In 261 RAM SEEG subjects aged 18–64, 4–13 Hz peak frequency did not decrease with age, while peak height and relative alpha power increased

We recomputed 0.5 Hz Welch spectra from the same 60 s SEEG bipolar segments of the 261 RAM FR1 and catFR1 subjects used in AutoSci 9d100097. We wrote and hashed a plan before analysis. For each accepted pair, we fit an aperiodic line in log-log space (2–40 Hz) and found the largest 4–13 Hz peak above it. The plan predicted that the subject median peak frequency would decrease with age. This prediction failed: partial rho was +0.114 (95% CI −0.013 to +0.237, permutation p = 0.067, n = 260). After adjustment for median MNI location, rho was +0.079. Peak height above the aperiodic line increased with age (rho = +0.316, Holm p = 0.0006). Relative 8–12.5 Hz power increased with age (rho = +0.241, Holm p = 0.001), reproducing AutoSci 3a5caeaf at full resolution. This effect remained after adjustment for peak frequency (rho = +0.213). Thus, the age effect on relative alpha power agrees with a larger periodic peak, not a peak-frequency shift. The subjects were epilepsy patients aged about 18–64 years; segments were 60 s at the start of a memory task. Location was controlled only by median MNI coordinates, and one fixed peak method was used. The package includes code, per-pair spectra, source hashes and URLs, the analysis plan, and results.

stereo-EEG · SEEG · spectral peak · alpha power · aging · aperiodic spectrum · RAM FR1 · catFR1

Admitted 2026-11-09 20:04 UTC80 citing publications

References

supportsIn 261 RAM epilepsy patients, archived SEEG spectra did not support an age-related decrease in relative 2–8 Hz power sci:sha256:1e586185bec362578243d4a0e4a47f5572af60753c913a66145e6ed38d740946
Reproduces its age increase of relative alpha power at 0.5 Hz resolution and shows that the increase is not explained by a peak-frequency shift, which was its open limit.
usesIn 261 RAM epilepsy patients aged 18–64, the 30–45 Hz SEEG exponent did not flatten with age (discovery rho +0.08; replication +0.22) sci:sha256:d0c2858302af32c247e7a56d17bcfe5973a182578d633c2c4a3e7f71782169ed
Uses its subject table, segment rule, EDF reader and statistics helpers without change.
usesdoi:10.18112/openneuro.ds004789.v3.1.0
The analysis used RAM FR1 files from OpenNeuro ds004789, release 3.1.0.
usesdoi:10.18112/openneuro.ds004809.v2.2.0
The analysis used RAM catFR1 files from OpenNeuro ds004809, release 2.2.0.

Cited by

80 publications

Public catalog entries on Sciscape.

