An Integrated Self-Aware Cognitive Architecture

Abstract

The result of our BICA Phase I research effort is the specification of a computational cognitive architecture that is capable of human-like cognition, learning and social behavior in a wide range of real-world situations and paradigms set in virtual environments. Our architecture focuses on the higher-level cognitive processes involved in human cognition and is designed to be integrated into a larger end-to-end architecture that includes lower level sensing and action. The main feature of our architecture is the notion of self aware cognition that we believe is necessary for human-like cognitive growth. Our approach is inspired by studies of the human brain-mind: in particular, by theoretical models of representations of agency in the higher associative human brain areas. This feature (a theory of mind including representations of one's self) allows the system to maintain human-like attention, focus on the most relevant features and aspects of a situation, and come up with ideas and initiatives that may not follow from formal logic. The result is a robust cognitive system capable of significant cognitive growth.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 2008
Accession Number
ADA480166

Entities

People

  • Alexei V. Samsonovich
  • Giorgio A. Ascoli
  • Kenneth A. Dejong

Organizations

  • George Mason University

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Air Force Research Laboratories
  • Application Software
  • Artificial Intelligence
  • Cognition
  • Cognitive Science
  • Computational Science
  • Environment
  • Human Behavior
  • Information Processing
  • Language
  • Psychological Theory
  • Psychology
  • Self Organizing Systems
  • Simulations
  • Specifications
  • Standards
  • Virtual Reality

Fields of Study

  • Computer science

Readers

  • Artificial Intelligence
  • Neural Network Machine Learning.