Consciousness Might Be a Broadcast Signal

The Hardest Problem Anyone Has Ever Tried to Solve

Somewhere in the three pounds of biological tissue between your ears, something extraordinary is happening. Billions of neurons are firing, exchanging chemicals, generating electrical fields. And somehow, out of that electrochemical storm, there is something it is like to be you.

You're not just processing information. You're experiencing it. The redness of red. The taste of coffee. The felt sense of existing as a person reading this sentence. This is consciousness. And explaining how it arises from physical matter is what philosopher David Chalmers famously called "the hard problem."

Most scientists, for most of history, refused to touch it. Consciousness was considered too squishy, too philosophical, too unmeasurable for serious science. You couldn't put awareness under a microscope. You couldn't run a controlled experiment on what it feels like to see blue.

Then, in 1988, a cognitive scientist named Bernard Baars proposed something radical. He didn't try to solve the hard problem. Instead, he asked a more tractable question: What is the brain doing differently when information is conscious versus when it isn't? And he came up with an answer so elegant that it has shaped consciousness research for nearly four decades.

He called it global workspace theory. And it starts with a theater.

The Theater of Your Mind (It's More Literal Than You Think)

Baars's metaphor goes like this.

Imagine a vast, dark theater. On the stage, there's a single bright spotlight. In the audience sit dozens of specialists: the memory system, the language processor, the visual cortex, the emotional centers, the motor planning circuits. All of them are watching the stage.

Backstage, in the dark, many performers are active simultaneously. They're processing information, running computations, doing their jobs. But they're doing it unconsciously. You're not aware of any of it. The visual system is computing edge detection. The auditory system is parsing sound frequencies. The motor system is maintaining your balance. All happening, all invisible to you.

Now one of those performers steps into the spotlight. Its information gets broadcast, from the stage, to the entire audience. Every specialist in the theater receives the signal simultaneously. That moment of broadcasting, that transition from backstage darkness to the illuminated stage, is what we experience as consciousness.

The spotlight is narrow. Only one thing (or a very small number of things) can occupy it at a time. But when something does make it to the stage, it becomes available to every system in the brain at once. That's why conscious experience feels unified and integrated even though it's generated by dozens of separate processing systems. The global workspace is the common channel through which all those systems can share information.

And the thing that makes this more than just a metaphor? Baars's theory made testable predictions. It said there should be a measurable difference in brain activity between information that reaches consciousness and information that doesn't. It predicted something like a neural "ignition" event when unconscious processing transitions to conscious awareness.

Those predictions turned out to be right.

Dehaene's Ignition: What the Brain Actually Does When You Become Aware of Something

In the late 1990s and early 2000s, French neuroscientist Stanislas Dehaene took Baars's theater metaphor and turned it into a neurobiological theory with specific predictions about brain circuits, timing, and measurable signals.

Dehaene's version, called global neuronal workspace theory (GNWT), proposes that the workspace is physically implemented by a network of neurons with long-range connections, primarily in the prefrontal and parietal cortices. These neurons have axons that stretch across the brain, connecting distant regions. They are the broadcasting infrastructure.

Here's how it works. When a stimulus arrives (say, an image flashing on a screen), it first activates local, specialized brain regions. The visual cortex starts processing the image. This initial processing is fast, automatic, and unconscious. The image is being analyzed, but you don't know it yet.

If the stimulus is strong enough, salient enough, or relevant enough to current goals, something remarkable happens. The local activation crosses a threshold and triggers a sudden, explosive burst of activity in the frontoparietal workspace neurons. Dehaene calls this ignition.

During ignition, the information that was being processed locally by the visual cortex gets amplified and broadcast throughout the brain via those long-range connections. Prefrontal regions light up. Parietal regions light up. The signal reverberates between distant brain areas in a sustained, self-reinforcing loop. And at that exact moment, the stimulus becomes conscious. You see it. You're aware of it. You can report on it, think about it, and use it to make decisions.

