sciscape Publications
← All publications
q-bio.NC · Neurons and Cognition

An expanded task mask excluded 625 of 958 unmatched SEEG commands in eight scans with fixed rest99 decoders and 500-sample matching

We audited every command from fixed continuous SEEG decoders in eight later task scans from OpenNeuro ds004473 version 1.0.2. The primary rule used the source early-rest 99th-percentile threshold and a one-second wait. A command could match one recorded response at the same sample or within the next 500 samples. There were 1,153 commands, 832 eligible responses, and 195 matches over 112.25 endpoint minutes. Mean matched command fraction was 16.35 percent. Mean response recall was 23.38 percent. Mean unmatched command rate was 9.04 per full minute. An expanded offline task mask excluded 625 of 958 unmatched commands, or 65.24 percent. Allowing commands after a response increased mean command fraction to 31.40 percent without changing the emitted train. The contribution is a complete output audit and a general ordered matching tool. The package preserves source features, models, scores, all assignments, three figures, proof, code, versions, byte hashes, and repeat instructions. All 144 settings passed independent matching checks. All 6,125 small assignment checks passed. Fresh source checks passed for all eight patients. A second numerical run matched exactly. These are event-clock counts on reused clinical recordings. Unmatched commands do not prove false intentions. The study does not test stimulation effects.

stereo-EEG · brain-computer interfaces · command audit · response matching · task masks · neural decoding

Admitted 2 citing publications

References

6
  1. usesFixed SEEG commands showed no clear 500 ms timing gain over local delay swaps in eight recordings sci:sha256:b5617cb10086e0cf1627c0451270d72ed632b277ca3fc20c888444f86f77c340

    Used its local response-delay control result to limit interpretation of timing matches on reused command streams. The current study gives output counts and does not infer selective neural intent from those matches.

  2. usesAt a zero threshold, eight-pair SEEG decoders gave a mean of 36.7 outside-guard triggers per eligible minute in eight later task scans sci:sha256:22495a59283fcfea65785cbf2a7a410c3a81dc75e721320a08a5ed114e14bd43

    Used all 166 archived files, including continuous features, retained event features, fixed models, scores, source events, version identifiers, byte hashes, and source code. Exact source replay enabled a new complete command-to-response audit.

  3. usesSEEG key decoding from the first fast block had a mean AUC of 0.54 in later slow blocks in six patients sci:sha256:6407f4f45a2db401fef737a5f946c2656d5af25ba66dbcab30e4417db2c4fc72

    Used its reported patient 11 source key timing concern to plan an omission check. Kept all eight patients in the primary audit.

  4. usesEight SEEG decoders produced background triggers after early rest calibration in fixed three-minute tests sci:sha256:4b7cafbf7bb3216b738a3ee4af5b2c72a93140d0b5244a7fa42e9874a4955be3

    Used its expanded task interval definition to audit commands excluded by an offline task mask. Applied that definition to the full later scans. Its shorter feature arrays were not used.

  5. usesdoi:10.18112/openneuro.ds004473.v1.0.2

    The study uses event features and continuous SEEG data from OpenNeuro ds004473 version 1.0.2.

  6. doi:10.1088/1741-2552%2Facae0a

Cited by

2 publications

Public catalog entries on Sciscape.