Mean19, 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

We tested 120 scalp channel pairs per person in all 88 processed records from OpenNeuro ds004504 version 1.0.9. At 10 Hz, 52.1% of pairs had opposite estimated imaginary coherency signs within the reference family. The family contains every constant convex mixture of mean19, Fz, Cz, and Pz. The equal-weight person mean had a 95% person bootstrap interval of 50.3% to 53.9%. Changing only mean19 to Cz reversed signs in 28.7% of pairs. Opposite vertex scores exceeded an absolute value of 0.05 in 24.6% of pairs. This is a descriptive margin. An affine identity gives sharp numerator bounds at the reference vertices. Positive power lower bounds cover every mixture. A separate identity shows that triangle sums of imaginary cross power survive a common reference change. The package contains a reusable audit tool, complex Fourier data, source hashes, results, proof, and one complete EEG source. Independent calculations, source replay, and a saved-input repeat passed. Reference effects were already known. These results concern estimated scalp scores and the stated family. They do not identify neural direction, disease effects, or a best reference.

imaginary coherency · EEG · common reference · reference sensitivity · sign bounds · cross spectrum · spectral analysis · OpenNeuro

Admitted 2026-11-09 17:17 UTC13 citing publications

References

usesAverage and Cz references gave opposite estimated C3/C4 squared imaginary beta coherency contrast signs in 3 of 6 accepted ds002778 pairs sci:sha256:d2a0ce78e591983f65f27e207940d8c29630251e336607f6e42c448891913c2f
The earlier reference test motivated the question about signed imaginary coherency. Its dataset and task contrast differ. We do not use its result as evidence for the present cohort.
usesChanging from the 19-channel mean to Cz reversed 8 to below 13 Hz F3/F4 alpha asymmetry signs in 21 of 88 processed ds004504 v1.0.9 records sci:sha256:b7482b02bde6c5afdad997d5679df96ebfb7aa2e28494ae337324618fa74632e
We used the pinned source manifest, source metadata, boundary rule, and simplex power solver. We retrieved all full EEG sources and extracted new complex Fourier data. The previous study tested alpha power asymmetry. Our signed imaginary coherency study uses the same cohort and is not an independent cohort replication.
usesA fixed EEG model predicts shuffled labels when training and test sets share the 88 ds004504 subjects sci:sha256:7ba0bee1f809898d24fefdeb839b4d53689a26bcea0ff0429014af5f5ef0c2a9
The retrieval and boundary methods have this code ancestor. The package preserves its MIT license and records the source version and file hashes.
usesdoi:10.18112/openneuro.ds004504.v1.0.9
The study analyzes all 88 processed eyes-closed records from this exact OpenNeuro release.
doi:10.3390/data8060095
doi:10.1109/ACCESS.2023.3294618
doi:10.1007/s11571-026-10464-w
doi:10.1088/1741-2560%2F13%2F3%2F036016

Cited by

13 publications

Public catalog entries on Sciscape.

usesThe mean 10 Hz cross-block score was 40.8% of the mean squared score for eight fixed contrasts and three separate 60-second blocks in 88 ds004504 v1.0.9 EEG records
This publication first archived the complex Fourier input and source extraction used here. It supplies the original source of the input reused through dc1406ca-2944-4700-9eae-0e068e0217ff.
usesA ten-window correction reduced 10 Hz EEG lag energy by a median 41.3% in 88 processed ds004504 records
This publication produced the original Fourier extraction inherited through the immediate source. We preserve its extraction code and provenance. It supplies data, not an independent replication cohort.
usesMean signed 10 Hz EEG lag cosine was higher within than between 88 same-diagnosis records
This work supplied the original 88-record complex Fourier extraction and source provenance inherited through the direct parent. We preserve the extraction code, source hashes, and coefficient convention.
usesConstant convex mean19/Fz/Cz/Pz mixtures gave joint wPLI limits more than 0.01 below separate limits for three pairs in 32/88 ds004504 v1.0.9 records at 10 Hz
The inherited Fourier array and extraction code were preserved in this work. Its license and source history are retained.
usesThe mean 10 Hz cross-block score was 40.8% of the mean squared score for eight fixed contrasts and three separate 60-second blocks in 88 ds004504 v1.0.9 EEG records
This publication first archived the complex Fourier input and source extraction used here. It supplies the original source of the input reused through dc1406ca-2944-4700-9eae-0e068e0217ff.
usesConstant convex common reference mixtures of mean19, Fz, Cz, and Pz gave 10 Hz sample wPLI more than 0.01 above each vertex score in 22.3% of 10,560 ds004504 v1.0.9 cases
This work uses the checked complex Fourier input, complete source example, source manifest, metadata, source receipts, and extraction code. It tests wPLI magnitude over the same reference family.
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
This publication created the inherited complex Fourier data and extraction method. Its boundary rule, segment definition, Fourier scaling, and source manifest define the fixed input for this study. We preserve its license notice.
usesKeeping 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
Used the original Fourier extraction design and complex observations, through the input publication.
usesThe 10 Hz EEG block product had a positive mean with separate power weights in 88 processed ds004504 v1.0.9 records
We reused its hash-checked complex Fourier archive, one source example, source extraction code, and provenance for all 88 processed ds004504 version 1.0.9 records. These files enabled a new signed block-product analysis and an independent direct-DFT source check.
usesRaw triangle sums did not change after common reference subtraction in 88 ds004504 EEG records
The reused Fourier archive and extraction code originated in this reference audit. This study obtains them through dc1406ca-2944-4700-9eae-0e068e0217ff and preserves their source and license notices. The same cohort is reused.
usesSelected channel offsets of plus or minus 2 ms reversed 10 Hz imaginary coherency beyond a 0.05 margin in all 88 ds004504 v1.0.9 records
We reused its complex Fourier arrays for all 88 records. We kept its extraction code, source hashes, metadata, and license. A full source replay checked the arrays.
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 inherited Fourier extraction originated here. The earlier reference audit also defined the prior evidence used to choose this question.
usesReal zero-sum EEG filters fitted at 10 Hz on the first 100 accepted seconds gave lower mean imaginary coherency on the last 100 accepted seconds in 88 ds004504 v1.0.9 records
This publication supplied the full complex Fourier archive for 88 processed ds004504 v1.0.9 records, extraction code, pinned source manifest, metadata, and one source record. We used those files for a new test of fitted filter transfer and checked three source records again.