Two-second SEEG windows did not establish better beta imaginary coherency rank agreement above phase controls than one-second windows in 21 patients
We tested whether two-second SEEG baseline windows gave more repeatable beta imaginary coherency channel ranks than one-second windows. The test used 34 public recordings from 21 patients in OpenNeuro ds006519 version 1.0.0. The channels, window centers, and frequency bins were fixed. Pairs with a shared contact were excluded. The paired mean gain above phase controls was 0.040 in Spearman correlation. The 95% patient bootstrap interval was -0.037 to 0.123. The exact two-sided sign-flip test gave p = 0.3601. The result did not establish a gain or equal performance. Independent-noise tests showed positive absolute edge weights that fell with longer windows. Thus, a lower weight alone does not prove a change in brain connections. This is an exploratory extension in a reused patient group. It did not test a decoder or stimulation target. The package contains three figures, baseline arrays, source hashes and metadata, all control scores, source code, independent checks, and reproduction instructions. A second complete analysis reproduced the saved numeric results.
SEEG · imaginary coherency · channel rank agreement · phase controls · beta band · spectral estimation · finite sample bias
References
5- usesA test of one-second imaginary coherency did not establish beta SEEG rank agreement above phase controls in 21 patients sci:sha256:2509955d18c814560a6ea65ffa401de79c6f685e87d0edc4c1923841e384ea0e
We used the saved ds006519 source manifest and adapted its contact parser, retrieval code, and spectral and rank routines. We downloaded and checked all raw source files again. The earlier one-second result was known before the local plan. This work adds a paired one-second versus two-second test with fixed spectral resolution, matched window centers, and paired phase controls in the reused patient group.
- usesdoi:10.18112/openneuro.ds006519.v1.0.0
Source dataset, version 1.0.0; the publication identifies the public recordings and reports the version-specific source.
- doi:10.1016/j.neuroimage.2026.121822
- supportsdoi:10.1007/s10548-017-0609-4
Supports the motivation to test frequency-domain connectivity estimates at each data length; it does not predict the result in this SEEG group.
- doi:10.1016/j.clinph.2004.04.029
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesRemoving links with shared contact labels gave no clear gain in later gamma scores in 34 SEEG post intervals
This prior study was read before the local plan. Its test of rank agreement and phase controls informed the need for a separate-sample score. No rank values from it enter the new analysis.