A BIDS-Formatted fNIRS Dataset of Hand Clenching and Finger Tapping

This dataset contains functional near-infrared spectroscopy (fNIRS) recordings from 100 participants performing two right-hand sensorimotor tasks: hand clenching and finger tapping. Data were collected with the NIRSport2 system (NIRx Medizintechnik GmbH, Germany; wavelengths: λ1 = 760 nm, λ2 = 850 nm) using an increased-density (8x8 optode setup; 20 standard-distance and one short-separation channel) fNIRS montage covering the left hemispheric sensorimotor cortex to capture contralateral sensorimotor activation. This release (v1.0.0) contains a computationally sub-selected 5x5 optode and 12-channel subset corresponding to a standard-density configuration, as widely used in applied and clinical fNIRS settings. Isolating this subset reflects realistic, hardware-constrained acquisition conditions. The remaining channels will be released at a later time as part of a separate, ongoing methodological comparison between standard- and increased-density montages. As the released subset is an unmodified portion of the same recordings, future extension with the remaining channels will be fully backward-compatible, and results based on the current release will remain valid. The dataset is organized according to the Brain Imaging Data Structure (BIDS, version 1.11.1) and includes raw fNIRS recordings, event files, and participant metadata. Additionally, a derivatives folder with preprocessed data, using a standard preprocessing pipeline, is included. The data is intended for use in machine learning benchmarking, sensorimotor research, and brain-computer interface (BCI) research. A manuscript describing this dataset, its acquisition, the preprocessing pipeline, a technical validation, and a proof-of-concept classification benchmark is currently under review. This description will be updated with the citation once available.