisation of the interactions between an agent and its environment inspired by classical formulations of dynamical systems-based approaches to embodied cognition (Beer, 1995) in a categorical framework via structure preserving relations, i.e. bisimulations, between an agent and its environment.

This work will shed (some) light on the relation between embodied cognition and predictive processing, showing cases where current predictive coding/processing frameworks appear as fundamentally limited from an embodied perspective.

Title: Category-Theoretic Approach to the Sense of Ownership: Implications for Life-Mind Continuity
Author: Ryuzo Hirota

Abstract: This presentation extends a category-theoretic framework previously developed for understanding autonomy in living systems (Hirota et al., 2023) to the phenomenon of the sense of body ownership, with implications for the continuity between life and mind. The rubber hand illusion, in which synchronous multisensory stimulation causes a fake hand to be experienced as part of one's body, reveals that the sense of ownership is not strictly delineated by the physical body but depends on the coherence of sensory information from different modalities. This finding is consistent with the principles of category theory, which characterizes objects not in terms of their intrinsic components but in terms of their relationships called “arrows” with other objects and themselves. In other words, in a category, the characteristic of an object is solely defined by what kind of arrows it 'connects' to each other. 

Here, I hypothetically propose a monoid (category with a single object) representing a sense of ownership, where the object represents the (feeling of) “self” and the arrows represent various sensory modalities. The composition of arrows represents the integrability or coherence of the multisensory information. In this view, the rubber hand illusion occurs when arrows representing visual and tactile inputs related to the rubber hand are composed because of their temporal and spatial coherence, resulting in a change in the characterization (range) of the self. This formalization suggests a deep continuity between the autonomous organization of living systems and the structure of embodied cognition. It enables us to formalize the role of membranes in autonomous living systems (i.e., self-distinction) and the sense of bodily ownership in subjective experience in a unified way: Their essence lies in “connecting” various loop-like relationships. The category-theoretic framework thus provides a unified way of describing the phenomenological mind and the dynamics of living systems.

Reference:

Hirota, R., Saigo, H., & Taguchi, S. (2023). Reformalizing the notion of autonomy as closure through category theory as an arrow-first mathematics. ALIFE 2023: Ghost in the Machine: Proceedings of the 2023 Artificial Life Conference. https://doi.org/10.1162/isal_a_00627

Title: Hybrid Embodied Agency in Human - AI Interactions

Authors: Anna Ciaunica, Shaun Gallagher

Abstract: For most of us, most of the time, our experiences seem to be tacitly accompanied by a sense of self – a sense of being an embodied agent within a world, among but distinct from others (Gallagher 2000). Everyday experience also seems to involve experiences of agency; namely, the feeling that I am in control of my own bodily actions, that I can leverage them to access and change the external world’ (Gallagher 2000; Haggard 2017). It is now well established that humans attribute human-like states to artificial others. However, the effect of interacting with artificial minds and bodies on the human sense of agency is less understood. In this talk we will present theoretical and empirical work looking at embodied joint agency in human/ human versus human/ robotic and virtual agents. Specifically, we will outline the key role of the human embodiment and sense of self in establishing joint agency with artificial others. We will argue that in order to establish a “sense of we” with both humans and artificial agents, humans need to feel connected to their own bodies first. We introduce the notion of ‘hybrid agency’ to describe these new, technologically mediated ways to embody and control in tandem human and artificial minds and bodies in real and virtual environments. Do we develop a sense of hybrid “we” in interacting with these artificial agents? If yes, in which sense this sense of “we” is different from the sense of togetherness that we have when we interact with biological systems? We will discuss key implications of these questions on recent efforts to design autonomous and interactive artificial others.