The critical prediction of GNWT is that the transition from unconscious to conscious processing is not gradual. It's sudden. There's a threshold, and crossing it produces a rapid, all-or-nothing amplification of neural activity. Below the threshold, the stimulus remains unconscious no matter how close it gets. Above the threshold, it ignites into full conscious awareness.

This was a bold prediction. And EEG turned out to be one of the best tools for testing it.

The P300: Consciousness Leaves a Signature in Your EEG

When the global workspace ignites, it leaves an electrical fingerprint that EEG can detect. And the most well-studied version of that fingerprint has a name: the P300.

The P300 is a positive voltage deflection that appears in the EEG signal approximately 300 to 600 milliseconds after a stimulus that reaches conscious awareness. It was discovered by Samuel Sutton and colleagues in 1965, decades before global workspace theory existed. But in retrospect, the P300 turned out to be almost exactly what GWT would predict: an electrical marker of the moment information gets broadcast across the brain.

Here's the evidence. In experiments where stimuli are presented at the threshold of awareness (using masking, very brief presentations, or attentional manipulation), the P300 appears only for stimuli that subjects report seeing. If the same stimulus is presented but the subject doesn't consciously perceive it, there is no P300. The earlier ERP components (P100, N100, even the N200) may still appear for unseen stimuli, reflecting the unconscious local processing that happens before ignition. But the P300 marks the boundary. It's the EEG signature of information crossing the ignition threshold and entering the global workspace.

Dehaene and his colleagues demonstrated this dramatically in a series of experiments using masked stimuli. They would flash a word on a screen for a very brief duration, immediately followed by a visual mask. By varying the timing of the mask, they could push the word right to the edge of awareness. In trials where subjects reported seeing the word, EEG showed a clear P300 accompanied by a sudden surge of gamma-band activity and enhanced long-range synchrony between frontal and posterior regions. In trials where subjects didn't see the word, despite identical physical stimulation, these late signatures were absent.

Same stimulus. Same eyes. Same brain. But a fundamentally different pattern of neural activity depending on whether the information reached consciousness or not.

Gamma Synchronization: The Sound of the Broadcast

The P300 isn't the only EEG signature of global workspace activity. There's another, arguably more interesting signal: late gamma-band synchronization.

gamma brainwaves (typically 30-100 Hz, though the range associated with consciousness is often 30-50 Hz) are the fastest oscillations in the brain. They're generated when large populations of neurons fire in tight temporal coordination, synchronizing their activity within windows of just a few milliseconds.

In the context of global workspace theory, gamma synchronization serves as the binding mechanism. When information ignites in the workspace, distant brain regions need to be coordinated. They need to be "talking about" the same thing at the same time. Gamma oscillations provide this coordination by synchronizing the firing of neurons across regions.

Here's what makes this relevant to consciousness. Early gamma activity (within the first 100-200 milliseconds after a stimulus) occurs regardless of whether the stimulus is consciously perceived. This early gamma reflects local processing, the specialized modules doing their thing backstage. But late gamma activity (300 milliseconds and beyond), particularly when it's synchronized between frontal and posterior regions, appears only for consciously perceived stimuli.

This late, distributed gamma synchronization is the electrical fingerprint of the broadcast. It's the workspace reverberating with the information that just entered it. And it's visible in EEG.

Research by Melloni, Rodriguez, and others has shown that the degree of long-range gamma synchronization between frontal and parietal electrodes predicts conscious perception on a trial-by-trial basis. When gamma synchrony is high, the subject reports seeing the stimulus. When gamma synchrony is low, the stimulus remains unconscious. The same physical stimulus, processed by the same brain, with the only difference being the degree to which distant brain regions synchronized their activity in the gamma band.

The Competition Backstage: What Determines What Becomes Conscious

If the workspace has limited capacity (and the evidence overwhelmingly says it does), then there must be a competition. Many processes vie for access, but only a few win. What determines the winner?

