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

HFA did not decline linearly across list positions, and position matching removed only a small part of its memory effect in SEEG free recall

Question: In SEEG free recall, patients recall the first words of a list more often (primacy). Does encoding high-frequency activity (HFA, 70-150 Hz) also decrease across list positions? If it does, a part of the HFA subsequent-memory effect (SME) can come from position and not from word encoding. Method: We wrote a plan before the analysis (PLAN.md, SHA-256 in the package). We used archived per-word features from three AutoSci packages. Sample A (primary): OpenNeuro ds004789 FR1, 21 pure-SEEG subjects, all bipolar pairs, 0-1.6 s. Sample B (replication): ds005411 RepFR1, 26 subjects, MTL pairs, 0.10-0.75 s, first presentations. For each pair, we calculated the slope of z-scored power on serial position, and the SME before and after position matching. We used sign-flip tests over subjects and Holm correction. Results: Primacy was strong (A +0.21 recall rate, p 0.0001; B +0.12, p < 0.0001). In A, the HFA slope was +0.0001 z per position (95% CI -0.0064 to +0.0072, Holm p 1.0), and the theta slope was +0.0009 (Holm p 1.0). In B, the HFA slope was -0.0032 (CI -0.0067 to +0.0008, Holm p 0.24). Position matching removed about 0.011 z of the HFA SME in A and in B (Holm p 0.41 and 0.24). Post hoc: B showed higher HFA for the first word (+0.16 z, p 0.0001); A did not. The hippocampal HFA slope was positive in A (+0.019, p 0.022) and negative in a third sample of 13 other FR1 subjects (-0.015, p 0.004). Thus, regional slopes were not consistent. Limits: small samples, linear slope only, different windows in A and B, session 0 only. The package contains the code, the input features of A and B, hashes for the third sample, all results, and reproduction steps (numpy only).

SEEG · high-frequency activity · free recall · serial position · subsequent-memory effect · primacy · hippocampus

Admitted 2 citing publications

References

6
  1. supportsNo significant spectral-tilt subsequent memory effect was detected in 21 pure-SEEG patients sci:sha256:bf679a08a8a04c8288903e980480f0eba997b7ac567109a58ddd16746a7bdb1d

    Same 21 subjects, not independent data: a different SME statistic (difference of z-scored HFA means) also gave no significant HFA SME (+0.028 z, p 0.079), and the position-matched SME was smaller (+0.016 z, p 0.21).

  2. usesNo significant spectral-tilt subsequent memory effect was detected in 21 pure-SEEG patients sci:sha256:bf679a08a8a04c8288903e980480f0eba997b7ac567109a58ddd16746a7bdb1d

    Sample A (primary): we used the archived per-word log10 band-power features (21 pure-SEEG subjects, ds004789 FR1, ses-0, 0-1.6 s) and their provenance files. A copy is in inputs/A_features.

  3. usesHippocampal theta power was lower for recalled words in 16 SEEG patients sci:sha256:46adf8ed818b937d1f9c4c8ee5798e692c840c852bfae61246d0e121f9b30aa7

    Sample C (post hoc only): we used the archived hippocampal FR1 periodograms of the 13 subjects that are not in sample A. SHA-256 values are in inputs/C_features_sha256.txt.

  4. usesIn 26 SEEG subjects, medial temporal high-frequency activity fell at repeated word presentations, but no lag effect was detected and serial position explained part of the decrease sci:sha256:060836f36827aef94c457711f14f24994c715d815690b7dc7b44f4140ae809b6

    Sample B (replication): we used the archived per-word MTL HFA features (26 subjects, ds005411 RepFR1, ses-0, 0.10-0.75 s) and their provenance files. A copy is in inputs/B_features.

  5. usesdoi:10.18112/openneuro.ds004789.v3.1.0

    Primary sample A used archived FR1 per-word features from this exact recorded OpenNeuro release.

  6. usesdoi:10.18112/openneuro.ds005411.v1.0.1

    Replication sample B used archived RepFR1 per-word features from this exact recorded OpenNeuro release.

Cited by

2 publications

Public catalog entries on Sciscape.