Keeping 17 pairs removed return paths in all 88 ds004504 v1.0.9 EEG graphs at 10 Hz from the first 20 accepted windows with the equal mean reference of 19 channels
We tested all 88 processed EEG records in ds004504 v1.0.9. Each main test used the first 20 accepted windows at 10 Hz. It used the equal mean reference of 19 channels. Every signed edge in each full graph had a return path. We kept the 17 pairs with the largest absolute imaginary coherency. Each of the 88 graphs then had an edge without a return path. Pure source nodes and pure sink nodes each occurred in 87 records. A proof covers real, strictly positive reference weights and common observations. It uses every referenced channel. Positive channel powers are required for coherency. An audit tool reports omitted flow that balances retained net flow. We downloaded all 88 source files again and rebuilt the Fourier input byte for byte. Independent code checked 616 spectra and 4,928 graph masks. A clean installation gave the same graph counts within the stated numerical limits. The package preserves code, full Fourier input, source hashes, metadata, results, proof, and reproduction commands. Edge arrows represent matrix signs. They do not establish neural causes.
EEG · imaginary coherency · reference weights · directed graphs · circulation · return paths · flow certificates · OpenNeuro ds004504
References
7- usesMean19, Fz, Cz, and Pz references gave opposite 10 Hz imaginary coherency signs in 52.1% of tested pairs in 88 ds004504 v1.0.9 records sci:sha256:094b90e4b977e5284645863ffa4e5a4798181fe7bf966f267482939912866202
Used the original Fourier extraction design and complex observations, through the input publication.
- usesSelected removal of at most two of the first 20 accepted ten-second windows reversed 10 Hz imaginary coherency beyond a 0.05 margin in 32 of 88 ds004504 v1.0.9 records under a fixed mean19 reference sci:sha256:70ac6860cd79884ac1cba8e563b97f2dec774efd9d0bffec987334ba8e971aa5
Used the complete Fourier input, source manifest, metadata, source receipts, and extraction code. The new rebuild verified all 88 source records.
- usesRaw triangle sums did not change after common reference subtraction in 88 ds004504 EEG records sci:sha256:a19e5a14a0cf0fe0b6e33b44c74607ab38e8a468cc87ebe7ce5112e8938f092e
Used the earlier mean reference and Hodge analysis to set the scope of the new method. This work adds positive weighted reference constraints, support graph checks, and removal certificates.
- usesdoi:10.18112/openneuro.ds004504.v1.0.9
The analysis used all 88 processed EEG records from this pinned dataset version.
- doi:10.3390/data8060095
Dataset descriptor cited for attribution and provenance of the analyzed EEG archive.
- supportsdoi:10.1088/1741-2560%2F13%2F3%2F036016
Prior evidence that EEG reference choice changes imaginary coherency and graph measures.
- doi:10.1016/j.clinph.2004.04.029
Background for imaginary coherency and its stated mixing assumptions.
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesConvex reference changes at total variation 1/19 reversed 359 of 11,138 tested mean-reference imaginary coherency signs at 10 Hz in ds004504 v1.0.9
The full Fourier archive, source manifest, metadata, receipts, and extraction code supplied the fixed input for this audit.