Search Abstracts | Symposia | Slide Sessions | Poster Sessions
A neuroscientific model of the entropy of qualia for semantics and consciousness
Poster Session B, Friday, October 25, 10:00 - 11:30 am, Great Hall 3 and 4
T R Williamson1,2,3,4, Agustín Ibáñez5,6; 1University of the West of England, Bristol, 2Southmead Hospital, North Bristol NHS Trust, 3University of Oxford, 4University of Southern California, 5Latin American Institute for Brain Health, Universidad Adolfo Ibanez, Santiago, Chile, 6Global Brain Health Institute, Trinity College Dublin
Studies of meaning in language and conscious experience are equally ancient, although both have been determined in the past to be too ambiguous or irrelevant for scientific investigation (Bloomfield, 1933; Chomsky & Ronat, 1979; Skinner, 1953; Watson, 1924). This may stem from the impression that they both share a kind of mysterious, opaque, and indeterminate nature (Nagel, 1974; Wittgenstein, 1953) that entails recourse to objective, ontological explanations aiming at characterising what subjective, immaterial experience fundamentally is (Crick & Koch, 1990; Lakoff & Johnson, 1980). We argue that using subjective experiences as the foundation for cognitive and neuroscientific explanation can frequently lead to category mistakes, where subjective properties (in the explanans) are mistakenly taken as true representations of a phenomenon (the explanandum). Our observation is that this mistake arises because semantics and consciousness are rooted, to a certain extent, in the (neural) underpinnings of subjective experience (i.e., qualia). Impossible to encapsulate fully, we propose a novel analytic-synthetic approach that introduces entropy to neuroscientifically explore semantics and consciousness as a first step. While work especially in the last year has explored the neurobiology of consciousness entropically (Carhart-Harris, 2018; Froese, 2024; Ji et al., 2024; Kringelbach et al., 2024), our approach is the first to apply entropy to a neuroscientific understanding of linguistic semantics. We start with defining neuropsychological states encapsulating processes for semantics (production, comprehension, planning, prediction, etc.) and consciousness (sensory perception, inference, motor action, etc.). Then we define how elements of discrete neurobiological representations or substrates that might compose those processes may be modelled entropically. The (information-theoretic) entropy of a given neuropsychological state may be determined by the probability of a given microscopic (in the statistical mechanics sense of entropy) element composing it. We conceive of specific semantic- or conscious-states as vector, containing finite sets of 0s and 1s corresponding to whether a specific neurobiologically determined, constitutive, microscopic element (for picking a red flower in a field, e.g., semantic-state: activation of visual sensorimotor properties of redness, semantic composition of linguistic items; conscious-state: activation of world knowledge of social appropriateness to pick a flower in a meadow vs. a private garden, volitional action initiating motor coordination of the hand or arm muscles, etc.) is involved in producing the qualitative experience associated with that state. Overarchingly, our proposals aim to highlight the synergy between meaning in language and consciousness that qualia facilitates and to propose a new avenue for studying this aspect of their natures. We conclude that this conceptualisation may enable new kinds of scientific questions, from investigations into what these neuropsychological elements might be to how measurable macroscopic constraints like brain health (Ibanez et al., 2024) may eventually be factors within an information-theoretic model to ask radically cross-disciplinary neurobiological questions.
Topic Areas: Multisensory or Sensorimotor Integration, Meaning: Lexical Semantics