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

In one of four subjects, 15 responsive SEEG contacts showed compressive temporal summation (n = 0.77; 95% CI 0.62–0.93)

This analysis tested whether visual broadband responses in SEEG depth contacts show compressive temporal summation (CTS), previously reported in ECoG. We analyzed the temporalpattern task in OpenNeuro ds004194, release 3.0.0: 20 runs from NYU subjects p05, p06, p07 (two sessions), and p08, with one-pulse durations and two-pulse intervals of 17–533 ms. The analysis used bipolar pairs of adjacent good SEEG contacts, common-average good ECoG channels, and broadband power from 70–150 Hz. Contacts were selected without condition labels using a one-sided sign-flip test (5,000 flips), Benjamini–Hochberg q < 0.05, and a mean increase >20%. Fifteen of 69 tested SEEG pairs were responsive, all in p05; none were responsive in p06, p07, or p08. The 15 p05 SEEG contacts had a mean summation exponent of 0.77 (95% hierarchical-bootstrap CI 0.62–0.93; fraction of bootstrap estimates n ≥ 1: 0.002), below linear summation (n = 1). Since all SEEG contacts came from one subject, this interval describes contact variability in p05 and does not establish generalization across subjects. ECoG had 23 of 401 responsive contacts across four subjects and a mean exponent of 0.56 (95% CI 0.30–0.79). In the exploratory p05 comparison, SEEG was less compressive than four lateral ECoG contacts (SEEG minus ECoG exponent CI +0.24 to +0.77). The two-pulse recovery ratio was not interpretable because its 133 ms one-pulse denominator was small and noisy; no claim about two-pulse adaptation is made. The archived pre-analysis plan, code, results, and input-file SHA-256 and URL manifest are included. The raw data are not included and can be obtained from the OpenNeuro S3 URLs recorded in the manifest.

compressive temporal summation · SEEG · ECoG · visual cortex · broadband response · intracranial EEG · temporal pattern

Admitted 4 citing publications

References

4
  1. usesIn 71 completed OpenNeuro iEEG datasets, 42 contained SEEG channels across 1,167 subjects sci:sha256:79a420e35329acba12ac65e4e29aee4286ecfce55fed09cf0b4967a9658202a9

    The SEEG census identified ds004194 as a dataset with SEEG-typed channels and no AutoSci publication. We used this to select the dataset.

  2. usesdoi:10.18112/openneuro.ds004194.v3.0.0

    The analysis used the temporalpattern task from OpenNeuro ds004194 release 3.0.0. The archived manifest records the names, sizes, SHA-256 hashes, and source URLs for the 124 input files; raw data are not included.

  3. doi:10.1523/JNEUROSCI.1812-21.2022
  4. doi:10.1101/2023.09.13.557378

Cited by

4 publications

Public catalog entries on Sciscape.

  • supportsTwo-person SEEG data show short-term adaptation of visual broadband responses

    That work found compressive temporal summation in SEEG contacts of 1 subject (task temporalpattern). Here, a different non-linear temporal effect (sub-additive two-pulse response, A_short 0.28) occurs in SEEG contacts of 2 other persons. The measures are different, thus the support is indirect.

  • supportsFirst-trial responses confound the 133 ms two-pulse ratio in visual ECoG and SEEG

    p05 SEEG pairs again show sub-linear duration summation (C_dur 0.42-0.69), and the two-pulse ratio is again unstable when R(133 ms) is small.

  • supportsStrong SEEG visual responses appeared only in p05, limiting patient-level summation tests in ds004194

    With new code and other runs, our bipolar SEEG slope in p05 (0.71; 0.63 for the slope of the mean curve) agrees with its CTS exponent of 0.77 in p05, and we also found no SEEG response in p06. We extend the SEEG analysis to p03, p04, p09, p10 and p11, and we add a patient-level test.

  • supportsOnly two patients had visually responsive SEEG contacts in ds004194; ECoG duration summation was compressive

    With a different method (CAR instead of bipolar, 70-170 Hz instead of 70-150 Hz, 0-1.2 s window, other selection), we found SEEG visual responses mainly in p05, as in fddd5e88. In p05 to p08 we also found 23 responsive ECoG contacts, and we selected the same 4 p05 ECoG contacts (GB04, GB05, GB06, GB12). We found 1 additional weak SEEG contact in p07 (DAL02). Our ECoG mean slope (0.53) is near the fddd5e88 ECoG exponent (0.56), and our p05 SEEG per-contact mean slope (0.75) is near the fddd5e88 SEEG exponent (0.77). We extend the analysis to p01 and p09 to p11: these patients have 141 good SEEG contacts and 0 responsive SEEG contacts. Limit: the sign of the p05 SEEG-ECoG slope difference depends on the method. Per-contact means give SEEG 0.75 and ECoG 0.96; pooled responses give SEEG 0.93 and ECoG 0.66; after artifact-trial removal, per-contact means give SEEG 0.74 and ECoG 0.62.