100 results for "model interpretability" · page 8 of 10 · ranked by relevance
[ and electroencephalography (EEG) dataset recorded from 30 participants viewing 21,600 video clips spanning 180 categories of human action, extending the previously released Human Action Dataset (HAD) fMRI resource. It was collected in the same participants and with the same stimuli as HAD-fMRI to enable combined spatiotemporal investigation of neural mechanisms underlying human action recognition. The dataset leverages the millisecond-level temporal resolution of M/EEG to complement the spatial precision of fMRI.
…or by more advanced transformer models of vision. We propose that the…
This dataset contains MEG data collected as part of the FLUX pipeline project, a standardized analysis and preprocessing pipeline for magnetoencephalography (MEG) research. The dataset was reorganized into BIDS format from an original OpenNeuro dataset to support reproducible MEG analysis workflows across multiple research sites.
…By implementing a cross-validation procedure, we trained and validated the model…
This dataset contains EEG recordings from 34 participants collected to investigate the adaptive recruitment of cortex-wide recurrent processing during visual object recognition. Participants viewed a stimulus set of 242 images, comprising 'challenge' and 'control' images selected based on discrepancies between human behavioral performance and AlexNet classification, while performing a rapid serial visual presentation task with a paper-clip detection component. The dataset includes derivatives with time-resolved decoding accuracy matrices for object identity, supporting analyses of recurrent cortical dynamics in visual processing.
[ and rest conditions in both movement execution and motor imagery modalities. The study investigates neural encoding of individual upper limb movements using low-frequency EEG signals (0.3-3 Hz) and achieves classification accuracies of 55-87% for executed movements and 27-73% for imagined movements. Source localization analysis identifies discriminative movement information in premotor areas, primary motor cortex, somatosensory cortex, and posterior parietal cortex, with applications toward non-invasive control of motor neuroprostheses and robotic arms.
…We include the ICA decomposition and dipole model in EEG.etc. The…
This dataset contains preprocessed EEG recordings from 6 healthy participants performing imagined speech tasks with three short word conditions (out, in, up). Data were acquired at 256 Hz using 64 channels and analyzed using Riemannian manifold methods and relevance vector machines for brain-computer interface applications. The motor imagery paradigm employed auditory and visual cueing, yielding 5,400 trials suitable for BCI research and benchmarking.