Cognition Without Neural Representation: Dynamics of a Complex System

Abstract

This paper proposes an account of neurocognitive activity without leveraging the notion of neural representation. Neural representation is a concept that results from assuming that the properties of the models used in computational cognitive neuroscience (e.g., information, representation, etc.) must literally exist the system being modelled (e.g., the brain). Computational models are important tools to test a theory about how the collected data (e.g., behavioural or neuroimaging) has been generated. While the usefulness of computational models is unquestionable, it does not follow that neurocognitive activity should literally entail the properties construed in the model (e.g., information, representation). While this is an assumption present in computationalist accounts, it is not held across the board in neuroscience. In the last section, the paper offers a dynamical account of neurocognitive activity with Dynamical Causal Modelling (DCM) that combines dynamical systems theory (DST) mathematical formalisms with the theoretical contextualisation provided by Embodied and Enactive Cognitive Science (EECS).

Rejecting Cognitivism: Computational Phenomenology for Deep Learning

“…Additionally the cognitivist-interpreted ANN becomes a model of the brain; in other words, the brain is thought to learn Turing-like mental representations of the world. Finally, NR takes cognitivism's representationalism to its extreme by making a claim about our conscious experience: we experience a brain generated model of these entities (Churchland and Sejnowski, 1990;Mrowca et al, 2018;Sitzmann et al, 2020); for criticism see (Zahavi, 2018;Hipólito 2022). Frith puts it in slogan form: "my perception is not of the world, but of my brain's model of the world" (Frith, 2007).…”

“…14 Following this second interpretation, a CP-informed neuroscience would reject the notion of neural representations, following neuroscientists that find it misleading (Freeman and Skarda, 1990;Brette, 2019), having identified it as the well-known map-territory fallacy (confusing models of reality with reality itself) (Korzybski, 1933). It might find common grounds with anti-representationalist theories such as the dynamical systems approach (Van Gelder, 1995;Freeman, 2000;Favela, 2021;Hipólito, 2022) or instrumentalist accounts of the Free Energy Principle (Friston 2021;Andews, 2021;van Es, 2021;Parr Pezzulo and Friston, 2022).…”

May the 4C's be with you: an overview of complexity-inspired frameworks for analysing resting-state neuroimaging data

“…Experimental evidence accumulated in recent years has revealed that variation in structure and function in the human brain (both for normal development and for disease) tends to be widely distributed, and hence there is often no simple one-to-one mapping between cognitive functions and individual brain regions [44][45][46][47]. An important paradigm shift in neuroimaging research was therefore to go beyond mere activation and focus instead on co-activation, i.e.…”

May the 4c’s Be with You: An Overview of Complexity-Inspired Frameworks for Analyzing Resting-State Neuroimaging Data

“…Experimental evidence accumulated in recent years has revealed that variation in structure and function in the human brain (both for normal development and for disease) tends to be widely distributed, and hence there is often no simple one-to-one mapping between cognitive functions and individual brain regions (Ecker et al, 2010;Hipólito, 2022;Thompson et al, 2001;Zhuo et al, 2019). An important paradigm shift in neuroimaging research was therefore to go beyond mere activation and focus instead on co-activation, i.e., on the interdependency between the activity of multiple regions.…”