← Citations

neuroscience

Global Workspace & The Bandwidth of Consciousness

Why the conscious channel is so much narrower than the brain

Tor Nørretranders' 1991 book The User Illusion popularized the now-cited bandwidth figures (about 11 million bits/sec arriving at the senses; ~40 bits/sec reportable). The numbers are necessarily approximations — channel capacity depends on how you cut it — but the order-of-magnitude gap is real and is the central architectural fact of conscious access. The question is not whether information is being thrown away; it is how, and by what.

Bernard Baars' Global Workspace Theory (1988) proposed that consciousness is the result of brief, brain-wide broadcasting of selected content from a 'workspace' that competes for limited capacity. Stanislas Dehaene extended this with the Global Neuronal Workspace model (Dehaene & Naccache 2001; Dehaene 2014), tying it to specific frontoparietal networks and identifying a neural signature: a late (~300ms) burst of widely-distributed, recurrent activity that distinguishes conscious from unconscious processing. Subliminally-presented stimuli evoke early sensory responses but fail to ignite the workspace; consciously-accessed stimuli show the late, distributed signature.

The competing major theory is Giulio Tononi's Integrated Information Theory (IIT), which proposes that consciousness is identical to a system's integrated information (Φ) — the irreducibility of its causal structure. IIT and Global Workspace make different predictions about, for instance, whether the cerebellum (massively interconnected but functionally segregated) is conscious. The two camps have been engaged in a series of adversarial collaborations (Melloni et al., currently ongoing) that are producing some of the cleanest comparative data the field has ever had.

There is also genuine work showing the conscious 'bandwidth' is not a single number. Block's distinction between phenomenal and access consciousness suggests there may be more lived experience than is reportable. Sperling's classic partial-report paradigm hinted at this: subjects appear to briefly 'see' more letters than they can report. Whether this constitutes additional conscious content or only richer iconic memory remains contested.

For practice: regardless of which theory wins, the architectural fact — most of what the brain does does not reach conscious access — is settled. The contemplative claim that 'you are not your thoughts' has a straightforward neural reading: the thoughts you have access to are the small subset that won the broadcast competition, not the substrate that produced them.

Primary sources