Title: Mapping the space of altered states of consciousness

Author: Paweł Motyka

Abstract: Altered states of consciousness are defined as a transient change in the overall pattern of subjective experience. While states of consciousness are increasingly better understood across a spectrum ranging from unconsciousness (e.g., coma or anesthesia) to the waking state, we still lack a unified approach for capturing the qualitative diversity within conscious wakefulness. A fundamental challenge is to identify the dimensions that can effectively structure the coarse-grained representation of the phenomenological landscape. Here we aimed to create the empirically grounded classification of pharmacologically induced altered states using a bottom-up approach. In a preregistered study (n = 739), participants assessed the dissimilarity between pairs of vividly remembered experiences associated with the use of various psychoactive substances. The ratings were given on a continuous labeled dissimilarity scale, originally used for capturing unimodal qualia spaces. We employed the multidimensional scaling technique (MDS) to create models representing the averaged dissimilarity ratings between states as distances in a geometric space. First, we observed that states induced by pharmacologically similar substances clustered closely, serving as a control validation for our methodology. The primary dimension, interpreted as the 'intensity of mind alteration', positioned baseline/ordinary state alongside states evoked by depressants, and stimulants at one end, while situating psychedelic- and dissociative-induced states at the other. The second dimension covered the inhibition-stimulation spectrum, ranging from states evoked by depressants and opioids to classical stimulants. The MDS analysis was complemented by other multidimensional reduction techniques (PCA, UMAP, and t-SNE), which enabled us to preserve subject-level information. Consequently, we derived a set of empirically grounded models mapping dissimilarity ratings for radically diverse experiences and reflecting the variability of these ratings across subjects. We propose that such avenue of research may eventually inform our understanding of the dimensionality, topology, and metrics underlying the landscape of consciousness states.

Title: Altered States of Viscereality: Augmenting Breathwork with Bio-Responsive Virtual Reality

Author: George Fejer

Abstract: This project explores enhancing breathwork with bio-responsive virtual reality (VR) to induce altered states of consciousness (ASC). Our goal is to design a bio-responsive VR interface that synchronizes breathing patterns with spatial changes—expanding during inhalations and contracting during exhalations—to influence conscious states. Previous research on breathing and embodied cognition has shown that people perceive larger quantities after inhaling compared to exhaling (Belli et al., 2021). This is likely because they associate the expansion of lung volume with larger quantities. In our VR scenario, we aim to directly link the subjective sense of spaciousness to lung mechanics, making participants feel as though the room is breathing, thereby inducing an altered state of consciousness. We plan to achieve this by either coupling a VR headset with a breathing mask (Tatzgern et al., 2022) or placing the VR controller on the belly to measure diaphragmatic breathing (Rockstroh et al., 2021). In one condition, we will couple the expansion of space to lung volume; in the other, we will couple the expansion of lung volume to contraction. Our hypothesis is that the subjective sense of space will be larger in the former condition.To evaluate the effectiveness of our VR setup, we will use psychometric time series tracing, behavioral assessments, body-size estimation, random number generation, and neurophysiological markers related to the representation of peripersonal space (Serino et al., 2018). In light of previous work, which has shown that Breathwork can effectively induce ASCs comparable to those produced by psychedelics (Bahi et al., 2024; Lewis-Healey et al., 2024), we aim to explore methods for embodied technologies to influence our conscious state,  with the hope of developing non-pharmacological methods for inducing ASCs with therapeutic applications.

References

Bahi, C., Irrmischer, M., Franken, K., Fejer, G., Schlenker, A., Deijen, J. B., & Engelbregt, H. (2024). Effects of conscious connected breathing on cortical brain activity, mood and state of consciousness in healthy adults. Current Psychology, 43(12), 10578-10589.

 Belli, F., Felisatti, A., & Fischer, M. H. (2021). “BreaThink”: breathing affects production and perception of quantities. Experimental Brain Research, 239(8), 2489-2499.

 Lewis-Healey, E., Tagliazucchi, E., Canales-Johnson, A., & Bekinschtein, T. (2024). Breathwork-Induced Psychedelic Experiences Modulate Neural Dynamics. bioRxiv, 2024.2002. 2019.580985.

 Rockstroh, C., Blum, J., & Göritz, A. S. (2021). A mobile VR-based respiratory biofeedback game to foster diaphragmatic breathing. Virtual Reality, 25(2), 539-552. https://doi.org/10.1007/s10055-020-00471-5

 Serino, A., Noel, J.-P., Mange, R., Canzoneri, E., Pellencin, E., Ruiz, J. B., Bernasconi, F., Blanke, O., & Herbelin, B. (2018). Peripersonal space: an index of multisensory body–environment interactions in real, virtual, and mixed realities. Frontiers in ICT, 4, 31.