The Crown captures brainwave data at 256Hz across 8 channels. All processing happens on-device. Build with JavaScript or Python SDKs.

Explore the Crown

Global workspace theory identifies several factors that influence which information wins the competition for conscious access:

Stimulus strength. A loud sound is more likely to reach consciousness than a quiet one. A bright flash beats a dim one. Stronger sensory signals activate local processors more strongly, making it more likely that the activation will cross the ignition threshold.

Attention. This is the big one. Attention acts as an amplifier in the competition for workspace access. When you direct attention to something, you bias the competition in its favor. An attended stimulus that might otherwise be too weak to cross the ignition threshold gets boosted above it. This is why you can hear your name in a noisy room (the cocktail party effect): your attention system has pre-amplified information matching your name, giving it a competitive advantage for workspace access.

Current goals and expectations. The prefrontal cortex, which is part of the workspace network itself, sends top-down signals that bias the competition based on what you're trying to do. If you're looking for your friend in a crowd, face-like stimuli get a competitive boost. If you're proofreading a document, letter and word patterns get priority.

Novelty and surprise. Unexpected stimuli have a natural competitive advantage. A sudden change in the environment, something that doesn't match the brain's predictions, generates a strong prediction-error signal that pushes toward ignition. This is adaptive: novel and surprising events are more likely to be important for survival.

Emotional salience. Threatening or emotionally charged stimuli get preferential access to the workspace. The amygdala, which processes emotional significance before conscious awareness is established, can boost certain stimuli past the ignition threshold even when attention is directed elsewhere. This is why you might "unconsciously" detect a snake in your peripheral vision before you consciously see it.

The Workspace in Sleep, Anesthesia, and Coma

One of the most powerful applications of global workspace theory is explaining why consciousness disappears in certain states.

During general anesthesia, the frontoparietal broadcasting network is selectively disrupted. Anesthetic agents don't shut down all brain activity. Local processing in sensory cortices continues. But the long-range connections that enable global broadcasting are interrupted. The workspace goes offline. Information gets processed locally but never ignited, never broadcast, never made available to the full range of cognitive systems. The result: unconsciousness, even though neurons are still firing.

EEG confirms this. Under anesthesia, the P300 disappears. Gamma synchronization between frontal and posterior regions collapses. event-related potentials show preserved early components (local processing continues) but absent late components (broadcasting has stopped). The workspace signature, specifically, is gone.

Deep non-REM sleep tells a similar story. The thalamocortical system enters a state of slow oscillation, where large populations of neurons alternate between synchronized "up states" (depolarized, active) and "down states" (hyperpolarized, silent). These slow waves disrupt the fine-grained, differentiated activity patterns needed for global broadcasting. The workspace can't maintain the complex, reverberant activity that ignition requires.

And then there's REM sleep, where things get interesting. During REM, some frontoparietal broadcasting capacity returns. Not at the level of wakefulness, but enough to support the vivid, immersive conscious experiences we call dreams. EEG during REM shows partial restoration of gamma synchronization and some long-range coherence, though with different patterns than wakefulness. This is consistent with global workspace theory's prediction: partial broadcasting capacity produces partial consciousness, dream consciousness that is vivid and experiential but lacks the full executive control of waking awareness.

GWT vs. IIT: The Great Consciousness Debate

Global workspace theory is the leading theory of consciousness. But it's not the only one. Its primary competitor is integrated information theory (IIT), developed by neuroscientist Giulio Tononi. And the debate between these two theories is one of the most fascinating conflicts in modern science.

GWT says consciousness is about what information does: it becomes conscious when it's broadcast globally. The critical thing is the broadcasting process, and it requires the specific prefrontal-parietal infrastructure that enables it.

IIT says consciousness is about what information is: any system with a high degree of integrated information (measured by a quantity Tononi calls Phi) is conscious to some degree. The critical thing is the intrinsic causal structure of the system itself. And crucially, IIT doesn't require a prefrontal cortex. In principle, any system with sufficient integrated information, biological or not, could be conscious.

