2026-07-22 · Preprint · bioRxiv
Testing the spatiotemporal predictions of global neuronal workspace theory and integrated information theory

A re-analysis of the COGITATE brain recordings finds the frontal 'flare-up' at picture offset that the original analysis missed.
The same public data, analysed with a different method that asks which brain area drives which, gives a different answer from the original paper on one key point. This is a warning: with this much freedom in how to analyse the data, the conclusion can depend on the method.
Dynamic causal modelling on the COGITATE MEG data (N = 100), task-irrelevant faces, split discovery-validation, expanding time windows. Support for GNWT's prefrontal ignition at both onset and offset (offset not seen in the original analyses). For IIT, fusiform-V1 feedback was not sustained, but intrinsic fusiform connectivity was broadly consistent with sustained processing; a post-hoc face-vs-object validation gave no strong evidence of face selectivity (pp < 0.95), so the IIT findings are read conservatively. Not peer reviewed.
What we do with it
This is the 'July 2026 reanalysis' the plan could not name. It goes into our multiverse on COGITATE (task #11) as one of the analysis paths, and shows why we hash the analysis set first. Planned
Read the original. Source checked against the original.