Human-Computer Interaction and Brain Measurement Using Functional Near-Infrared Spectroscopy

Functional near-infrared spectroscopy (fNIRS) is an emerging non-invasive, lightweight imaging tool which can measure blood oxygenation levels in the brain. We present an experiment showing how it could be used in humancomputer interaction. We use machine learning techniques to analyze fNIRS data to classify different levels of mental workload. Using our fNIRS measurements and machine learning algorithms, we were able to distinguish different levels of user workload and interaction styles in a computer task with average accuracy as high as 83% across subjects. Our results show the feasibility of using fNIRS in HCI research. ACM Classification: H5.2 [Information interfaces and presentation]: User Interfaces. Graphical user interfaces. General terms: Design, Human Factors, Experimentation