Pre-stimulus theta and high-frequency activity did not predict later recall in 59 SEEG participants

We tested a pre-registered prediction that hippocampal theta power in the 0.75 s before word onset is higher for words later recalled. We analyzed word-epoch byte ranges from OpenNeuro ds004789 FR1, session 0, using bipolar SEEG from 59 participants (51 with hippocampal and 39 with ventral temporal depth pairs). Within each participant, OLS controlled serial position, list, and recall of the previous word. None of four primary pre-stimulus tests was significant after Holm correction; the hippocampal theta mean was negative (-0.026 z, p = 0.143, Holm p = 0.426). Hippocampal high-frequency activity was also lower for later recalled words before correction (-0.031 z, p = 0.034; Holm p = 0.136). In secondary analyses, post-onset theta was lower for recalled words (hippocampus p = 0.0004; ventral temporal cortex p = 0.0001). Pre-stimulus high-frequency activity was lower when the previous word was recalled, which may reflect carry-over from that word. The pre-stimulus prediction was not supported. Results are limited to the specified 0.75 s window, frequency bands, atlas-defined regions, and the first participants processed before a fixed stop time. The source EDF release is uncertain; byte-range hashes and source URLs are preserved.

subsequent memory effect · theta · high-frequency activity · SEEG · free recall · hippocampus · ventral temporal cortex · OpenNeuro ds004789