Modelling aspects of consciousness: a topological perspective
arXiv Preprint Archive – November 10, 2020
Source: arXiv
Summary
The brain's inability to fully grasp its own consciousness may be mathematically inevitable. Using topological modeling in neuroscognitive biology (q-bio.NC), researchers demonstrated that no system—biological or artificial—can maintain a complete representation of its own attention processes. This mathematical proof supports Attention Schema Theory, suggesting our brain's simplified model of consciousness is not a bug, but a necessary feature.
Abstract
Attention Schema Theory (AST) is a recent proposal to provide a scientific explanation for the basis of subjective awareness. In AST, the brain constructs a representation of attention taking place in its own (and others') mind (`the attention schema'). Moreover, this representation is incomplete for efficiency reasons. This inherent incompleteness of the attention schema results in the inability of humans to understand how their own subjective awareness arises (related to the so-called `hard problem' of consciousness). Given this theory, the present paper asks whether a mind (either human or machine-based) that incorporates attention, and that contains a representation of its own attention, can ever have a complete representation. Using a simple yet general model and a mathematical argument based on classical topology, we show that a complete representation of attention is not possible, since it cannot faithfully represent streams of attention. In this way, the study supports one of the core aspects of AST, that the brain's representation of its own attention is necessarily incomplete.