Neuropsychology and Explainability of AI: A Distributional Approach to the Relationship Between Activation Similarity of Neural Categories in Synthetic Cognition
This work addresses the explainability of AI for researchers and practitioners by offering a novel bridge between human psychology and neural networks, though it appears incremental as it builds on existing concepts without demonstrating broad empirical gains.
The paper tackles the problem of explaining artificial neural networks by proposing a neuropsychological approach that uses human cognitive concepts like categorization and similarity to align synthetic explanatory frameworks with human thought. The result is a proposed process of synthetic cognition involving categorical convergence of highly activated tokens, explained as a superposition of categorical sub-dimensions in the input vector space.
We propose a neuropsychological approach to the explainability of artificial neural networks, which involves using concepts from human cognitive psychology as relevant heuristic references for developing synthetic explanatory frameworks that align with human modes of thought. The analogical concepts mobilized here, which are intended to create such an epistemological bridge, are those of categorization and similarity, as these notions are particularly suited to the categorical "nature" of the reconstructive information processing performed by artificial neural networks. Our study aims to reveal a unique process of synthetic cognition, that of the categorical convergence of highly activated tokens. We attempt to explain this process with the idea that the categorical segment created by a neuron is actually the result of a superposition of categorical sub-dimensions within its input vector space.