 Tatzgern, M., Domhardt, M., Wolf, M., Cenger, M., Emsenhuber, G., Dinic, R., Gerner, N., & Hartl, A. (2022). Airres mask: A precise and robust virtual reality breathing interface utilizing breathing resistance as output modality. Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems,

Title: The Embodiment of Embeddedness through Huatou Chan: Data and Insights from Chinese Buddhism, Deep Ecology, and Neuroendocrinology

Author: Brianna Morseth

Abstract: Coining the term ecology, zoologist Ernst Haeckel’s definition encompasses its panoramic scope: “By ecology, we mean the whole science of the relations of the organism to the environment including, in the broad sense, all the ‘conditions of existence.’” Meanwhile, eco-philosopher Arne Naess inextricably embeds identity within the interrelational matrix of the broad manifold of nature. Existentially relational in context, ecological theory has since been applied to a range of fields, bridging philosophy and practice while integrating scientific methods toward the study of interconnection and its neural correlates. Especially keen on contemplating and challenging the sense of self in relation to broader ecologies are the varieties of Buddhism. However, no empirical studies have yet examined the neuro-ecology of Chinese Buddhist contemplative practices, particularly in the context of a monastic ecosystem. Thus, employing ecological and neuroendocrinological methods, this paper reports on data collected through monastic fieldwork in Taiwan on the effects of huatou 話頭, a form of meditation from Chan Buddhism invoking contemplative doubt through variations on the inquiry “Who am I?” Based on quantifiable trends in recycling, food waste habits, and enzyme-linked immunosorbent assay of saliva samples, results indicate an increase in both environmentally sustainable behavior and oxytocin, a hypothalamic neuropeptide hormone expressed in primarily interrelational contexts, as a function of huatou Chan and time on retreat over a one-month window of observation. Findings are discussed in the context of deep ecology and existing research in neuroendocrinology and phenomenology on the relationship between oxytocin and sense of self. This study is the first to empirically measure environmentally sustainable behavior and oxytocin as dependent variables in the context of Buddhist meditation practice. Hence, it offers an empirically testable framework linking oxytocin to sense of self and behavior and provides a foundation for future studies in the growing field of contemplative science.

Title: The emergence of agency or: How phenomenological methods can guide scientific research

Authors: Alessandro Solfo, Cees van Leeuwen

Abstract: In his manifesto for neurophenomenology, Varela (1996) suggests three “large issues” for testing his nonreductive method for the study of human consciousness: “attention,” “present-time consciousness,” and “body image and voluntary motion” (pp. 341–342) (i.e., sense of agency). Whereas the first two issues have been adequately addressed, sense of agency remains unexamined. We adopted a generalization of neurophenomenology—front-loaded phenomenology—to trace the emergence of agency. Unlike neurophenomenology, front-loaded phenomenology allows explorations at the limits of life (e.g., fetal stages). Our study consists of two stages: phenomenological reduction and scientific research. Phenomenological reduction redefines agency by identifying its necessary structure (essence), free of the inconsistencies of previous accounts. By apodictic self-evidence, we found that, in essence, the sense of agency is constituted by two complementary aspects: the feeling of being engaged in movement (bodily agency) and that of controlling, through movements, events temporally related to those movements (extended agency). This redefinition was front-loaded into scientific research designs. Through a systematic review of ontogenetic studies, we situated the emergence of bodily agency in the coordination patterns that humans encounter from intrauterine life (e.g., maternal–fetal heartbeat coordination). In a controlled experiment, we investigated the emergence of extended agency. When participants coordinated their finger tapping with periodic flashes, they felt to be controlling, through tapping, the flashes. Results indicated that sensorimotor coordination, operationalized as fractal time series, is necessary and sufficient for extended agency. Our study demonstrates that both aspects of agency stem from coordination and, more broadly, that phenomenological methods enable rigorous descriptions of experience that strengthen the validity of behavioral research. 

Reference

Varela, F. J. (1996). Neurophenomenology: A methodological remedy for the hard problem. Journal of Consciousness Studies, 3(4), 330–349.