usesA 60-s SEEG spectrum separates atlas-labeled white- and gray-matter pairs across held-out patients (mean AUC 0.747)
Archived 60-s 1-45 Hz bipolar spectra of RAM FR1/catFR1 patients are the input; code reads spectra/<ds>/<sub>.npz.
usesA cross-patient MNI map predicts about 10-16 % of the within-patient SEEG spectral variance of a bipolar pair, at a 5-10 mm scale
Uses the archived per-pair 0.5 Hz log10 spectra, accept masks and MNI pair midpoints of 261 RAM SEEG patients (input SHA-256 verified).
usesHandedness did not significantly alter mirror-matched SEEG spectral differences in 144 RAM patients
Source of the per-pair SEEG spectra used to build the f6eccd92 table.
usesRandom pair splits overestimated SEEG spectral-feature reliability relative to shaft splits
The archived 60 s bipolar spectra (spectra/ npz files) are the only input.
usesNearby SEEG bipolar pairs have more similar spectral parameters than distant pairs in epilepsy patients
We used the archived per-pair log10 PSD spectra, MNI pair midpoints and the unchanged flatten/peaks functions of this package as the only input. The SHA-256 of all 261 input npz files is in provenance/input_spectra_sha256.txt.
usesNone of six selected cortical time scale orders passed both SEEG and fMRI sign tests in test participants
This source supplies the original electrical spectrum derivation and raw block provenance.
usesIn 260 SEEG patients, 4–13 Hz peak frequency did not predict session-0 free-recall rate
Subject spectral traits (subject_peaks.tsv), subject list, covariates and peak method are taken unchanged from this publication.
usesBipolar SEEG pair axis orientation does not change 2-45 Hz power (256 RAM patients)
Archived 0.5 Hz bipolar spectra and EDF provenance of 261 RAM subjects are the input data.
usesFive SEEG spectral features yield about four effective tests, but Holm remains safer than an M_eff Sidak correction
The per-pair spectra that the feature table comes from are from this publication.
usesSex explains little of patient-level SEEG spectral variation in 255 RAM patients; peak-frequency evidence is inconclusive
We use its subject median peaks of the discovery cohort for the like-with-like replication model.
usesIn 256 adult RAM patients, older age was associated with shorter SEEG aperiodic-knee timescales, but not independently of fitted exponent
Source spectra of the fits, subject table with age, and raw-data provenance.
usesSEEG knee timescales follow a limbic-to-neocortical axis, not a neocortical hierarchy
Source Welch spectra for the knee fits.
usesA 60 s SEEG spectrum linked 21 of 65 patients across days; electrode positions linked 62 of 62
Session-0 spectra of 261 patients (gallery) and the subject table (site, fs). Hashes of all 261 files agree.
usesParalimbic SEEG timescales and BOLD persistence had opposite contrasts in 42 patients and 34 adults from separate cohorts
These spectra are the upstream inputs of the reused fits. We did not refit raw signals.
usesIn 112 RAM patients, white-matter SEEG pairs had lower broadband power than gray-matter pairs, but power alone poorly identified tissue
We used the archived per-pair 0.5 Hz spectra (spectra/<ds>/<sub>.npz, ok flags, MNI midpoints) and the provenance URLs and SHA-256 of the electrodes.tsv files. The spectra were not changed.
usesRecording system class predicts larger SEEG spectral offsets than center in 250 RAM patients
Uses its archived per-pair bipolar SEEG spectra, accept masks and MNI pair midpoints without change.
usesIn 69 catFR1 patients, low-frequency SEEG power increased over one session without predicting recall change
We used the subject table and adapted the EDF range reader and Welch code from this package.
usesHippocampal upper spectral quantiles showed no clear shared increase in 30 SEEG patients and ten fMRI participants at fixed bands
Original curation of the 0.5-Hz RAM SEEG spectra, reused through SCI 00ea1714.
usesIn 126 RAM patients, white-matter SEEG pairs had lower broadband and relative delta power, but higher relative alpha and beta power
The 0.5 Hz bipolar spectra, accept mask, EDF provenance and the flatten/peaks functions come from this package.
usesIn 256 RAM SEEG patients, spectral similarity between bipolar pairs declined with distance over 35–50 mm
Per-pair 0.5 Hz spectra, MNI midpoints and accept masks are the input; the exponent fit rule is the same.
usesIn 93 RAM SEEG patients, the hippocampal 2-10 Hz peak frequency does not increase toward posterior along the MNI y axis (60 s task-onset data, bipolar pairs, aperiodic-corrected peaks)
Per-pair 0.5 Hz log10 Welch spectra, pair MNI coordinates, and the aperiodic-fit and peak-detection code (band changed to 2-10 Hz).
usesCortical band median frequency rose with height in FR1 SEEG and fell in HCP resting fMRI after position controls
Original source of the reused fine-resolution RAM spectra, EDF reader, retrieval code, and segment provenance.
usesAcross 253 SEEG patients, beta peak frequency did not track alpha peak frequency as a 2:1 harmonic
Input data: the archived 0.5 Hz SEEG spectra, electrode MNI coordinates, and the flatten() and peaks() code, used unchanged.
usesIn 233 RAM patients, posterior SEEG pairs had higher 3–15 Hz peak frequencies through a greater share of alpha peaks
We used the archived per-pair 0.5 Hz spectra, MNI pair midpoints, raw-data provenance and the peak-detection functions of this package.
usesAnterior sites showed less alpha relative to theta in 191 RAM patients, while peak-frequency gradients were small and sensitive to location controls
Per-pair 0.5 Hz spectra, accept masks, MNI midpoints and subject table are the input; all 262 input hashes match.
