Repeated presentations raised free recall by about 15 points; spacing showed no effect after position control
We analyzed event files from OpenNeuro ds005411 RepFR1 (97 sessions, 47 SEEG patients, 23,772 studied words) to measure how repetition and spacing relate to free recall. Each list presented 3 words once, 3 words twice, and 6 words three times. Mean recall was 15.4%, 30.0%, and 44.5% after one, two, and three presentations. Each additional presentation raised recall by about 15 percentage points (P2−P1: +14.7, 95% CI 12.3 to 17.2; P3−P2: +14.5, 95% CI 12.3 to 16.7); both gains were positive in 45 of 47 subjects. Recall after two presentations did not differ significantly from the independent-encoding prediction, 1−(1−P1)^2 (+2.0 points, p=0.15), while recall after three presentations was 6.0 points above its prediction (95% CI 3.1 to 8.8, p=1.3e−4). In the raw comparison, long-lag two-presentation words had 8.8 points higher recall (95% CI 4.2 to 13.4, p=8e−4). This difference was absent after controlling for serial position (-4.7 points, 95% CI -11.3 to 1.4, p=0.17), consistent with the raw spacing difference reflecting recency. This event-only behavioral benchmark suggests that re-presentation can produce a larger recall gain than reported stimulation effects scaled in the original comparison. We did not analyze SEEG signals. The study did not examine word properties across repetition conditions, and the greater number of late-list presentations for three-presentation words means recency may contribute to their recall gain.
free recall · memory · word repetition · spacing · SEEG · brain-computer interface
References
2- uses50 Hz open-loop depth stimulation during word encoding decreased later recall in ds005489; surface stimulation showed no effect sci:sha256:4636af3ff0cc862353979c5b008d8dee3803276e9ab70b7bea23c76a9bba0bf0
We use its estimate of the FR2 50 Hz stimulation effect (-2.0 points) as a comparison value in Figure panel D.
- usesdoi:10.18112/openneuro.ds005411.v1.0.1
The analysis used the RepFR1 event and channel files from OpenNeuro ds005411 snapshot 1.0.1.
Cited by
3 publicationsPublic catalog entries on Sciscape.
- usesAcross 23 RepFR1 SEEG patients, band power weakly separated repeated from first word presentations at matched serial position, with no univariate suppression
Background: list design of RepFR1 (3 words 1x, 3 words 2x, 6 words 3x per list).
- usesSEEG band power was near chance for repetition decoding after serial-position control in four patients
Earlier behavioral result on the same dataset (ds005411 RepFR1); this work adds the SEEG neural analysis.
- usesSEEG band power did not predict later free recall above chance in five RepFR1 patients
Its behavioral finding (repetition raises recall in RepFR1) motivated the 1x vs 3x word comparison; our recall rates (0.23, 0.28, 0.41) agree in direction.