Title: How the Brain Performs Phenomenology: Neural Correlates of Phenomenological Attitude toward Conscious Experience

Authors: Satoshi Nishida, Hiro Taiyo Hamada, Takuya Niikawa, Katsunori Miyahara

Abstract: Phenomenology is considered one of the most promising approaches to studying conscious experience. It posits that a rigorous study of conscious experience requires a transition from the “natural attitude” (NA) to the “phenomenological attitude” (PA). NA is our ordinary stance, in which our attention is focused on external objects and events, whereas PA is a distinctive, reflective stance in which our attention is focused on our conscious experience itself. Despite its theoretical importance in both philosophy and science of consciousness, the neural mechanisms underlying PA remain unknown. This uncertainty undermines the scientific credibility of phenomenological approaches to consciousness research. To address this, our study investigated these neural mechanisms using fMRI experiments with a novel behavioral task. This task required participants to switch between NA and PA in relation to their stimulus-evoked subjective experiences. Participants were presented with two sentences designed to induce either NA or PA and were asked to select the one that best captured their experience. We found that activation patterns in action-related brain regions—the premotor cortex, posterior parietal cortex, supplementary motor area, and cerebellum—successfully discriminated between NA and PA. Furthermore, the activation strength in these regions was lower in PA, suggesting that PA involves the suppression of action-related neural processes. These findings provide the first empirical evidence of the neural correlates of PA, establishing valuable guidelines for future research in consciousness, grounded in an integrative framework of neuroscience and phenomenology.

Title: Clinical Neurophenomenology: Integrating Psychiatry and Enactive Cognitive Science.

Author: Aleš Oblak

Abstract: In psychiatry, we are faced with the so-called integration problem, the question of how to combine various perspectives (psychological, biological, social) on mental disorders into a coherent theory. In recent years, several frameworks were proposed tackling this problem (most notably, the Research Domain Criteria and enactive psychiatry). Typically, integrative frameworks for psychiatric disorders emphasize the need for explanatory pluralism and non-hierarchical theories. In this way, the proposed solutions for tackling the integration problem bear similarities to the project of neurophenomenology in consciousness science. Neurophenomenology refers to the systematic integration of first- (i.e., phenomenological) and third-person perspectives (i.e., neuroscientific) in order to address the hard problem of consciousness (i.e., the question of how conscious experience can arise from non-conscious matter).

I will present a neurophenomenological project that attempts to bridge the gap between the lived experience of emotional dysregulation (as a transdiagnostic marker of psychiatric disorder as well as one of the more problematic symptoms) and its associated neurocognitive networks (as a potential target for treatment with methods such as TMS). I will present two studies, focusing, respectively on a) qualitative phenomenology; and b) electrophysiology. With the use of qualitative phenomenological methods, I have identified the three central experiential components of emotion regulation: agency, temporality, and sensemaking. In the second step, with the use of electrophysiological methods, the behavioral and neural correlates of emotion regulation in health and depression were discovered.

In the concluding sections, I provide some methodological considerations regarding neurophenomenology of psychiatric disorders, specifically, how it can be integrated into high-precision, individualized psychiatry.

Title: On becoming aware that one is dreaming: A micro-phenomenological investigation of signal-verified lucid dream experiences

Authors: Ema Demšar, Mahdad Jafarzadeh Esfahani, Martin Dresler, Thomas Andrillon, Jennifer Windt

Abstract: Lucid dreaming (LD) is an experience in which one is aware that they are currently dreaming. LD has been reported for millennia across different cultural contexts, with some contemplative traditions even developing techniques to cultivate LD and to harness lucidity for engaging in contemplative practices while asleep. With the development of the eye-signaling protocol, where dreamers signal the onset of lucidity with intentional eye movements during REM sleep, the neurophysiology of LD can now be studied in laboratory settings. However, neuroscientific progress has been hampered by the underexplored phenomenology of LD. While it appears that lucidity is not an all-or-nothing phenomenon, but comes in different types and is often characterized by complex temporal dynamics, there has as of yet been no systematic investigation of the detailed phenomenology of LD. In my research, I use the micro-phenomenological interview method to investigate the structures and dynamics of LD experience.

