Access Consciousness: Information Availability

Core Idea

Access consciousness involves information being globally available for reasoning, report, and behavioral control. A mental state is access-conscious if its content can be used by reasoning and behavior-producing systems, distinct from phenomenal consciousness which concerns subjective feel.

Explainer

You already know that consciousness in general raises the question of whether there is something it is like to be in a mental state — the puzzle of subjective experience. And if you've encountered the phenomenal/access distinction, you know that Ned Block drew a line between two different things philosophers might mean by "conscious." Access consciousness is one side of that line, and it has a precise, functional definition: a mental state is access-conscious if its content is globally available — broadcast widely enough in the cognitive system to be used by reasoning, verbal report, and the voluntary guidance of behavior.

Think of the mind as having many specialized processors running in parallel: perceptual systems, memory, motor planning, language production, and executive reasoning. Most of what these systems compute never becomes available to the whole system — peripheral processing remains local and isolated. A mental state achieves access consciousness when it enters a global workspace: a broadcast mechanism that makes its content available across multiple processors simultaneously. You can talk about it, reason with it, use it to guide voluntary action. The Global Workspace Theory of consciousness, associated with Bernard Baars, is essentially an account of access consciousness — it proposes that the brain has an architecture in which certain representations are globally broadcast while others remain local and inaccessible.

What makes this definition philosophically significant is the contrast with phenomenal consciousness — the felt quality of experience, the redness of red, the ache of pain. A mental state can in principle have phenomenal character without being access-conscious, and perhaps vice versa. The case of blindsight is the most discussed empirical support for this dissociation: patients with damage to primary visual cortex deny seeing anything in part of their visual field (no phenomenal experience), yet can correctly guess properties of stimuli presented there above chance (information is somehow accessible). Their performance suggests that some form of information processing occurs without phenomenal awareness, though how cleanly this maps onto access consciousness is debated.

The distinction matters most for the hard problem. The hard problem asks why physical processes give rise to subjective experience — why there is *something it is like* to be in a particular state. This question targets phenomenal consciousness. Access consciousness, by contrast, is relatively tractable: we have plausible functional and neural accounts of how information gets globally broadcast and becomes available for reasoning and report. When someone asks whether a robot or AI system is conscious, the philosophically interesting question is almost always about phenomenal consciousness — whether there is something it is like to be the system — not about access consciousness, which any sufficiently sophisticated information-processing system might achieve. Keeping these concepts clearly separated is essential for making precise claims about minds, machines, and the gaps in our understanding of either.