von Luhmann, A. | bioRxiv Science

bioRxiv2026

Decoding Spatial Attention in the Cocktail Party Problem Using Wearable Whole-head High-Density fNIRS

Spatial attention is critical for solving the cocktail party problem, a longstanding problem in neuroscience and artificial speech recognition. The ability to decode where humans are attending in a cocktail party like scene would empower applications in brain computer interfaces and assistive devices such as hearing aids. Here we demonstrate that, in an overt attention task, the attended spatial location can be decoded robustly from single trial hemodynamic responses, using a wearable whole head high density fNIRS system. We also identify critical brain regions that make the highest contribution to decoding accuracy. Specifically, we find that decoding based on a small fraction of channels within the left and right inferior parietal lobule (IPL), achieve maximal decoding accuracy comparable to all channels. These results open the way for the design of novel BCIs and assistive devices integrated with fNIRS, that can be steered by spatial attention.

neuroscience↗

bioRxiv2024

ninjaCap: A fully customizable and 3D printable headgear for fNIRS and EEG brain imaging

SignificanceAccurate sensor placement is vital for non-invasive brain imaging, particularly for functional near infrared spectroscopy (fNIRS) and diffuse optical tomography (DOT), which lack standardized layouts like EEG. Custom, manually prepared probe layouts on textile caps are often imprecise and labor-intensive. AimWe introduce a method for creating personalized, 3D-printed headgear, enabling accurate translation of 3D brain coordinates to 2D printable panels for custom fNIRS and EEG sensor layouts, reducing costs and manual labor. ApproachOur approach uses atlas-based or subject-specific head models and a spring-relaxation algorithm for flattening 3D coordinates onto 2D panels, using 10-5 EEG coordinates for reference. This process ensures geometrical fidelity, crucial for accurate probe placement. Probe geometries and holder types are customizable and printed directly on the cap, making the approach agnostic to instrument manufacturers and probe types. ResultsOur ninjaCap method offers 2.2{+/-}1.5 mm probe placement accuracy. Over the last five years, we have developed and validated this approach with over 50 cap models and 500 participants. A cloud-based ninjaCap generation pipeline along with detailed instructions is now available at openfnirs.org. ConclusionsThe ninjaCap marks a significant advancement in creating individualized neuroimaging caps, reducing costs and labor while improving probe placement accuracy, thereby reducing variability in research.

neuroscience↗