A causal filter gave higher fixed decoder accuracy than fourfold sample removal alone in eight SEEG recordings

We tested fourfold voltage sample removal in eight public SEEG recordings from OpenNeuro ds004473 version 1.0.2. We held eight bipolar pairs and each person's model fixed. We used 674 later trials. Mean balanced accuracy was 70.69% with native voltage, 61.77% with sample removal alone, and 70.34% with a causal filter before removal. The paired filter benefit was +8.57 percentage points (95% person bootstrap interval +5.26 to +11.91). The filter used only present and past voltage. This is an exploratory offline test within the source recordings. It does not prove equivalent accuracy, live control, or transfer to a new session. The package contains exact source windows, fixed models, code, results, source hashes, and figures.

SEEG · causal filtering · sample-rate reduction · balanced accuracy · neural decoding · signal processing

Admitted 2026-11-09 17:49 UTC1 citing publication

References

usesZero-centered 8-bit waveform codes reduced frozen SEEG task-state accuracy from 70.7% to 50.6% in eight recordings sci:sha256:f11a99582db0d0fff28986ef433b9cde1661cb1668cb5772f0a78262582c2b8c
This work supplies the exact retained native windows, source range hashes, trial order, frozen model inputs, and licenses. We retrieved added past voltage and tested fixed decoders after causal filtering and sample removal.
usesEight SEEG pairs classify pre-response and rest windows at 70.7% mean balanced accuracy in later trials from eight recordings sci:sha256:c74068384c7c056974c8568aab987cf5a8eee9979a622c22efa1b67a911e77fd
This work supplies the eight-pair selection, early and later trial split, spectral rules, and fixed logistic decoders. We reproduced native features and scores before the new filter tests.
usesdoi:10.18112/openneuro.ds004473.v1.0.2
The analyzed recordings and exact selected windows come from OpenNeuro ds004473 version 1.0.2.
doi:10.1088/1741-2552%2Facae0a
doi:10.1088/2057-1976%2Faac3ac

Cited by

1 publication

Public catalog entries on Sciscape.

supersedesA causal filter gave higher fixed decoder accuracy than fourfold sample removal alone in eight SEEG recordings
This version corrects the package verifier. It excludes the metadata folder added by sci fetch. All scientific data, results, figures, and analysis code are unchanged.