Depth stimulation recall cost did not significantly habituate across encoding lists, but a full loss remains possible

We tested whether the recall cost of open-loop 50 Hz stimulation through SEEG depth contacts during word encoding becomes smaller over repeated stimulation lists. The outcome was serial-position-matched recall of stimulated words minus recall of non-stimulated words in the same stimulation lists. In FR2 (29 subjects), the overall cost was -1.9 percentage points (95% CI -3.8 to -0.2; p = 0.048). The pre-specified late-minus-early contrast was +1.0 points (95% CI -3.1 to +4.8; p = 0.65), so the study did not find habituation. The confidence interval does not exclude a full loss of the cost in the late half, and the test cannot show that the cost is stable. In catFR2 (10 subjects), the late-half cost was larger, with a non-significant contrast of -6.6 points (95% CI -12.7 to +0.6; p = 0.10). The per-list slope and across-session tests also did not show habituation. In FR2, the cost was -3.2 points for the first word of a two-word stimulation block and -0.7 points for the second; their difference was not significant (p = 0.16), and stimulation of the first word may affect the second. Analyses use behavioral and stimulation labels from events-derived per-word tables, not neural signals. The sample was small, the early and late halves also differed in time-on-task and fatigue, clinicians selected sites and amplitudes, and secondary tests were not corrected for multiple comparisons.

SEEG · word encoding · free recall · open-loop stimulation · recall cost · habituation · RAM FR2 · catFR2