-
This document argues that, from the perspective of cognitive science and artificial intelligence, the way we view “concepts” and “categories” has shifted from fixed “entities” to dynamic “processes.” In particular, it challenges the traditional view of concepts and knowledge as having sensory independence and certain stable structures and patterns, questioning this from a new perspective.
-
According to the author, concepts are not independent of the senses, but are inseparable from the sensory level, and he proposes that they do not exist as stable entities, but are created and renewed in a process. From this perspective, they argue that concepts are organized in the process of interaction with situations, and knowledge is fragmented and dispersed across different sensory modalities.
-
To concretize this new perspective, the author introduces the ideas of Barsalou’s theory of perceptual symbol systems and symbol emergent robotics. Through these theories, the author attempts to clarify the process-based view of knowledge, in which knowledge and concepts are generated through the interaction of fragmented and distributed groups of representations.
-
This section describes Barsalou’s perceptual symbol system theory. This theory, proposed by cognitive psychologist Lawrence Barsalou, argues that concepts and categories in the mind are represented not as abstract symbols or linguistic labels, but as sensory and motor reactivations of experience. In other words, when we think of concepts such as “dog” or “run,” Barsalou’s theory is that they exist not merely as words or abstract concepts, but as simulated experiences in the mind that reproduce the experience of seeing a dog or running. This is also related to “embodied cognition,” which embodies the idea that cognition is not abstract information processing, but is deeply connected to our physical and sensory experiences.
-
Next, we will discuss symbol emergent robotics. This is an area of research on robots that communicate socially like humans, and studies the ability of robots to spontaneously generate symbols and language. The term “emergence” here refers to the ability of a robot to self-organize and spontaneously create new symbols and means of communication in order to adapt to a new environment. Language emergence is a central feature of human cognition and sociality, and by reproducing it in robots, we aim to gain new insights into the origins of human cognition and language.
-
These two theories provide answers to the question of how human cognition forms concepts and how this leads to social communication. They offer an approach that seeks to understand the depths of human cognition and communication by linking the perspective that human cognition is deeply connected to sensory and physical experience (Barsalou’s theory of perceptual symbol systems) and how that experience leads to social communication (symbol emergent robotics). The approach is based on a deep understanding of human cognition and communication.
relevance
This page is auto-translated from /nishio/実体ベースの概念からプロセスベースの概念へ using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.