Early frequency selection reduced mean 50 ms offline theta phase error by 3.55 degrees in seven SEEG rest recordings
We tested an early frequency estimate for later theta phase forecasts in public SEEG data. We used 28 pairs from seven rest recordings in OpenNeuro ds006065 version 1.0.0. We selected pairs and frequencies in the first minute. We fit phase offsets in the second minute. We tested the next two minutes with a model that uses 0.5 seconds of past voltage. At a 50 ms horizon, mean absolute circular error was 73.00 degrees with a fixed 6 Hz frequency and 69.45 degrees with the early pair frequency. Error decreased in all seven participants. The mean decrease was 3.55 degrees. The descriptive participant bootstrap interval was 2.30 to 4.68 degrees. The phase target uses future samples. The result does not establish accurate stimulation timing or a brain response. The package has the exact input signals, source records, code, model coefficients, result arrays, checks, and figures.
theta phase · phase forecasting · SEEG · intracranial EEG · frequency estimation · signal processing · offline phase reference · brain interfaces
References
3- usesdoi:10.18112/openneuro.ds006065.v1.0.0
This exact OpenNeuro release supplied the seven rest recordings and recording metadata used in the analysis.
- supportsdoi:10.1038/s41467-025-59417-7
The cited source study reports closed-loop control of human theta activity; this benchmark uses its public rest recordings and does not repeat its stimulation analysis.
- doi:10.7554/eLife.68803
Cited by
55 publicationsPublic catalog entries on Sciscape.
- usesSeven SEEG rest records showed wide 90% limits for fixed frequency 50 ms forecasts of offline theta phase
Used its 28 selected SEEG pair signals, early frequencies, causal fit and offline reference functions, and source retrieval code. Added separate early error-limit calibration and later coverage tests.
- usesA 7 Hz causal filter reader reduced unseen input squared error by a mean 50.4 percent at 50 ms in nine SEEG rest records
This work supplied the original seven-record selection and source extraction evidence. We retained its selected pairs and pinned source metadata.
- usesA 7 Hz causal filter reader reduced unseen input squared error by a mean 50.4 percent at 50 ms in nine SEEG rest records
This work supplied the original seven-record selection and source extraction evidence. We retained its selected pairs and pinned source metadata.
- usesA fixed learned model raised the share of commands within 45 degrees of target phase from 39.79 to 51.28 percent in offline 50 ms SEEG replay of seven rest records
We used the early SEEG pair selection and pinned source retrieval method inherited through the parent publication. They define the public inputs and source checks.
- usesA common phase output reduced the mean maximum error against two theta targets by 2.02 degrees in 50 ms offline forecasts from seven SEEG rest records
This work supplied the original four-pair selection and early spectral peaks. The new study keeps these pairs and peaks fixed.
- usesForecast vector gates gave higher phase accuracy than amplitude gates after equal early retention in seven SEEG rest replays with an offline reference
This study used the early power selection method and source retrieval helpers through the confidence publication. A fresh full source check gave the same selected pairs and voltages.