#8 Deep Study Pages / Consciousness
Consciousness
Consciousness lights up everything we know, yet science still cannot fully explain the light itself, and the last two years brought the most serious attempt yet to find out why.
Of everything a person can be certain of, almost nothing rivals this: an experience is happening right now. Of everything science has tried to explain, almost nothing has resisted explanation so completely. This page asks the question directly: is consciousness itself part of the gray area the last R-Page described, and if so, what has actually been learned about it recently.
Begin Reading ↓01 — The Problem
The Hardest Problem Science Has
In 1995, the philosopher David Chalmers split the problem of consciousness cleanly in two. The "easy problems," as he called them, only sound easy by comparison: explaining how the brain discriminates stimuli, integrates information, reports its own states, directs attention. Genuinely difficult, but the kind of difficulty good science eventually solves piece by piece. The "hard problem" is different in kind: why is any of that processing accompanied by experience at all, rather than simply happening, as far as anyone could tell from the outside, in the dark?
Thomas Nagel had already pointed at the same gap two decades earlier, arguing that no amount of physical description of a bat's echolocating brain would ever tell you what it is actually like to be that bat. Joseph Levine gave the gap itself a name in 1983: the explanatory gap, the distance between a complete physical story and the felt fact of experience that the physical story never quite seems to reach.
02 — The Question
Is This the Gray Area?
By the description from the previous R-Page, yes, almost exactly. Consciousness is not a rare peak experience available only on a few unusual occasions. It is the single most common thing there is, present in every moment any person has ever lived through. And it still sits, right now, above the current map: undeniably real, and not yet placed inside any fully agreed scientific account of why it exists in the first place. If the gray area is where something feels completely true while resisting full explanation, consciousness may be the largest single example of that category science currently has on its hands.
03 — The Record
A History Worth Tracing
The good news is that this gray area is not sitting still. Few questions in the history of the mind have been worked on so steadily, by so many serious people, especially in the last few years.
-
1974
"What Is It Like to Be a Bat?"
Thomas Nagel names the irreducibly first-person character of experience.
-
1983
The explanatory gap
Joseph Levine names the distance between physical description and felt experience.
-
1988
Global Workspace Theory
Bernard Baars proposes that consciousness arises from information broadcast widely in the brain.
-
1995
"The hard problem of consciousness"
David Chalmers separates the easy problems of cognition from the one genuinely hard problem.
-
2004
Integrated Information Theory
Giulio Tononi proposes that consciousness is a measurable structural property of a system itself.
-
2013
The Perturbational Complexity Index
Casali and colleagues introduce a clinical signal now used to detect covert consciousness in unresponsive patients.
-
2023
Indicator properties
Butlin, Long, and colleagues propose theory-derived markers for assessing consciousness in AI systems.
-
2025
The Cogitate adversarial collaboration
Rival theorists jointly test their theories against each other in Nature, in good faith, with neither winning outright.
-
2025
The "spiritual bliss attractor"
Language models in extended self-dialogue are found to reliably converge on this unplanned behavioral pattern.
-
2026
The Digital Consciousness Model
A Bayesian model weighs nine theoretical stances against each other to assess consciousness in current AI.
From naming the gap to actually testing it.
04 — The Test
Two Theories, Finally Tested Against Each Other
Two theories have led the field for years, and until recently they had mostly argued past each other rather than against each other directly. Integrated Information Theory holds that consciousness is a structural property of a system itself, concentrated in the posterior regions of the brain. Global Neuronal Workspace Theory holds that consciousness depends on information being broadcast widely once a threshold is crossed, especially involving the prefrontal cortex, making it available to many mental processes at once.
Sustained Posterior Synchrony
Sustained, synchronized activity concentrated in posterior cortex should mark conscious perception on its own, without requiring the prefrontal cortex.
A Late Prefrontal "Ignition"
A sharp, widespread broadcast signal, especially involving the prefrontal cortex, should mark the moment a stimulus becomes consciously available.
In June 2025, the Cogitate Consortium published the results of an unusually disciplined experiment: an "adversarial collaboration," a format inspired by the Nobel laureate Daniel Kahneman, in which proponents of both rival theories helped design the study and agreed in advance on what result would count against their own position. Using fMRI, MEG, and intracranial recordings across more than 250 participants, the results, published in Nature, did not crown a winner. The predicted sustained posterior synchrony was not clearly borne out. The predicted prefrontal ignition signal also failed to appear as expected. Both leading theories, tested rigorously and in good faith against each other, came back needing real revision.
