Global Workspace Theory

Global Workspace Theory
Name	Global Workspace Theory
Field	Cognitive science, Neuroscience
Year	1988
Authors	Bernard Baars
Related	Cognitive architecture, Working memory, Attention

Global Workspace Theory. It is a cognitive architecture and theory of consciousness developed by Bernard Baars. The model proposes that conscious awareness arises from a fleeting integrative memory capacity, analogous to a global workspace in a computational system, which broadcasts information to a multitude of specialized, unconscious processes.

Overview

The central metaphor of the theory is a "theater of consciousness," where a bright spot on a stage, illuminated by the spotlight of attention, represents the contents of conscious experience. This stage corresponds to a limited-capacity global workspace, while the audience in the dark represents a vast array of unconscious neural networks and specialized processors. Key collaborators and developers of related models include Stanislas Dehaene, Jean-Pierre Changeux, and the late Bernard J. Baars. The framework has been influential in projects like the LIDA model developed by researchers including Stan Franklin.

Theoretical foundations

The theory integrates concepts from cognitive psychology, computer science, and neuroscience. It draws heavily on earlier work on working memory by Alan Baddeley and the notion of a central executive. Architecturally, it posits that the brain's thalamocortical system, particularly the frontal and parietal regions, implements the global workspace. This architecture allows for the integration of information from specialized modules, such as those for vision processed in the occipital lobe or language in Broca's and Wernicke's areas.

Empirical evidence

Substantial support comes from cognitive neuroscience experiments using techniques like fMRI and EEG. The neural correlates of conscious access, as described by Stanislas Dehaene and Jean-Pierre Changeux in their Global Neuronal Workspace hypothesis, show widespread cortical activation when a stimulus becomes conscious, contrasting with weaker, localized activity for unconscious processing. Key experiments include studies on binocular rivalry, the attentional blink, and visual masking conducted at institutions like the French National Institute of Health and Medical Research and the Collège de France.

Relation to consciousness

The theory specifically addresses phenomenal consciousness and access consciousness, providing a mechanistic account for how some mental contents become reportable and capable of guiding a wide range of behaviors. It contrasts with theories that posit consciousness as a fundamental property, such as those suggested by David Chalmers, or those focusing on higher-order thought advocated by David Rosenthal. The framework is also compatible with aspects of Integrated Information Theory proposed by Giulio Tononi, though they differ in fundamental approach.

Criticisms and alternatives

Critics argue the theory may be more descriptive than explanatory, a "homunculus" theory that does not fully solve the hard problem of consciousness. Alternative frameworks include panpsychism, Orchestrated Objective Reduction by Roger Penrose and Stuart Hameroff, and predictive processing models associated with Karl Friston. Some, like Daniel Dennett with his Multiple Drafts model, reject the need for a central workspace altogether, favoring a more decentralized account of mental activity.

Applications

The architecture has inspired developments in artificial intelligence, particularly in creating cognitive architectures for autonomous agents, such as the LIDA model. In clinical neuroscience, it informs research into disorders of consciousness like the vegetative state and coma, with applications in diagnosis and neurorehabilitation. The theory also provides a framework for understanding anesthetic mechanisms and the loss of consciousness under agents like propofol.

Category:Cognitive science Category:Consciousness Category:Neuroscience