When does cortical coherence carry shared content? Compatibility and self-reconstruction as testable conditions
Abstract
Synchrony accounts treat collective phase order as the mechanism that integrates distributed cortical activity. We show formally that shared content and a self-model each need a condition beyond phase order, and state it. Three conditions are separated: coherence (collective phase order), compatibility (agreement of independently decoded local descriptions where regions overlap) and self-reconstruction (accurate recovery of the global state from an internal encoding). Compatible local descriptions determine a unique global state. Coherence bounds disagreement between overlapping descriptions only through shared encoders of bounded sensitivity, and the bound grows as the square root of the population read: on one simulated cortical sheet it constrains at most 10 sites, chaining through overlapping patches extends it to 0.312 mm, and across distant regions the link from coherence to content is an assumption. A contracting self-map fixes one state but reconstructs accurately only states within a diameter set by its error and contraction factor. Lean 4 verifies these implications; one model satisfies all connecting hypotheses at once, showing consistency, not cortical realization. The conditions yield a disconfirmable comparison: at matched phase coherence and single-region decoding accuracy, decoders fitted independently on two regions should agree more, beyond their separate accuracies, on perceived than on unperceived stimuli classified by a no-report marker. Choosing the overlapping regions by what each decodes makes the unity correlate operational relative to a pre-registered decoder and criterion. An exploratory EEG analysis is a measurement check only. Unity and a minimal self remain interpretations that need an independent experiential endpoint.
Related
- Hippocampal Neuronal Overflow and Cortical Reconstruction: A Hypothesis on the Formation of Neural Networks Unique to the Human Brain
- CortexDecoder: fMRI-to-image Reconstruction from Cortical Flatmaps
- Top2All: Benchmarking sparse-to-transcriptome reconstruction from high-expression profiles across sex- and sex-steroid-defined conditions
- Evidence-Calibrated Reanalysis of an Otosclerosis Locus Generates a Testable Myeloid State Hypothesis
- From seven combination hypotheses to one testable interaction: a gated agentic AI QSP workflow applied to healthy ageing interventions
BibTeX
@misc{suarez2026when,
title = {{When does cortical coherence carry shared content? Compatibility and self-reconstruction as testable conditions}},
author = {Sebastian Bobadilla Suarez},
howpublished = {PsyArXiv},
year = {2026},
month = sep,
doi = {10.31234/osf.io/fydt3_v1},
url = {https://osf.io/preprints/psyarxiv/fydt3_v1/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - When does cortical coherence carry shared content? Compatibility and self-reconstruction as testable conditions
AU - Sebastian Bobadilla Suarez
PY - 2026
DA - 2026/09/27
PB - PsyArXiv
DO - 10.31234/osf.io/fydt3_v1
UR - https://osf.io/preprints/psyarxiv/fydt3_v1/
AB - Synchrony accounts treat collective phase order as the mechanism that integrates distributed cortical activity. We show formally that shared content and a self-model each need a condition beyond phase order, and state it. Three conditions are separated: coherence (collective phase order), compatibility (agreement of independently decoded local descriptions where regions overlap) and self-reconstruction (accurate recovery of the global state from an internal encoding). Compatible local descriptions determine a unique global state. Coherence bounds disagreement between overlapping descriptions only through shared encoders of bounded sensitivity, and the bound grows as the square root of the population read: on one simulated cortical sheet it constrains at most 10 sites, chaining through overlapping patches extends it to 0.312 mm, and across distant regions the link from coherence to content is an assumption. A contracting self-map fixes one state but reconstructs accurately only states within a diameter set by its error and contraction factor. Lean 4 verifies these implications; one model satisfies all connecting hypotheses at once, showing consistency, not cortical realization. The conditions yield a disconfirmable comparison: at matched phase coherence and single-region decoding accuracy, decoders fitted independently on two regions should agree more, beyond their separate accuracies, on perceived than on unperceived stimuli classified by a no-report marker. Choosing the overlapping regions by what each decodes makes the unity correlate operational relative to a pre-registered decoder and criterion. An exploratory EEG analysis is a measurement check only. Unity and a minimal self remain interpretations that need an independent experiential endpoint.
ER -