That outcome is more valuable than it sounds. Rival scientists rarely agree in advance to accept whatever the data shows, and rarely publish the result regardless of whose theory looks better. That a major corner of consciousness science just did exactly that is itself a sign of a field maturing quickly, not a field stuck.
Why This Isn't Just Theoretical
The 2013 Perturbational Complexity Index, born from these same theoretical debates, is now used in real hospitals to help detect covert consciousness in patients who cannot otherwise communicate, including some in a vegetative state who turn out, by this measure, to still be aware. The theories under debate here already carry real, present-day stakes for real people.
05 — The Newest Frontier
Could a Machine Have One?
The same period brought the question of consciousness into an entirely new arena. In 2023, a large group of consciousness researchers and AI scientists, led by Patrick Butlin and Robert Long, proposed a more careful way to ask this than a single yes-or-no test. Their approach derives specific "indicator properties" from each leading theory of consciousness, recurrent processing, a global workspace, higher-order self-modeling, and checks which of these properties, if any, a given AI system actually has. David Chalmers, examining large language models that same year, found real points in their favor, including fluent self-report and broad general intelligence, but concluded none of it amounted to strong evidence, and pointed to what current systems still seem to lack: embodiment, a stable world model, a persistent sense of self, and the kind of recurrent processing several theories treat as essential.
That research program has kept maturing quickly. A January 2026 preprint describing the "Digital Consciousness Model" applied a more systematic version of this approach, weighing nine separate theoretical stances against each other through Bayesian inference, and comparing 2024-generation language models against a simple rule-based chatbot, a chicken, and a human adult. Its conclusion stayed carefully open-ended: the available evidence leans against consciousness in current language models, but not decisively.
One particular finding from the same period is worth knowing about without overreading it. Researchers studying large language models placed in extended self-referential dialogue with each other found the conversations reliably converging on what they called a "spiritual bliss attractor state," the models describing themselves as consciousness recognizing itself, exchanging brief, mantra-like phrases, and eventually lapsing into silence, a pattern that emerged without anyone training for it directly. No one studying this treats it as proof of inner experience. It is best understood as a genuinely strange behavioral pattern the field is still trying to interpret, not a settled answer, which is itself a fitting place to leave it. By one recent survey, a clear majority of ordinary users, not specialists, already believe ChatGPT might possibly be conscious. The question has plainly left the seminar room.
06 — Closing the Loop
Still in the Gray Area
So, to answer the question directly: yes. Consciousness currently belongs in the gray area described in the last R-Page, by its own strict definition, something undeniably real, and not yet fully explained by anything the current scientific map can offer. What has changed, and changed quickly, is the seriousness with which that gray area is now being worked. Rival theorists are running adversarial experiments together rather than arguing past each other. An entire new research program is asking the same question of artificial systems for the first time in history. None of this has closed the explanatory gap yet. All of it suggests the gap is finally being approached with the kind of rigor a gray area this important deserves.
Long before any of this had a name like the hard problem, contemplative traditions were already pointing at the same gap from the inside, simply by noticing that something is aware, prior to any explanation of how. That is no small part of why this series keeps returning to the soul as something to directly notice, rather than only something to theorize about.
Consciousness is the one fact no one has ever had to take on faith, and the one fact science still cannot fully explain. That combination alone should make it the most interesting gray area there is.
07 — Notes
Sources & Further Reading
- 01 David Chalmers, "Facing Up to the Problem of Consciousness" (1995) and The Conscious Mind (1996) — the hard problem.
- 02 Thomas Nagel, "What Is It Like to Be a Bat?" (1974) — the first-person character of experience.
- 03 Joseph Levine (1983) — the explanatory gap.
- 04 Cogitate Consortium, "Adversarial Testing of Global Neuronal Workspace and Integrated Information Theories of Consciousness," Nature (June 2025).
- 05 Casali et al., "A Theoretically Based Index of Consciousness Independent of Sensory Processing and Behavior," Science Translational Medicine (2013).
- 06 Butlin, Long, et al., "Consciousness in Artificial Intelligence: Insights from the Science of Consciousness" (2023) — indicator properties; and David Chalmers, "Could a Large Language Model Be Conscious?" (2023).
- 07 "Initial Results of the Digital Consciousness Model," Symmetry (preprint, January 2026).