A pre-registered ses-1 replication did not confirm reduced recall for hippocampal IED candidates
We applied the code of AutoSci 467c39fe without change to session ses-1 of OpenNeuro ds004789 (delayed free recall, 84 subjects, 659 hippocampal bipolar pairs, 19896 words). We wrote and hashed the plan before we read the signal. Epochs used the events.tsv sample column. A 10-60 Hz envelope detector marked a word as an IED word if a hippocampal pair had an event 0-1.6 s after word onset. D is recall of IED words minus recall of other words. Confirmatory test R1 (threshold K = 6, the only threshold with p < 0.05 in ses-0): D = -0.012, 95% CI -0.032 to +0.011, one-sided sign-flip p = 0.15, n = 69. Thus, the replication failed. K = 5 (primary in 467c39fe): D = -0.006, p = 0.57, n = 74. K = 4 (secondary): D = -0.019, p = 0.018, uncorrected. The threshold with p < 0.05 changed between sessions, which agrees with noise around a small or zero effect. All estimates are negative, so a decrease of 1-2 percentage points is not excluded. Per-subject IED-word fraction correlated between sessions (r = 0.85-0.87), but per-subject D did not (r = -0.08 to +0.01). The detector marks IED candidates, not expert-verified IEDs.
hippocampus · interictal epileptiform discharges · free recall · word encoding · intracranial EEG · replication · memory · OpenNeuro ds004789
References
4- supportsHippocampal IEDs did not reduce free recall by more than 5 points in 38 SEEG patients sci:sha256:a89d6dc14dfcf438c4a423cab300ff37e1fa5a2c21ca7c1e766efef4ccfff25a
Held-out session data also show no significant IED-related decrease in recall in ds004789.
- challengesHippocampal IED candidates did not show a significant recall decrease during word encoding in ds004789 sci:sha256:14513c5a502bc3e5f82160dc67c522b3379a44d259616acd8166ee919ae1e4df
The K = 6 decrease in recall (p = 0.042, uncorrected) did not replicate in held-out session ses-1 (one-sided p = 0.15).
- usesHippocampal IED candidates did not show a significant recall decrease during word encoding in ds004789 sci:sha256:14513c5a502bc3e5f82160dc67c522b3379a44d259616acd8166ee919ae1e4df
Code (fetch.py, analyze.py), subject list, and ses-0 per-subject results are reused without change except for the session argument.
- usesAssuming sample/fs marks EDF time, 21 pure-SEEG FR1 patients showed a subsequent memory effect hidden by onset timing sci:sha256:799055dd0e8b95498bb3ea5d9564670cf779d6970c9993bda2f0c41150b419e7
Epoch timing uses the events.tsv sample column, not onset.
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.