In one pRF run of ds004194 subject p05, 25 SEEG contacts and 146 ECoG contacts did not identify bar position above chance from one sweep pass without averaging

This single-run pilot tested whether depth (SEEG) contacts carry bar-position information that one pass of a population receptive field (pRF) bar sweep can identify, compared with surface (ECoG) contacts in the same patient. We analyzed the dataset authors’ ECoGBroadband derivative (512 Hz; common average reference per electrode group) for subject p05, session nyuecog01, pRF task, run 01, from OpenNeuro ds004194 version 3.0.0. The analysis included 25 good SEEG and 146 good ECoG contacts. Each of 56 bar steps occurred twice. No SEEG contact and 5 ECoG contacts (3.4%) had reliable pass-to-pass profiles (p < 0.01). Template matching produced mean absolute errors of 8.88 steps for SEEG and 9.90 for all ECoG; chance error was 9.31 steps, and the SEEG shuffle test gave p = 0.18. The SEEG result was not different from matched 25-contact ECoG subsets; 8% of subsets had lower error. Thus, no contact set identified bar position above chance from one pass. This finding is limited to one subject and run, with no trial averaging, and contacts were not selected for anatomy or visual response. The SEEG contacts were not checked against an atlas, so the result does not test SEEG contacts known to be in visual cortex.

SEEG · ECoG · visual cortex · bar-position decoding · pRF · broadband power · single-run analysis