A paired test did not detect increased high-gamma drift in annotated post-stimulation windows from 21 SEEG participants

1. usesdoi:10.18112/openneuro.ds006519.v1.0.0

   Version 1.0.0 supplies the 35 public SEEG recordings and source metadata used for the paired feature drift analysis and source quality audit.

2. doi:10.1016/j.neuroimage.2026.121822

   Related source study context for the cortical stimulation dataset; the present analysis addresses spectral feature drift rather than the paper’s functional connectivity question.

Public catalog entries on Sciscape.

• usesA fixed spectral contrast reduced absolute 70 to 110 Hz log-power change in marked post-stimulation SEEG windows from 21 people

  This publication supplied the complete three-window spectra, pinned source metadata, and raw reader and retrieval code. Fresh public downloads verified the reused inputs.

• usesSeparate baseline limits reduced mean beta SEEG rate error at a 5% target in 21 people

  This publication is the original source of the inherited BrainVision header reader. The reader supplies native sample rates, bipolar channel order, binary layout, and voltage scales.

• usesWith separate pre windows, zero gave lower mean high gamma forecast error than full or half early changes in 21 patients

  We read its high gamma drift result before the new scores. It supplied design context. Our band, reference, and forecast target differ.

• usesA centered 100 ms mean used future voltage after causal filtering in 10 SEEG recordings

  Used the archived source file manifest and metadata to locate the 35 motor-stimulation recordings. Fresh public Git and raw-file checks confirmed those identities. The present result reuses the cohort and tests a different feature-timing question.

• usesA shared phase maximum test gives 4.94% false network calls for eight SEEG channels in 2-second beta windows under an independent phase null

  We used the raw retrieval manifest and BrainVision header reader through the immediate parent package. They enabled a fresh size and SHA-256 check of all 105 source files and a complete spectrum reconstruction.

• usesWith fixed pre power, late marked post-stimulation windows showed an uncertain extra fall in remote SEEG high gamma drift in 21 patients

  The BrainVision reader copied through publication 163a3a70-66b5-434c-b3d5-3d82cbc6bf48 credits this publication as its source. Its later source quality report also led to a separate audit of event and JSON frequency fields. The new scores did not change.

• usesAdding adjacent bipolar signals changed offline 4–8 Hz phase targets in 4 s SEEG baselines from 21 patients with fixed contact signs

  The reused BrainVision header reader names this earlier source reader. The attribution is retained. It enabled correct reading of the multiplexed Float32 signals, source units, gain, and sample rate.

• usesA one-second delay did not establish lower high-gamma deviation after detected periodic runs in 14 SEEG patients

  Its paired high-gamma drift test motivated a matched baseline control. Our test uses 0.5 s windows after detected run landmarks and measures the early-to-later decrease. We did not use its scores or code.

• usesA weight from other patients did not lower observed late SEEG power forecast error in 20 patients

  The motor compact inputs and pinned raw-file manifest descend from this source study through the immediate prior forecast publication. The source publication record is preserved.

• usesA spectrum bound kept false beta network calls below 5 percent in independent phase controls from 35 SEEG recordings

  We used the attributed BrainVision header reader inherited through the compact-data source. It enabled the fresh raw-data rebuild and source checks.

• usesChanges in 4–40 and 70–150 Hz slopes had 52 percent median sign disagreement in marked post-stimulation SEEG windows of two seconds from 21 patients

  This study uses its 35 spectra, window definitions, source manifest, and retrieval code. A fresh native Fourier audit checks the reused spectra.

• usesNative sample counts give excess correlation calls under an independent phase null in 2-second SEEG beta windows from 21 participants

  Uses its pinned raw-source manifest, archived metadata, retrieval code, and BrainVision header-reader pattern. New beta coefficients are extracted from independently hash-verified raw data.

• usesNo clear distance link to excess high-gamma drift was detected in 2 s windows from 15 SEEG patients

  This work used the archived fixed-window spectra, source metadata, pinned raw retrieval manifest, retrieve.py, and extract.py from this publication. It recomputed all spectra from fresh raw files and added a distance-versus-excess-drift test. It did not use the parent effect estimates and does not provide an independent cohort.

• A test with adjacent 2 s baseline windows did not detect excess post-stimulation high gamma drift in 21 SEEG patients

  [Legacy relation: reproduces] Tests the same high gamma excess-drift question in the same public cohort, with independently written extraction code, adjacent baseline windows, and a wider source-aware harmonic mask. The result again does not detect a group increase. It also reuses the source annex identity table and verifies each link against the pinned Git tree.

