A shared SEEG reader scored 47.6% balanced accuracy in six people on one film with test people and sections excluded
We tested a reader with 12 inputs from SEEG power in six people who watched one film. Training used neither the test person's data nor the test section. The first minute set input scales without labels. Mean balanced accuracy was 47.6% on 3120 test windows. A descriptive interval over people and sections was 42.5% to 50.8%. The fixed reader did not show performance above chance. An added exploratory check used personal models with eight selected contact pairs. It scored 59.9% on the same windows. This check changed input count and contact selection. It does not isolate an effect of contact identity. The package has features, source checks, saved models, code, and figures. The labels describe film blocks. They do not describe intended speech. The study did not test brain stimulation.
SEEG · brain-computer interface · cross-person transfer · music and speech blocks · balanced accuracy · stereoelectroencephalography
References
3- usesFull section tests did not detect a mean block accuracy change after 96–104 Hz bin removal in six SEEG recordings of one film sci:sha256:0336278385fc639046cd9956fe37aeb1c2151076aa6918a4aca321eeb8cdc8e5
This study uses the six-person SEEG cohort, original four power bands, source metadata, voltage audit samples, and extraction code from this publication. It tests a shared reader with person and section exclusions. It adds a separately declared personal contact check.
- usesdoi:10.18112/openneuro.ds003688.v1.0.7
The study used the six-person SEEG recordings in OpenNeuro ds003688 version 1.0.7; the recorded source metadata identify this exact version and its CC0 license.
- doi:10.1038/s41597-022-01173-0
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesEarly sign consistency gave mixed later error changes with eight pairs in ten SEEG recordings of one film
This publication supplied the source metadata, six-person four-band features, and raw extraction recipe. We adapted the recipe, removed the six-person cap, independently retrieved all ten full raw recordings, and exactly matched the six reused feature arrays.
- usesThree early sections of one film reduced mean later probability error in six SEEG recordings with eight selected pairs and equal sample counts
This package supplied the six exact SEEG feature arrays, source metadata, raw audit evidence, and eight-pair reader rule. We fitted new models with equal calibration sample counts and a fixed later test.