A society of mind approach to cognition and metacognition in a cognitive architecture

Venkatamuni, Vijayakumar Maragal

Abstract

This thesis investigates the concept of mind as a control system using the "Society of Agents" metaphor. "Society of Agents" describes collective behaviours of simple and intelligent agents. "Society of Mind" is more than a collection of task-oriented and deliberative agents; it is a powerful concept for mind research and can benefit from the use of metacognition. The aim is to develop a self configurable computational model using the concept of metacognition. A six tiered SMCA (Society of Mind Cognitive Architecture) control model is designed that relies on a society of agents operating using metrics associated with the principles of artificial economics in animal cognition. This research investigates the concept of metacognition as a powerful catalyst for control, unify and self-reflection. Metacognition is used on BDI models with respect to planning, reasoning, decision making, self reflection, problem solving, learning and the general process of cognition to improve performance.

One perspective on how to develop metacognition in a SMCA model is based on the differentiation between metacognitive strategies and metacomponents or metacognitive aids. Metacognitive strategies denote activities such as metacomphrension (remedial action) and metamanagement (self management) and schema training (meaning full learning over cognitive structures). Metacomponents are aids for the representation of thoughts. To develop an efficient, intelligent and optimal agent through the use of metacognition requires the design of a multiple layered control model which includes simple to complex levels of agent action and behaviours. This SMCA model has designed and implemented for six layers which includes reflexive, reactive, deliberative (BDI), learning (Q-Ieamer), metacontrol and metacognition layers.

Citation

Venkatamuni, V. M. (2008). A society of mind approach to cognition and metacognition in a cognitive architecture. (Thesis). University of Hull. Retrieved from https://hull-repository.worktribe.com/output/4213883

• Thesis Type: Thesis
• Deposit Date: Mar 8, 2013
• Publicly Available Date: Feb 22, 2023
• Keywords: Computer science
• Public URL: https://hull-repository.worktribe.com/output/4213883
• Additional Information: Department of Computer Science, The University of Hull
• Award Date: Apr 1, 2008