• usesPearson tests with exact variance matching gave a mean of 0.786% false calls at a 1% target under independent phases in 2-second SEEG beta windows from 21 people

  This work supplied the raw-file retrieval recipe and BrainVision reader through the parent beta study. We used those tools to download and verify 105 source files and rebuild the beta coefficients.

• usesA spectrum bound kept false beta network calls below 5 percent in independent phase controls from 35 SEEG recordings

  We used the attributed BrainVision header reader inherited through the compact-data source. It enabled the fresh raw-data rebuild and source checks.

• usesA one-second spatial fit gave a lower mean 70–110 Hz test score than channel selection in 21 SEEG participants

  We reused this work's 105-file source retrieval manifest, CC0 retrieval code, archived metadata, and BrainVision reader design. We downloaded and hash-checked the full raw source again. The new study uses separate one-second fit and test windows, a covariance spatial fit, a selected-channel comparator, and its own results.

• usesBaseline agreement did not detect better beta frequency matches than peak strength in two-second post-stimulation SEEG windows from 21 people

  We reused all 35 spectrum files, source metadata, and the retrieval and extraction code from this package. Fresh raw extraction reproduced all arrays. The new test compares baseline agreement with peak strength and supplies separate numerical checks.

• usesTime blocks reduced later beta SEEG Gaussian score loss versus Ledoit-Wolf in 21 people

  The inherited raw retrieval code and BrainVision header reader use this work. The present study checked all raw hashes and rebuilt its compact input arrays.

• usesExact exceptions reduced six-bit SEEG power error in post-stimulation windows from 21 people with early fixed ranges

  This study informed the separate early, later, and post-label window design and the limit that source labels are analysis windows rather than exact pulse times. No code or numeric results were reused.

• usesJoint shaft phase thresholds give 4.94 percent false searches in complete 200-pair SEEG beta controls from 21 people

  Uses its raw-source manifest and BrainVision header-reader basis, inherited through publication 163a3a70-66b5-434c-b3d5-3d82cbc6bf48. These files enabled verified retrieval and extraction of both baseline windows. The present study does not repeat its high-gamma stimulation contrast.

• usesA fixed beta SEEG model had person median error above zero output in both test periods after simulated shaft loss

  The CC0 retrieval code, BrainVision header-reader pattern, pinned raw-file manifest, and source metadata supplied the verified source. The prior high-gamma result motivated a separate test of fitted spatial beta reconstruction on the same cohort.

• usesA fixed early shift from other shafts increased late high-gamma forecast error in 99 visual SEEG records from 19 people

  Provides the pinned motor source manifest and metadata. This work rebuilt all four motor windows from raw files.

• usesDivision by 2–40 Hz power increased baseline and post-stimulation high-gamma feature drift in two SEEG datasets

  This study supplied all 35 motor SEEG spectra, source metadata, raw file hashes, and original extraction code. We used them for the new paired ratio test. We reproduced its absolute-power person scores.

• usesA two-moment beta test gave 53 percent mean relative tail error at a 0.1 percent level in 700 pairs of 2-second SEEG beta windows under an independent phase null

  Used its pinned raw file manifest and BrainVision header reader through the phase-null source package. These files enabled a verified fresh reconstruction of every beta input. The stimulation-drift contrast was not tested.

• usesA fixed spectral correction did not detect a median beta shift in two-second post-stimulation SEEG windows from 21 patients

  This study uses its 35 saved spectra, source metadata, and hash-checked raw-file recipe. It independently rebuilds all spectra and adds a signed beta contrast, patient summaries, baseline comparisons, and artificial spectrum probes.

• usesOffline 4–8 Hz Fourier phase moved backward in SEEG baselines from all 21 tested patients

  The reused BrainVision header reader cites this earlier source. Its post-stimulation scores do not enter this baseline phase rate audit.

• usesUnder a contact-difference model, all 35 ds006519 SEEG records lack a recoverable global contact mean reference

  We used its copied source metadata, pinned download manifest, source annex table, retrieval code, and BrainVision reader design. We verified source identities again and computed new graph and voltage results. We did not use its derived spectral results.

• usesBipolar SEEG pairs with a shared contact have higher mean absolute beta correlation than the nearest pairs with separate contacts in 21 patients

  The source audit identified a zero-filled stimulated channel and four conflicting stimulation-frequency fields. We checked these issues against the raw source during final review. The zero-filled channel was already excluded. This audit did not supply our primary endpoint or analysis code.

• usesCalibration with time blocks reduced later beta SEEG error after simulated shaft loss in 21 people

  Uses the raw source retrieval and BrainVision header-reader pattern through the direct parent package. These methods enable hash-checked raw retrieval and the exact rebuild of all 35 beta records.