In one animal micro-SEEG subject, high-impedance contacts showed sub-pitch 300–3000 Hz correlation lengths under common-average reference
In one likely pig (animal) micro-SEEG subject from OpenNeuro ds004819 v1.0.0, we analyzed three 1 Hz spine stimulation runs (0.2, 0.4, 0.6 mA), 44 contacts, and within-column pairs at 120–3840 µm. Correlations were measured in six frequency bands under scalp and common-average references. Pre-specified P1 (shorter 300–3000 Hz than 1–4 Hz length constant in every run × column unit) and P2 (lower 300–3000 Hz nearest-neighbor correlation in every unit) were not supported. P3 (decreasing length constant across bands) was supported with scalp reference but not common-average reference. Under common-average reference, column B had 300–3000 Hz length constants of 63–74 µm, below the 120 µm pitch and therefore only an upper bound, with nearest-neighbor r of 0.37–0.45. Its 1–150 Hz length constants were 127–508 µm. Under scalp reference, shared reference signal dominated column B correlations below 150 Hz. Low-impedance column A remained highly correlated with little distance decline, and its fits were degenerate; this is consistent with a column-specific shared signal such as coupling or common noise, not local neural signals. The subject is not a human patient, its animal species is uncertain, and column A should not be used for spatial-scale estimates without further checks. The package preserves the original plan, code, metadata, hashes, fit table, curves, log, and summaries; raw EDF files are not included.
micro-SEEG · intracranial EEG · inter-contact correlation · electrode impedance · spatial correlation · frequency bands · common-average reference · animal recording
References
2- usesdoi:10.18112/openneuro.ds004819.v1.0.0
Source dataset snapshot for the raw micro-SEEG recordings and associated metadata analyzed here.
- doi:10.1101/2022.11.08.515705
Publication associated with the OpenNeuro dataset, as cited in its dataset metadata.
Cited by
3 publicationsPublic catalog entries on Sciscape.
- usesSpine stimulation did not produce a current-dependent pulse-locked response in one animal micro-SEEG recording
Uses the same subject, the same contact rule (impedance < 1000), the EDF reader code and the reported input hashes (runs 03, 05, 07 agree). It adds the evoked-response analysis of all 8 runs, which d6e126f6 did not do.
- One non-human ds004819 micro-SEEG subject showed column-specific correlations and no consistent spinal-stimulation response
[Legacy relation: reproduces] With different code, all 8 runs (not 3) and a common-median reference, column A again had a common high-frequency signal at all distances and column B had distance-dependent correlation.
- supportsRecorded-reference SEEG correlations persisted beyond 60 mm, while bipolar data did not support a shorter high-frequency length constant
That work found that 300-3000 Hz correlation did not decrease faster with distance than 1-4 Hz correlation in animal micro-SEEG, and that a reference offset dominated scalp-referenced correlation. In human macro-SEEG (ds005670), we also found that 70-150 Hz did not have a shorter length constant than 1-4 Hz (common average), and that the recorded contact reference added a distance-independent correlation of 0.17-0.67.