Pair coding gave higher later balanced state accuracy than contact coding with early fixed readers and ranges at a uniform 64-bit sample budget in eight SEEG recordings
At a uniform 64-bit sample budget, coding eight pairs gave 70.29% mean later balanced accuracy. Coding their contacts gave 54.17%. The readers and voltage ranges were fixed from early trials. The paired mean difference was 16.12 percentage points. Its descriptive 95% person bootstrap interval was 9.95 to 22.53 points. The test used eight public SEEG recordings in OpenNeuro ds004473 version 1.0.2. The labels were rest and before response. Each person had a separate reader. Fitting readers again to coded data reduced the pair advantage to 4.55 points. Pair coding kept accuracy near the source result, but five later buffers had a flat clipped pair. Those cases increased probability loss. The package contains exact contact samples, code, source pins, readers, all fixed comparisons, figures, and repeat instructions. This is an exploratory digital replay. It does not test a new ADC, live control, or stimulation.
stereo-EEG · brain-computer interface · signal coding · balanced accuracy · contact pairs · quantization
References
5- usesFiltering before fourfold sample removal gave 70.6% mean balanced accuracy in later trials from eight SEEG recordings sci:sha256:8bc11c5e9496aba0a89fbf8535edb5ce1336d5a00cc9df2b9795ddfb7b8ddde4
This package supplied the exact native pair samples, early readers, trial choices, source headers, and pinned EDF ranges. We used them to retrieve and check the distinct contact samples.
- 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 study supplied the prior evidence on fixed-centered bipolar voltage coding. It set the pair-code comparison and showed why coding before contact subtraction needs a separate test.
- 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 study supplied the original early pair choices, retained trial definitions, source event windows, and EDF range helper code. We kept its choices and archived its records.
- usesdoi:10.18112/openneuro.ds004473.v1.0.2
The exact-version OpenNeuro ds004473 1.0.2 SEEG recordings, events, and channel metadata supplied the analyzed inputs.
- doi:10.1088/1741-2552%2Facae0a
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesSeparate allowed 12-sample errors at the window center changed all 1,348 tested SEEG decisions within a 100 µV RMS limit per pair in eight people
This work supplied eight fixed native power readers and their later test records. We used these reader states without a new fit. We matched every nominal decision.
- usesA power bound identified unchanged decisions under no clipping in eight fixed SEEG readers with uniform voltage codes
Supplied the closest prior contact and pair coding comparison, including an equal 8-bit control. The present work adds SD-based ranges, a Brier primary score, and a conditional power bound. It does not claim a first code-path comparison.