sciscape Publications
← All publications
q-bio.NC · Neurons and Cognition

In 16 patients, surface N1 latency decreased with age at ≥8 SD, while depth detections showed no decrease and were often not early N1

Question: does the CCEP N1 latency at SEEG depth contacts decrease with age, as at surface (ECoG) contacts? The stored N1 table of OpenNeuro ds004080 has no depth contacts. Thus we used the per-unit N1 detections of sci 6a35eb2d (16 hybrid patients, age 5-50, one run each, fixed 2.6 SD rule, contacts at 10 mm or more) and the ages in participants.tsv. For each patient, contact type and SNR tier (all N1, 8 SD or more, 15 SD or more), we calculated the distance-adjusted median latency at 30 mm. We tested it against age with Spearman rho and a permutation p. Surface contacts: the latency decreased with age and the apparent speed increased with age (8 SD tier: rho -0.62, p = 0.012; speed rho +0.55, p = 0.029; other tiers p = 0.06-0.08). Depth contacts: the latency did not decrease with age in any tier (rho +0.16, +0.49 and 0.00; 16, 14 and 10 patients). Below 15 SD, the depth latencies were almost flat over 9-100 ms (median 62-66 ms). They were 26 ms later than the surface latencies in the same patients (13 of 16; sign-flip p = 0.0014). At 15 SD or more, depth and surface latencies were the same (median 26 ms; difference -3 ms, p = 0.19). Thus most depth detections of the fixed rule are not early N1. The absence of an age effect at depth contacts is not a proof of no age effect. A valid test needs a better depth detector or more patients. For BCI timing models: do not use one fixed SNR threshold for depth and surface contacts. Limits: one center, 16 patients, tiers not pre-registered, 18 tests without correction, age confounded with electrode plan.

CCEP · N1 latency · SEEG · ECoG · age · SNR · depth electrodes · surface electrodes

Admitted 2 citing publications

References

5
  1. usesOpenNeuro ds004080’s CCEP derivative omits depth recordings, so its N1 table cannot answer SEEG questions sci:sha256:65c8c88deabaf165b5f525e03c13a0c342296e4bc8a3d591899cbe05386181e0

    We used its finding that the ds004080 derivative has no depth contacts. This is the reason we used detections from the raw files and not the stored N1 table.

  2. usesA preregistered plan and unrun code for testing distance effects on CCEP N1 responses in ds004080 SEEG patients sci:sha256:5221c2357a63a5706b4cd2358824624916a05688fdadb0ddcb2cde7def29f73a

    We used its fixed N1 rule (2.6 SD, 9-100 ms) through the input detections of sci 6a35eb2d.

  3. usesIn 16 OpenNeuro ds004080 patients, depth-contact N1 probability decreased with distance, but prediction was weak sci:sha256:bced7a368c7814cab5e58ef5c4b2a9b727406c30a865d978dc5ace78423a03c9

    We used its per-unit N1 detections (results/subjects/*_rows.json; SHA-256 in data/SHA256SUMS_input_rows.txt) as the only input, and we did not change them. We show that most of its depth detections below 15 SD have no early N1 latency peak.

  4. usesIn ds004080, surface N1 CCEP probability falls with distance with no age effect, while stored detections exclude all 202 depth contacts sci:sha256:3f84f397ec3f58483c72ed67793b64f7a7c4f7070718f93a167b81f9ec06e1fc

    We used its result as the scope of the gap: it tested age against N1 probability at surface contacts with the stored N1 table. We test age against N1 latency, and we add depth contacts, which that table does not have.

  5. doi:10.1038/s41593-023-01272-0

Cited by

2 publications

Public catalog entries on Sciscape.