In my talk, I will present findings from a neurophenomenological EEG study employing an LD induction protocol, where participants utilized eye signals to verify lucidity. Subsequent micro-phenomenological interviews (N=38, 60–90 minutes) investigated experiences associated with lucid insight and the onset and progression of lucidity. A qualitative micro-phenomenological analysis of a subset of 24 interviews that focused on signal-verified REM-sleep LD episodes identified a) distinct patterns of temporal progression leading to lucid insight and b) salient phenomenological structures characterizing phases of pre-lucidity and full dream lucidity. My talk will focus on the quality of attention and awareness in LD, meta-cognitive pre-lucidity processes, and experiences of embodiment, agency, and selfhood surrounding the onset of lucidity. I will conclude by considering the relevance of these findings for future research into dream lucidity, as well as discussing the application of micro-phenomenology and the neurophenomenological framework in dream and sleep research.

Title: Investigating the phenomenology of objectless sleep experiences

Author: Adriana Alcaraz Sánchez

Abstract: Indo-Tibetan philosophical traditions, including that of the Advaita Vedānta, as well as the Dzogchen in Indo-Tibetan Buddhism, posit the existence of a particular state of awareness during deep sleep, that of “sushupti” or “clear light sleep” (Padmasambhava & Gyatrul, 2008; Prasad, 2000, Ponlop, 2006). According to these traditions, such a state departs from any ordinary state of consciousness insofar as it is said to lack conscious content and merely involve a state of “consciousness-as-such”. However, despite the frequent mentions of this state in the key texts of these traditions, descriptions of sushupti and clear light sleep are highly heterogeneous, thus, questioning whether a single state is being described, or instead, several ones.

In this presentation, I will give an overview of my work on “objectless sleep experiences”, experiences that I describe as conscious states during sleep that appear to one as lacking a distinct object of awareness. To that aim, I will briefly present the results of two studies exploring the phenomenological blueprints of objectless sleep experiences, one of which uses a qualitative research tool inspired by the micro-phenomenological interview (Alcaraz-Sánchez et al., 2022; Alcaraz-Sánchez, Forthcoming). I will then show how such experiences of objectless awareness during sleep can be situated within other associated states across sleep and waking, including minimal forms of dreaming, deep meditative experiences, and experiences of absorption. I will end by introducing a multidimensional framework aimed at guiding research on elusive forms of altered states of consciousness, including clear light sleep and other minimal forms of awareness during sleep and waking (Alcaraz-Sánchez, Under Review). 

Title: Exploring the dynamic structures of lived experience through micro-phenomenology

Abstract: Micro-phenomenology starts from the observation that the naïve descriptions we give of our experience are usually poor and unreliable. However, this does not mean that our experience is out of reach. Accessing it requires a particular expertise, which consists in carrying out specific acts. Micro-phenomenology is a new discipline aiming at eliciting such acts in the context of an interview, making it possible to collect rigorous and very fine-grained descriptions of a given experience. However the method is not limited to only describing individual tokens of experience. It also includes an analysis method aiming at identifying generic characteristics which constitute the structure of the experience under study. This work of description and analysis highlights usually unrecognized micro-dynamics, where the structures we usually consider as essential to experience fade. The first part of my presentation will focus on the methodological and epistemological aspects of micro-phenomenology. In the second part, I will give an overview of these dynamic structures

Title: Accounting for the phenomenal structure of experience: space and time

Author: Matteo Grasso

Abstract: In this talk I present an outline of IIT’s account of phenomenal space and time. First, I will introduce IIT’s methodology to study consciousness scientifically, which consists in describing the structure of experience, and finding a correspondence with the cause-effect structure unfolded from its substrate. I will then present IIT’s account of two contents of experience: the extendedness of visual space and the flow of time.

Title: Everything Everywhere All at Once? Exploring the richness of visual experience beyond words

Author: Qianchen Liang

Abstract: Subjectively, our visual experience seems rich and detailed. However, previous studies implies our experience is limited to 3~4 items at a glance. These studies tended to focus on small, easy-to-label details in the stimuli, without exhausting all we could experience. Here, to probe richness of experience, we consider semantic details together with perceptual qualities that we have no words for. For each of 100 natural scene images, we synthesised an image with equated image summary statistics. In Experiment 1, 100 online participants described each image in 5 words. Using a large language model, we confirmed that 27 out of 100 pairs had indistinguishable semantic labels. Using these semantically-equated pairs, in (registered) Experiment 2, we will perform a perceptual task on online participants. We will also use an immersive display with in-lab participants, to test in situations mimicking the real-world environment. How rich can our experience be beyond verbal description?

