Neurosymbolic Systems of Perception and Cognition: The Role of Attention

Abstract

A cognitive architecture aimed at cumulative learning must provide the necessary information and control structures to allow agents to learn incrementally and autonomously from their experience. This involves managing an agent's goals as well as continuously relating sensory information to these in its perception-cognition information processing stack. The more varied the environment of a learning agent is, the more general and flexible must be these mechanisms to handle a wider variety of relevant patterns, tasks, and goal structures. While many researchers agree that information at different levels of abstraction likely differs in its makeup and structure and processing mechanisms, agreement on the particulars of such differences is not generally shared in the research community. A dual processing architecture (often referred to as System-1 and System-2) has been proposed as a model of cognitive processing, and they are often considered as responsible for low- and high-level information, respectively. We posit that cognition is not binary in this way and that knowledge at any level of abstraction involves what we refer to as neurosymbolic information, meaning that data at both high and low levels must contain both symbolic and subsymbolic information. Further, we argue that the main differentiating factor between the processing of high and low levels of data abstraction can be largely attributed to the nature of the involved attention mechanisms. We describe the key arguments behind this view and review relevant evidence from the literature.

Toward Visual Syntactical Understanding

“…System 1 performs fast, automatic pattern recognition, while system 2 executes slower, step-by-step deliberation. In this regard, DL is considered as a system 1 process and rule-based approaches are part of system 2 [50]. Aligning with this theme, our proposed approach can also be viewed from a neuro-symbolic lens as depicted in Fig.…”

Embodied Experience in Socially Participatory Artificial Intelligence

“…However, subsymbolic or statistical representations are closer analogues to human perception and automaticity, and its “body” can be more distributed and porous than what is needed for symbol processing embodiment, which attempts to emulate human deliberative processing (Kahneman 2013; Latapie et al. 2022). One can think of symbols as logical variables that can refer to any object in the world.…”

“…Both symbolic and subsymbolic processing require embodiment in order to engage the world (Brooks et al 1998;Smith 2019;Gill 2019;Cruz 2019;Lumbreras 2023;Herzfeld 2023), as does human cognition. However, subsymbolic or statistical representations are closer analogues to human perception and automaticity, and its "body" can be more distributed and porous than what is needed for symbol processing embodiment, which attempts to emulate human deliberative processing (Kahneman 2013;Latapie et al 2022). One can think of symbols as logical variables that can refer to any object in the world.…”