We present a brain-computer interface (BCI) that detects, analyzes and responds to user cognitive state in real-time using machine learning classifications of functional near-infrared spectroscopy (fNIRS) data. Our work is aimed at increasing the narrow communication bandwidth between the human and computer by implicitly measuring users' cognitive state without any additional effort on the part of... read morethe user. Traditionally, BCIs have been designed to explicitly send signals as the primary input. However, such systems are usually designed for people with severe motor disabilities and are too slow and inaccurate for the general population. In this paper, we demonstrate with previous work1 that a BCI that implicitly measures cognitive workload can improve user performance and awareness compared to a control condition by adapting to user cognitive state in real-time. We also discuss some of the other applications we have used in this field to measure and respond to cognitive states such as cognitive workload, multitasking, and user preference. Copyright 2015 Society of Photo-Optical Instrumentation Engineers (SPIE) One print or electronic copy may be made for personal use only. Systematic reproduction and distribution, duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the paper are prohibited.read less
Beste F. Yuksel, Evan M. Peck, Daniel Afergan, Samuel W. Hincks, Tomoki Shibata, Jana Kainerstorfer, Kristen Tgavalekos, Angelo Sassaroli, Sergio Fantini, Robert J. K. Jacob, "Functional near-infrared spectroscopy for adaptive human-computer interfaces", Proc. SPIE 9319, Optical Tomography and Spectroscopy of Tissue XI, 93190R (12 March 2015); doi: 10.1117/12.2075929; http://dx.doi.org/10.1117/12.2075929.