These theories make different predictions, and several large-scale experiments are currently underway to test them against each other. The Templeton World Charity Foundation funded a series of "adversarial collaborations" in which proponents of GWT and IIT pre-registered their competing predictions before running the same experiments. The initial results, published in 2023, provided partial support for both theories and definitive evidence for neither. The P300 and frontoparietal broadcasting predicted by GWT were confirmed in some paradigms. But the posterior cortical involvement emphasized by IIT was also observed in others.

The honest answer, as of 2026, is that we don't know which theory is right. We may need a new theory that incorporates insights from both. Or we may need entirely new experimental approaches that can distinguish between them more clearly.

What we do know is that both theories take EEG seriously as a tool for studying consciousness. The electrical signatures of neural activity, captured in real time with millisecond precision, provide some of the strongest evidence for or against any theory of consciousness. The temporal dynamics of how information propagates through the brain, when it ignites, how it reverberates, and which regions synchronize with which, are exactly what you need to study if you want to understand how awareness works.

Measuring the Workspace With Consumer EEG

For most of the history of consciousness research, studying the neural signatures of awareness required million-dollar fMRI scanners or dense EEG arrays with 128 or 256 electrodes in carefully controlled laboratory settings. The global workspace wasn't something you could observe in the wild.

That barrier is lowering. Consumer-grade EEG devices now have enough temporal resolution and spatial coverage to capture the key signatures that global workspace theory predicts.

The Neurosity Crown, with 8 channels at CP3, C3, F5, PO3, PO4, F6, C4, and CP4, covers the frontal and parietal regions that form the core of Dehaene's global neuronal workspace. Its 256Hz sampling rate provides the temporal resolution needed to detect P300-like components and gamma-band dynamics. The on-device N3 chipset processes signals in real time, which means you can observe consciousness-related patterns as they happen.

Through the Crown's SDKs, developers can access raw EEG data and power spectral density across all frequency bands. This opens the door to building applications that track workspace-related metrics: frontal-parietal coherence in the gamma band, the presence and latency of P300-like responses, and the balance between local processing (early ERPs) and global broadcasting (late ERPs).

The focus and calm metrics the Crown already provides are, in a sense, workspace-adjacent measures. A high focus score reflects a state where the frontoparietal network is actively engaged, which is the same network that forms the global workspace. A state of sustained focus is a state where the workspace is consistently broadcasting task-relevant information while suppressing competing distractors.

For researchers and curious builders, the MCP integration takes this further. Imagine an AI that can track the neural signatures of your conscious processing in real time, one that can detect when information has "ignited" in your workspace based on EEG signatures and adapt its outputs accordingly. This isn't consciousness science fiction. It's a plausible near-term application of the same science that Baars and Dehaene built their theories on.

The Spotlight Sweeps On

Bernard Baars started with a metaphor: a theater, a spotlight, a darkened audience. Stanislas Dehaene turned that metaphor into circuits and predictions. EEG gave those predictions teeth by providing millisecond-resolution evidence of when and how the brain transitions from unconscious to conscious processing.

We still don't fully understand consciousness. We don't know why physical processes give rise to subjective experience. The hard problem remains hard. But we now know more about the mechanics of consciousness than at any other point in human history. We know there is an ignition threshold. We know it involves frontoparietal broadcasting. We know it leaves electrical signatures that can be measured in real time.

And perhaps the most striking implication of global workspace theory is how much of your mental life happens outside the spotlight. The vast majority of your brain's processing, the pattern recognition, the motor planning, the emotional evaluation, the memory retrieval, happens backstage, in the dark, without ever reaching your awareness. You are a small, bright point of consciousness floating on an ocean of unconscious computation.

The spotlight is narrow. It can only illuminate one thing at a time. But it's the only light you've got. And for the first time, you can watch it sweep.