Concept study / Information & adaptationConceptual illustration. Shapes, colors, and motion are not measured data.
Global workspace approaches emphasize broad availability of information; integrated-information approaches emphasize a system’s intrinsic causal organization. Other families include higher-order and recurrent-processing accounts. These labels cover differences within each family as well as disagreements between them. Seth and Bayne’s review provides a useful map of this landscape. The Hill project should treat that map as a guide to research questions, not as a menu from which to select a preferred cosmic explanation. A credible comparison begins with the exact version of a theory, the phenomenon it aims to explain, and the measurement that its proponents accept as informative.
An example of constructive disagreement
The Cogitate study preregistered contrasting predictions and collected multimodal data from 256 participants. Its results supported some expectations while challenging important tested predictions of both leading frameworks. The study focused on empirical implementations that could be measured, rather than resolving every mathematical or philosophical commitment. We propose following that discipline in the atlas: distinguish a challenged prediction from the rejection of an entire theoretical family. Record what was tested, what result was observed, and which alternative interpretations remain available.
Build a prediction table
For each theory, the proposed table would list the target phenomenon, necessary conditions, predicted observations, and possible confounds. An empty prediction cell is meaningful: it identifies where a position has not yet supplied a discriminating test. A score based only on citation counts or attractive diagrams would be misleading. We suggest allowing readers to compare pairs of theories on one specific question at a time. That keeps the comparison understandable and avoids suggesting that one experiment ranks every possible explanation of consciousness.
How Hill’s proposal enters fairly
If Hill’s framework connects consciousness to nested geometry or retained information, it should submit the same materials as any competing account: definitions, a mechanism, a quantitative prediction, and a failure condition. Merely being compatible with an existing observation is insufficient because many models can be made compatible after the fact. We propose stating the strongest simple alternative and identifying the experiment that separates them. If there is currently no such experiment, the proposal remains exploratory and philosophical rather than an established physical theory.
The mathematical lens
A useful experiment targets a measurable difference between specified models.
Turn it into a question.
- Question
- Which observation would distinguish two named versions of a theory?
- Method
- Write a comparison protocol with divergent predictions, confounds, and interpretation rules before analysis.
- What would count against it?
- If both theories accommodate every possible result unchanged, the proposed test lacks discriminatory power.
Compare predictions and scope, not the appeal of a theory’s vocabulary.
Follow the original work.
These sources support the mechanisms and limits discussed here. They do not endorse Hill’s framework. Notes distinguish research proposals, reviews, and observations.
Continue the investigation
Edition 02 · 18 October 2026 · AI-assisted synthesis. See editorial method.