Using a neural net to debug the global neuronal workspace

The conversation about AI and consciousness is mostly science fiction and panic. The more interesting story is what happens when you use a neural network not to build a mind, but to build a working model of the wiring inside your own. These simulations are giving the system the first clear, mechanical explanation for that 'aha' moment when a solution suddenly clicks into place.

You’ve been wrestling with a problem all morning. A tricky email, a budget that won’t balance, a piece of code that keeps failing. You’ve stared at it, poked it, and sworn at it. Nothing. You give up and go for a walk. As you’re turning a corner, looking at a tree, the answer arrives. Not as a slow deduction, but as a complete, fully-formed insight that lands in your head all at once. It feels like magic, a gift from the ether. It is not. It is a predictable, physical event in your brain’s processing, and AI models now offer a mechanical proof.

Your brain's internal broadcast system

For decades, the leading theory of how conscious thought works has been the Global Neuronal Workspace. The idea is that your brain is not a single, unified computer. It’s more like an orchestra where hundreds of musicians are quietly practicing their own parts. One section works on rhythm, another on harmony, another on the melody of a feeling in your left foot. Most of this work happens locally, inaudible to the rest of the ensemble. This is your 'unconscious' processing—not a Freudian cellar of repressed desires, but a vast, silent, parallel-processing engine.

Consciousness, the theory goes, is what happens when the conductor points to one section. The information is taken from its silo and broadcast across a brain-wide network, becoming available to all the other players—memory, language, and planning. This is the 'global workspace'. A fleeting sensation becomes a conscious thought when its melody hits the main stage and gets the attention of the whole orchestra. It’s the difference between the violas quietly tuning and the main theme being played, full-blast, by every instrument at once.

The AI model of the 'aha' ignition

This theory was elegant but hard to prove in a living brain. So, researchers built an artificial neural network modelled on this exact design. They gave it specialised 'unconscious' processors and a central 'workspace' network. Then they gave it problems to solve, just at the edge of its ability to perceive them.

What the model showed was stunningly familiar. For subliminal stimuli—information that was too faint or too fast—activity remained trapped in the specialist processors. The AI was technically 'seeing' it, but it wasn't 'aware' of it. The digital violas were playing, but no one else heard them.

But when the stimulus was strong enough and held for long enough, a sudden, non-linear phase transition occurred. Activity would abruptly ignite across the global workspace network in an all-or-nothing cascade. This ignition event is the neural signature of conscious access. It’s the mechanical reality of your 'aha' moment: a piece of information finally gathering enough energy to break out of its local circuit and go global. It isn't a slow crescendo. It's a switch being flipped.

The metabolic price of going global

That ignition event is not free. Broadcasting information across the entire brain is one of the most metabolically expensive things it can do. It requires a sharp, localised surge in glucose and oxygen to fuel the cascade of firing neurons. Your brain, which already consumes about 20% of your body's total energy budget despite being only 2% of your body weight, has to make careful decisions about when to spend this kind of currency.

This is why you don’t have world-changing insights every five minutes. Your brain is a ruthless accountant. If it's already running a deficit—due to poor sleep, chronic stress, or the constant drain of context-switching between tasks—it simply cannot afford the energy spike required for ignition. The information stays siloed in local, unconscious circuits. The problem remains unsolved, not because you aren't smart enough, but because your brain literally does not have the fuel to light the fuse. Your 'aha' moment is waiting for a solvent budget.

Common Questions

Does this mean an AI can be conscious?

No. The model is a simplified simulation of information flow, not a sentient being. It simulates the mechanics of a 'broadcast' but has no body, no lived experience, no internal state, and no self-awareness. It's a tool for testing a hypothesis about wiring, like using a wave tank to understand how tsunamis form. You can model the physics without creating an ocean.

So brain fog is just an energy crisis?

It's a very useful way to frame it. Brain fog feels like a personal failing, a character flaw. This model reframes it as a hardware problem. The 'ignition' event is metabolically expensive. If you are low on energy from poor sleep, chronic stress, or a day of back-to-back meetings, your brain may lack the resources to push information into the global workspace. The data is there, but it's stuck in local, 'unconscious' processing. You can't access it. That feeling of wading through mental treacle is the feeling of a system that can't afford to go global.

