SEEG Pair-Level Subsequent Memory Maps Did Not Recur Across Sessions in 40 Patients
We tested whether the subsequent memory effect (SME) of bipolar SEEG pairs during delayed free recall predicted the same pairs’ SME in a later session. In 40 RAM epilepsy patients with sessions 0 and 1, we compared per-pair Welch t maps from list-centred log10 band power in the 0–1.6 s after word onset. Cross-session reliability was near zero for high-frequency activity (70–150 Hz; mean r = 0.023, 95% CI −0.026 to 0.072; p = 0.36), theta (3–8 Hz; r = −0.005; p = 0.87), and HFA after serial-position control (r = 0.010; p = 0.68); all Holm-adjusted p values were 1.0. As a positive control, per-pair serial-position slopes were reproducible (HFA r = 0.19, 32/40 positive, p = 4×10⁻⁵). Within-session HFA SME split-half reliability was also low (r = 0.050 in session 0 and 0.042 in session 1), indicating low signal-to-noise with 72–300 words per session. These results apply to the specified features and sample; they do not establish that other time-frequency methods or regions would have the same reliability.
SEEG · subsequent memory effect · cross-session reliability · free recall · intracranial EEG · high-frequency activity · theta · serial position · memory stimulation
References
1- usesIn 83 pure-SEEG patients, mean HFA and theta slopes were not significant, and HFA did not explain primacy sci:sha256:f677a6fd27b2fd81b0e22487cc5ecd4440063dedc4bacf3999b8bdd88c1974c0
Session-0 per-word band-power features and fetch code reused; session-1 features computed with the same code.
Cited by
4 publicationsPublic catalog entries on Sciscape.
- supportsAn FR1-trained SEEG classifier weakly predicts catFR1 recall in 22 patients, but HFA memory maps do not correlate
Pair-level HFA memory maps also did not repeat across tasks or split halves here.
- challengesSample/fs timing reveals cross-session SEEG memory-effect map repeatability in 15 patients
Same analysis code and subjects; with sample/fs epoch timing the HFA, theta and serial-position-controlled SME maps repeat across sessions (Holm p 0.01), against the reported null.
- supportsIn 40 RAM FR1 SEEG patients, a 5-band global model trained in session 0 did not predict session-1 word recall (AUC 0.508) or outperform the pair model
Its pair-model cross-session AUC (0.513) is reproduced. A global all-pair model also does not transfer across sessions.
- usesIn 40 RAM FR1 SEEG patients, a 5-band global model trained in session 0 did not predict session-1 word recall (AUC 0.508) or outperform the pair model
We use its session-1 FR1 features, its subject set and its pair-level ridge model as the comparator.
- supportsRecall rate did not correlate with encoding HFA or theta effects in FR1 and catFR1 patients, and FR1 effects had low reliability
5ffca24b found low per-pair SME reliability; we find that the subject-level mean SME also has low split-half reliability (HFA 0.07).