Cognitive architecture is the theory about the structures of the human mind and how they work together to manage intelligent behavior in any complex environment. A goal of cognitive architecture is to use the research of cognitive psychology to create a complete computer model of cognition.
As a blueprint for intelligent agents, cognitive architecture aims to have artificial computational system processes that act like natural cognitive systems, or humans. The theory focuses on combining cognitive science and artificial intelligence (AI).
In 1960, Ed Feigenbaum created one of the first possible models for cognitive architecture, Elementary Perceiver and Memorizer (EPAM), and attempted to understand from that model how a number of fundamental aspects of the human mind work. Herbert A. Simon, Feigenbaum's teacher and a founder in the field of AI, recognized that his student's work had provided a possible model for a cognitive architecture.
Successful cognitive architecture models include:
- Active Control of Thought-Rational (ACT-R) was developed at Carnegie Mellon University by John Robert Anderson. ACT-R attempts to study how the brain organizes itself into singular processing modules to reduce cognitive functions to the most basic operations that can enable cognition.
- Learning Intelligent Distribution Agent (LIDA) was created by Stan Franklin and colleagues at the University of Memphis. The architecture was created as an integrated model that attempts to broadly model human cognition, from perception and action to high-level reasoning.
- Soar was developed by John Laird, Allen Newell and Paul Rosenbloom at Carnegie Mellon University and drew from ACT-R and LIDA. The model's objective was to develop generalized intelligent agents for a range of tasks that also functioned as the building blocks to emulate human cognitive capacity.
With the growing popularity of AI and machine learning technologies, research into cognitive architecture will only further refine the theory and its current and potential uses.