Output phase controls rejected 512 of 512 null networks under independent contact phases in two SEEG records after maximum beta coherence correction
We tested maximum beta coherence ranks in two public resting SEEG records from OpenNeuro ds005670 version 1.0.0. Each fixed network had 12 bipolar signals and 66 edges. We held measured contact amplitudes fixed and drew independent phases. At a nominal 5% network level, controls drawn after subtraction rejected 512 of 512 null networks. Controls drawn before subtraction rejected 2.73% and 6.25% in the two records. Separate contact pairs and a shorter interval provided checks. A rank proof gives weak familywise error control for valid contact controls under the complete phase null. The package contains source code, compact spectra, all control maxima, quality masks, source hashes and retrieval instructions. This is a conditional method test. It does not measure clinical error rates or neural communication.
stereo-EEG · beta coherence · bipolar montage · maximum statistic · phase randomization · surrogate controls · familywise error · Monte Carlo calibration
References
6- usesRandom contact phases produced median beta coherence near 0.25 in two SEEG records with shared contacts sci:sha256:5475f5dcbaa8cba081f40b61758c931f3ee2b341fb86740808dd2315d85f7337
We used the shared-contact coherence result to choose the network question. We reused the CC0 EDF reader, retrieval code, source provenance and metadata. We made a fresh original-reference extraction and did not use the earlier feature or result arrays.
- usesFull contact ranks controlled beta PSI test size under independent uniform segment rotations in two SEEG records sci:sha256:19e2564c1684522307784f87d688041f7e37a9eed74a2080f731dccc4f4954a6
The full description informed the distinction between complete contact ranks and output controls. This work tests maximum coherence across a fixed network. No code or data arrays were copied from the earlier PSI report.
- usesdoi:10.18112/openneuro.ds005670.v1.0.0
The two resting SEEG records analyzed in this publication are from this exact OpenNeuro dataset version.
- doi:10.1093/braincomms%2Ffcae165
Background on how re-referencing and connectivity choices affect intracranial EEG network results.
- doi:10.1016/j.neuroimage.2014.01.060
Background on permutation inference and maximum-statistic correction across multiple tests.
- doi:10.1016/S0167-2789(00)00043-9
Background on surrogate tests and the need for random controls to preserve the target null model.
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesIndependent uniform contact and cell phases with fixed magnitudes give exact joint covariance and a five percent network error bound for fixed real SEEG montages
We reused both original-reference contact spectrum files, fixed contact and window selection, source metadata, provenance, and EDF retrieval and extraction code. Its complete-null contact rank argument informed the calibration control. We freshly rebuilt the spectra from both raw recordings. The present results use signed imaginary sums and new joint covariance tests.