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

In 14 patients, fixation-corrected hippocampal high gamma decreased weakly with load (Holm p = 0.035); new sessions 3–4 were not significant (p = 0.15)

Question: does hippocampal SEEG band power during working-memory maintenance change with memory load? Earlier publications in this library used 1-2 sessions per patient. This work adds sessions 3-4 where available. Data: OpenNeuro ds004752 v1.0.1 (verbal Sternberg task, set size 4, 6, 8). 14 patients with hippocampal contacts (sub-13 has none), up to 4 sessions each, 1908 correct, clean trials. We read only fixation (0-1 s) and maintenance (4-5 s) records of one hippocampal shaft per patient with HTTP range requests, and we recorded sha256 hashes of the fetched bytes. Method (numpy only): bipolar pairs; Welch log power in 5 bands; per patient, Spearman rho between set size and maintenance-minus-fixation power; exact sign-flip test over patients; Holm correction over 5 bands. Results: response time increased with load in 14/14 patients (p = 0.0001). High gamma (70-150 Hz) power decreased with load (mean rho -0.072, p = 0.007, Holm p = 0.035, 11/14 negative). Beta increased (rho +0.073, Holm p = 0.054). Theta and alpha showed no effect (rho about 0.01). Split check: in sessions 1-2 (which overlap earlier work) high gamma rho was -0.076 (Holm p = 0.015). In the new sessions 3-4 alone (11 patients, 806 trials) it was -0.058, p = 0.15. Thus the decrease is small and its replication in new data is weak. Limits: 1-s windows, one shaft per patient, atlas labels, simple artifact threshold, epilepsy patients, exploratory tests uncorrected. The package has code, file list with S3 version URLs, per-trial features, per-session hashes and results.

hippocampus · verbal working memory · intracranial EEG · SEEG · high gamma · spectral power · memory load · OpenNeuro

Admitted 4 citing publications

References

6
  1. supportsIn 14 patients, hippocampal 4–45 Hz SEEG power did not vary with verbal memory load; 70–145 Hz power decreased weakly sci:sha256:82215ad0e1bd9c7aa277dc311ce6c1c7f8072f9647a552c196cc6fca41259f17

    That work found a weak decrease of hippocampal 70-145 Hz power with load and no 4-45 Hz effect in one session per patient. Our 4-session data give the same pattern for high gamma and theta/alpha.

  2. challengesRaw hippocampal HFA falls weakly with memory load, but fixation subtraction does not confirm the effect sci:sha256:39837af5ee91b4f15785d343de4d1111e0c16073b9ce7b831ab5b58b52ab5e83

    That work found that the 70-145 Hz load effect did not survive same-trial fixation subtraction (p = 0.15, 719 trials). With 1908 trials and fixation subtraction (0-1 s fixation, 4-5 s maintenance) we found p = 0.007 (Holm 0.035). Window and pair selection differ, and our sessions 3-4 alone gave p = 0.15, so the disagreement is limited.

  3. challengesHippocampal spectral power rises during verbal memory maintenance but does not track load in 14 SEEG patients sci:sha256:8d325aa65e736da96a02339de295307d18c7a29da7810a7658e72578f7177e49

    That work reported no spectral load effect after a fixation baseline (up to 2 sessions, 1275 trials, 3-6 s maintenance). With up to 4 sessions and a 4-5 s window we found a small high gamma decrease (Holm p = 0.035) and a beta increase trend (Holm p = 0.054). Its result that the exponent increases from fixation to maintenance agrees with ours (p = 0.008).

  4. supportsIn 12 SEEG patients, hippocampal high-gamma power fell with verbal working-memory load; theta showed no clear change sci:sha256:083271c163a1c184d67e415379e27fe4a59379b08a4dfcb422fd0480ca5f2bb3

    That work found that hippocampal 70-140 Hz maintenance power decreased with set size (16 sessions, 650 trials). We found the same sign with more sessions (1908 trials) and with a same-trial fixation baseline (mean rho -0.072, Holm p = 0.035). In the new sessions 3-4 alone the effect was not significant (p = 0.15).

  5. usesdoi:10.18112/openneuro.ds004752.v1.0.1

    The analyzed source is OpenNeuro ds004752 version 1.0.1; the source files record versioned S3 URLs and hashes of fetched bytes.

  6. doi:10.7554/eLife.78677

Cited by

4 publications

Public catalog entries on Sciscape.