Title: Input-Output Functionalism and Underdetermination in Consciousness Science

Author: Francesco Ellia

Abstract: Paraphrasing Forrest Gump, we might say that for the input-output functionalist, “conscious is as conscious does.” Essentially, input-output functionalism evaluates consciousness in other systems based on the functions displayed in those systems. This view is explicitly advocated by Cohen and Dennett (2011), who consider unscientific any theory that attempts to explain consciousness without exclusively relying on functions. It is also implicitly supported by Doerig and colleagues (2019), who deem theories that dissociate subjective experience from functions as either unscientific or false. While consciousness science is arguably at a pre-paradigmatic stage (Ellia and Negro, in preparation), this view is generally widespread in contemporary neuroscience of consciousness, with most theories adopting a functionalist perspective (Doerig et al. 2019).

However, input-output functionalism presents severe problems, from risking collapse into behaviorism (Tsuchiya et al. 2020), to leading to misguided considerations on falsificationism (Negro 2020; Ellia 2021), to committing the fallacy of misplaced objectivity (Ellia, Hendren et al. 2021). It is not a trivial question to ask about the capacity of input-output functionalism to discriminate between conscious and unconscious systems in a world where technological advancements have introduced artificial systems capable of mimicking complex cognitive functions. In this talk, I explore another shortcoming of input-output functionalism when applied to clinical cases beyond thought experiments (Ellia 2021), demonstrating how input-output functionalism is underdetermined—it cannot reliably distinguish between different cases of consciousness despite claiming it can. I argue that structural approaches are preferable instead (Ellia, in preparation).

Title: Towards blinded classification of loss of consciousness: distinguishing wakefulness from general anaesthesia and sleep in flies using a massive library of univariate time series analyses

Author: Angus Leung

Abstract: Studies frequently discover new time-series features as potential markers of consciousness. However, these features often lack principled, theoretical justifications as to why they should be related to consciousness. This raises the issues of 1) the particular time-series features being selected due to researcher biases towards specific types of analyses, while other potential features could have been explored but were not, and 2) how to interpret identified features to understand neural activity underlying consciousness, especially when features are identified from recordings which summate across large areas such as electroencephalography. To address the first concern, we take the approach of staying agnostic to particular analysis types and apply as many univariate time-series features as feasible, from various research fields (~7000) as potential consciousness markers. To address the second issue, we apply this approach to high-quality neural recordings from a relatively simple brain, the fly brain (Drosophila melanogaster) obtained during wakefulness, anaesthesia, and sleep. Using a registered report format and multiple independent sets of flies (N = 62) to evaluate the performance of each feature in distinguishing wakefulness from loss of consciousness, we find that features related to fractal geometry seem to generalise across datasets, while features related to well-known markers of consciousness, such as those related to spectral power or complexity, fail to generalise across datasets.

Title: Reverse Engineering Agentive Control in Biological Neural Systems

Author: Jingyue Xu

Abstract: Sensorimotor enactivism posits that perceptual experience is constituted by the dynamic coupling of the sensorimotor system and its environment, which comprises the system’s actions on the environment. When an agent is confronted with the Selection Problem of actions, implementing agentive control is the process of finding a solution to this problem. How do our brains endogenously generate voluntary action control over our bodies and mental faculties? This presentation brings forth a new theoretical perspective, interfacing neuroscience, control theory, and philosophy of mind, by taking a system-level approach to modeling the brain’s control of its internal processes. In a feedback control system, a controller is implemented to define a closed-loop map and subsequently a solution given its current state. The enactive brain can be similarly viewed as an online control system operating with a “pre-defined” controller design that at least yields local stability and robustness (namely, the agentive state does not dissolve in an arbitrarily near future). Therefore, to characterize the source of agentive control is to reverse engineer the internal controller design, which does not necessarily have a physical mapping, in the brain. To do this, I will deploy a data-driven version of System Level Synthesis (SLS), which is a novel system-level parameterization of linear systems in control theory. SLS is compatible with systems that are large-scale, localized, and endowed with internal communication delays, all of which are problems that theoretical neuroscientists are constantly faced with. Data-driven SLS offers a unique approach to learning the controller in the brain using physiological data as state/input signals without requiring an explicit identification of the system model of the brain. Given the near-ubiquitous layered architecture of the brain across the animal kingdom, one can approximate the source and structure of agency in various networks/sub-systems of neurons by appropriately constraining the agentive state of the subject, and thereafter characterize the performance of the brain controller in terms of its closed-loop behavior.