Can I force an 'aha' moment to happen?

No, and trying to is counterproductive. The ignition is the result of unconscious processing, not the cause of it. Staring at a problem and demanding a solution is like shouting at the orchestra to play louder. It just creates noise. The real work happens when you create the conditions for the quiet, background processing to complete its task and gather enough signal strength to make the leap. This means managing your brain's energy budget, not brute-forcing the insight.

What to do this week

You can't command an insight, but you can set the table for one. This week, experiment with creating the conditions for ignition instead of trying to force it.

• Schedule a 'do nothing' break. Find one ten-minute slot in your day. Your only job is to stop consuming information. Stare out a window. Watch the kettle boil. Walk to the end of the road and back without your phone. This isn't about 'mindfulness'; it's about creating a silent pause for the background processes to run their course without new inputs interrupting them.
• Isolate one task. Pick one important, cognitively demanding task. For 45 minutes, turn off everything else. No email, no Slack, no phone. The goal isn't just to 'focus'. The goal is to stop spending metabolic currency on context-switching, saving it for the actual problem. You are funding the ignition event for that one task.
• When an insight arrives, rewind the tape. The next time you have an 'aha' moment in the shower or while walking the dog, pause. What were the exact conditions? You were likely relaxed, distracted, and engaged in a low-effort physical activity. You weren't actively 'thinking'. Start a log of these moments. You're not logging the insight itself, but the recipe that produced it.

Where this fits in the Kokorology system

This model of conscious access plugs directly into the core Kokorology principles of load and regulation. Your capacity for insight is not a fixed trait; it's a direct readout of your nervous system's current state and available resources.

A nervous system stuck in a sympathetic (fight-or-flight) state is noisy and expensive. It diverts glucose and oxygen to the muscles and suppresses non-essential, long-term projects—like the quiet, background processing that precedes an 'aha' moment. The system is too busy scanning for threats to allow for deep, creative synthesis. There is no budget for ignition.

Conversely, a well-regulated nervous system that can easily shift into a parasympathetic (rest-and-digest) state has the metabolic surplus and the low-noise environment required for these processes. The 'aha' moment is a luxury item on the nervous system's menu. It can only be ordered when all the essential bills—safety, homeostasis, basic energy needs—have already been paid. Improving your interoception and learning to regulate your autonomic state is therefore not just about feeling calmer; it's about restoring the physiological conditions required for your best thinking.

Closing

Working with your brain's natural processing rhythms, instead of fighting against them, is the most effective way to do better work with less effort. This isn't a productivity hack; it's basic operational maintenance for your own hardware.

• For practitioners: Learn to apply these principles to client work and burnout prevention in The Kokorology Practice.
• For individuals: Book a one-on-one Nervous System Audit to map your own cognitive energy patterns and resource deficits.
• Read the free guide on Somatic Journaling to start tracking the internal signals that precede brain fog and insight.

TL;DR

Consciousness research is no longer just philosophy. AI models of the brain's 'Global Neuronal Workspace' show that conscious awareness is a discrete physical event: a sudden 'ignition' where information is broadcast brain-wide. This explains the 'aha' moment, reframing it from a mysterious insight into a predictable, energy-dependent process. Supporting this process means managing your brain's metabolic budget through sleep, genuine rest, and nervous-system regulation—not by 'focusing harder'. Brain fog is what happens when your system can't afford the price of ignition.

Sources

• Dehaene S, Kerszberg M, & Changeux JP (1998). A neuronal model of a global workspace in effortful cognitive tasks. PNAS.
• van Vugt MK, et al. (2018). The dynamics of a computational model of the global neuronal workspace. NeuroImage.
• Melloni L, et al. (2023). Towards a realistic global neuronal workspace model. Nature Communications.
• Craig AD (2009). How do you feel — now? The anterior insula and human awareness. Nature Reviews Neuroscience.
• Maslach C (2016). Understanding the burnout experience: Recent research and its implications for psychiatry. World Psychiatry.