Global Workspace Theory (GWT) is a significant cognitive architecture and theoretical framework that aims to explain consciousness. Introduced in 1988 by cognitive scientist Bernard Baars, GWT was developed to provide a qualitative explanation for a wide range of conscious and unconscious processes. It has become influential in modeling how consciousness and higher-order cognition emerge from the competition and integrated flow of information across
numerous parallel neural processes within the brain.
Baars drew inspiration for GWT from the "blackboard system" concept found in early artificial intelligence system architectures. In these AI systems, independent programs would share information through a central "blackboard." GWT applies this idea to the brain, proposing a central hub for information exchange. This theory stands as one of the leading models attempting to demystify the complex phenomenon of consciousness.
The Brain as a Parallel Processor
At its core, GWT posits that the brain is composed of many specialized processes or modules that operate simultaneously, with a significant portion of this activity remaining unconscious. The global workspace itself is not a physical location but rather a functional hub. Its primary role is to broadcast and integrate information, allowing it to be disseminated across these various modules. This functionalist approach classifies GWT as a theory of consciousness that focuses on the function of mental states rather than their physical realization.
When sensory input, memories, or internal representations capture attention, they gain entry into this global workspace. Once there, they become accessible to a multitude of cognitive processes. Elements within the brain are constantly competing for this attention, and those that succeed in gaining access to the global workspace have their information distributed and coordinated throughout the entire cognitive system. This mechanism facilitates top-down control over functions like attention, working memory, planning, and problem-solving through this shared information.
The Fleeting Nature of Conscious Experience
GWT suggests that the global workspace involves a fleeting memory, lasting only a few seconds. This is considerably shorter than the 10-30 seconds typically associated with classical working memory. The contents of the global workspace are proposed to correspond to what we are consciously aware of, and this information is then broadcast to numerous unconscious cognitive brain processes, referred to as "receiving processes."
Other unconscious processes, operating in parallel with limited communication among themselves, can form coalitions that act as "input processes" to the global workspace. Since messages broadcast globally can trigger actions in receiving processes throughout the brain, the global workspace may be instrumental in exercising executive control to perform voluntary actions. Both individual processes and allied groups of processes vie for access to the global workspace, aiming to disseminate their messages to other processes to recruit more support and increase their chances of achieving their goals.
Beyond the Spotlight: Contextual Systems and Adaptability
Incoming stimuli require temporary storage to compete for attention and conscious access. Research suggests the existence of a sensory memory buffer that maintains stimuli for a few hundred milliseconds, with meaningful information extraction severely impaired after 300 milliseconds and most data lost after 700 milliseconds. Baars emphasizes that working memory is closely linked with conscious experience, though not identical to it. Conscious events might involve additional necessary conditions, such as interaction with a "self" system and an executive interpreter in the brain, as suggested by various researchers.
Despite these nuances, GWT effectively models several characteristics of consciousness, including its role in handling novel situations, its limited capacity, its sequential nature, and its ability to trigger a vast array of unconscious brain processes. The theory also lends itself well to computational modeling, with implementations like Stan Franklin's IDA model. GWT also accounts for "behind the scenes" contextual systems, such as the dorsal cortical stream of the visual system, which shape conscious content without ever becoming conscious themselves. This architectural perspective leads to specific neural hypotheses, such as the idea that sensory events in different modalities may compete for consciousness if their contents are incompatible, like out-of-sync audio and video tracks in a movie. The theory of practopoiesis offers an alternative view, suggesting that the global workspace is primarily achieved through the rapid adaptive mechanisms of nerve cells, where neurons learn when and how to adapt to their sensory context, rather than relying heavily on fixed connectivity.