Title: Telic states: towards a computational phenomenology of purposeful behavior

Authors: Nadav Amir, Yael Niv, Angela Langdon

Abstract: Goals fundamentally shape how we experience the world. For example, when we are hungry, we view objects in our environment according to whether or not they are edible (or tasty). Alternatively, when we are cold, we view the very same objects according to their ability to produce heat. Computational theories of purposeful behavior, such as reinforcement learning, assume that agents employ fixed and pre-defined state representations to capture behaviorally-relevant features of their environment. Here, we propose an alternative view in which state-representations are not assumed veridical, or even pre-defined, but rather emerge from agents' goals through interaction with their environment. We introduce a notion of goal-directed, or telic, states – defined as goal-equivalent experience distributions – and use it to develop, from first principles, a parsimonious framework of purposeful behavior, driven by rich and complex features of experience, beyond the maximization of extrinsically defined reward functions. By coupling states with goals, this framework provides a fresh perspective on recent findings in cognitive neuroscience indicating that neural activity previously thought to encode prescriptive (or normative) aspects of experience also conveys information about descriptive (or positive) ones, and vice versa. It also provides a novel approach to the problem of goal formation as a tradeoff between the granularity and the controllability of telic state representations. Taken together, this framework provides a step towards a computational phenomenology of state-representation learning through the lens of goals. 

Title: The sense of agency from active causal inference

Author: Acer Chang

Abstract: This study investigates the active component of the sense of agency (SoA), positing that SoA is fundamentally an outcome of active causal inference regarding one's own actions and their impact on the environment. Participants controlled visual objects via a computer mouse, with tasks designed to test their ability to judge control or detect controlled objects under varying noise conditions. Our findings reveal that participants formed high-level, low-dimensional action plans that were idiosyncratic across but consistent within individuals to infer their degree of control. Employing transformer-LSTM-based autoencoders, we captured these action plans and demonstrated that the geometrical and dynamical properties of these action plans could predict behavioural profiles in the tasks with remarkable accuracy. This suggests that participants' sense of control is shaped by actively altering action plans, viewed as generating causal evidence through intervention. Further, participants proactively expanded the diversity of their action plans, facilitating the exploration of available action plan options while accumulating causal evidence for the inference process. Contrarily, patients with schizophrenia exhibited reduced action plan diversity, suggesting impaired active control inference and detection of self-relevant cues. These findings offer a more comprehensive understanding of the sense of agency, deeply rooted in the process of active causal inference. 

Title: Cleaning up the Perceptual Spill: Phenomenal overflow in the context of predictive coding models of cortical function

Author: Krzysztof DOLEGA

Abstract: Ned Block’s (2008) ‘overflow’ hypothesis about perceptual experience is the claim that phenomenal consciousness is richer than the processing capacities of cognitive functions. According to this view, subjects enjoy a detailed phenomenal experience of the whole visual field, but are unable to reliably access all the perceptual information presented in consciousness. In this paper, I present a new argument against the overflow hypothesis based on structural features of the human visual system as well as state-of-the-art predictive coding algorithms used for modeling perceptual information processing. I begin by discussing the morphology of the early visual system, where some form of signal compression is already necessitated at the ganglionic layer of the retina (Srinivasan et al., 1982). While this does not rule out the possibility of overflow taking place further downstream, it speaks against the idea that subjects could experience the full array of the available visual information at any point in time. Furthermore, hierarchical predictive coding models of cortical function (Rao & Ballard, 1999; Spratling, 2008, 2017) show that the kind of compression present in very early sensory processing can be found across the perceptual hierarchy. This speaks against the possibility of overflow taking place at later stages of processing for two reasons. Firstly, the compression involved in predictive coding is realized by attenuating bottom-up signals congruent with context-dependent expectations in a way that does not allow for a later recovery of the previously removed components of the sensory signal. Secondly, under predictive coding, it is the context-dependent expectations and not the sensory signals that serve as the vehicles for the content of perceptual experience. Thus, predictive coding models overturn the traditional view of bottom-up perceptual processing, leaving no room for phenomenal overflow to take place. I conclude by considering some possible objections.