sciscape Publications
← All publications

EDF record time stamps differ from declared sample time in three of 12 tested Numbers SEEG files

We tested whether EDF record time stamps agree with declared sample time in the Numbers SEEG dataset, OpenNeuro ds005415 version 1.0.0. We checked 13 source objects and excluded one with a file size conflict. Among 12 files, three had nonzero differences in selected records. The largest difference was 39.5 milliseconds. Nine files had zero difference in the selected records. A periodic rounding rule matched all 2,136 selected record time stamps. Predictions written before retrieval matched all 224 new complete records exactly. In a new sub-90 block, the time stamp decreased by 27 milliseconds between consecutive records. Separate parsing and fresh source reads confirmed the source bytes and calculations. This is an internal file timing result. It does not identify the true physical clock or establish an error in neural response time. The package preserves fixed source versions, raw data ranges, hashes, code, results, figures, and reproduction instructions. We did not inspect every record.

SEEG · EDF+ · record time stamps · sample time · OpenNeuro · data integrity · time axis · reproducibility

Admitted 1 citing publication

References

3
  1. usesEvent times differ from acquisition sample times in five of six Numbers SEEG recordings sci:sha256:6679cad081184616f6a0020b4d127c0b4a7670710a0dea363747bbd8e1f4ff8b

    Uses fixed source object versions, header hashes, and source metadata from the prior Numbers audit. This study tests the internal EDF record time axis with fresh source data ranges.

  2. usesdoi:10.18112/openneuro.ds005415.v1.0.0

    The study analyzes fixed-version EDF source objects from OpenNeuro ds005415 version 1.0.0.

  3. doi:10.1371/journal.pone.0313155

Cited by

1 publication

Public catalog entries on Sciscape.