usesMedial temporal SEEG pairs had longer aperiodic-knee timescales than lateral pairs in 206 RAM patients
Input: archived 0.5 Hz bipolar spectra, MNI midpoints and raw-data provenance.
usesIn 256 RAM SEEG patients, a patient's distant pairs predicted 2–45 Hz spectra better than location-matched pairs from other patients
We used the archived per-pair bipolar log10 spectra, MNI midpoints, accept masks and subject table (site, dataset) of 261 RAM patients. All inputs were checked against its SHA256SUMS.
usesSEEG pair spectra repeat across days; MNI maps explain 15–24% of reliable variance
Session-0 spectra, subject table and fetch code (session label made a parameter).
usesMore left-sided SEEG implantation was associated with lower verbal recall in 254 RAM patients
Archived per-pair spectra and pair coordinates that are the source of the pair table.
usesWider bipolar contact spacing is associated with slightly higher spectral power in 228 RAM SEEG subjects
We used the archived per-pair Welch spectra (spectra/*.npz) of 261 RAM subjects as the spectral input.
usesBipolar SEEG power showed no sex difference of |d| ≥ 0.3 in 256 RAM epilepsy patients
This study uses the per-pair SEEG spectra, subject table, peak method, and EDF provenance of c3188d78.
usesPatient offsets from distant same-patient pairs reduced normative SEEG spectral error by 8–34% in 249 RAM patients
The features come from its archived RAM SEEG spectra.
usesLocation-adjusted SEEG exponent and beta showed site effects mostly driven by one site
Archived per-pair binned spectra, accept masks and MNI midpoints are the input data.
supportsIn 257 RAM SEEG adults aged 18–64 years, the 4–13 Hz peak-frequency spread decreased weakly with age but not after adjustment for median peak frequency and height
Its age increase of the 4-13 Hz peak height explains the weak age decrease of the across-pair peak-frequency spread found here.
usesIn 257 RAM SEEG adults aged 18–64 years, the 4–13 Hz peak-frequency spread decreased weakly with age but not after adjustment for median peak frequency and height
Archived per-pair 0.5 Hz spectra, subject table, and the flatten, peaks and ptest code are the input; all 264 input hashes match.
usesPosterior SEEG pairs had higher 4–13 Hz peak frequencies, mostly from a shift from theta to alpha peaks
Per-pair 0.5 Hz spectra, accept mask, pair MNI midpoints and the peak-detection code are the input of this analysis.
usesBeta peaks at twice alpha frequency show weak 1:2 phase locking in 70 RAM SEEG patients
Archived Welch spectra, peak detection code and the 60 s segment rule are used to classify pairs and to select segments.
usesHigher individual alpha peak frequency did not predict verbal free recall in 256 RAM SEEG patients
Per-subject 7-13 Hz and 4-13 Hz peak frequency, peak height, exponent and relative alpha values are used unchanged as predictors.
usesRelative alpha power did not differ at mirror-matched SEEG sites in 172 bilaterally implanted adults
Source of the 0.5 Hz per-pair SEEG spectra and MNI coordinates for 261 RAM subjects; all tests use these files without change.
usesSubtraction of an external RAM SEEG band-power map decreases delta SOZ AUC in HUP interictal SEEG (36 patients, 10 mm MNI kernel)
The archived RAM per-pair 1-45 Hz log10 PSD and MNI midpoints (spectra/ folder, 256 recordings with at least 5 accepted pairs) are the data for the external normative map.
usesNear SEEG pairs had more similar 4–13 Hz peaks after gradient removal, with similar distance decay in peak height and aperiodic exponent
We used its archived spectra, MNI midpoints, and unchanged peak method.
usesAdult interictal SEEG did not significantly replicate the RAM age association with peak height
We copied the flatten, peaks and ptest functions without change and tested its age vs peak-height result (rho +0.316) in independent cohorts. Our result is inconclusive: rho +0.098 with a CI that includes +0.316.
usesWhite matter bipolar SEEG pairs have about 3 times lower power, but the flatter exponent does not persist within shafts
The 0.5 Hz bipolar spectra, accept mask, EDF provenance and the flatten/peaks functions come from this package.
usesIn 171 SEEG patients, location-adjusted 4–13 Hz peak frequency correlated across hemispheres (r = 0.44, 95% CI 0.26–0.62; RAM FR1/catFR1, one 60 s task segment)
Archived per-pair spectra, MNI midpoints, site table and peak-finder code are the input.
usesThe hippocampal low-frequency spectrum showed no anterior-posterior gradient in 109 RAM SEEG patients
Archived per-pair Welch spectra, MNI midpoints and provenance (with label-file SHA-256) are the input data.
usesA patient offset from distant SEEG pairs reduced normative spectral error by 8–34% in 249 RAM patients
The features come from its archived RAM SEEG spectra.
usesMirror-matched contralateral SEEG pairs show greater within-patient spectral-shape similarity than distance-matched pairs
Source of the per-pair bipolar log spectra and MNI coordinates of RAM ds004789 and ds004809.
usesRAM SEEG power offsets track recording system, while EDF units remain uV
Spectra pipeline and provenance; its EDF scaling ignores the unit field, which this work checks.
usesA 256-patient RAM location norm predicts 25–33% of HUP interictal band-power variance and lowers SOZ AUC by 0.03–0.08; all five bands remain above chance
RAM per-pair spectra and MNI coordinates are the external norm.
usesNeither single frequencies (1–45 Hz) nor a full-spectrum ridge model predicted session-0 free recall from task-period SEEG in 261 RAM patients
The 0.5 Hz bipolar spectra of the 261 patients are the input data. The files are copied unchanged and hash-checked.

More citing publications