Steady-state visual evoked potential (SSVEP), which uses blinking light stimulation to estimate the attending target, has been known as a communication technique with severe motor disabilities such as ALS and Locked-in-syndrome. Recently, it was reported that pupil diameter vibration based on pupillary light reflex has been observed in the attending target with a constant blinking frequency. This fact suggests the possibility of a noncontact BCI using pupillometers as alternatives to contacting scalp electrodes. In this study, we show an increment in the number of communication channels by stimulating both eyes alone or in combination with different frequencies. The number of selective targets becomes twice the number of frequencies using this method. Experiments are conducted by recruiting three healthy participants. We prepare six target patterns comprising three frequencies and detect the target using a coefficient of correlation of power spectrum between the pupil diameter and stimulus signal. Consequently, the average classification accuracy of the three participants of approximately 83.4% is achieved. The findings of this study demonstrate the feasibility of noncontact BCI systems.
Objective: Recently, pupil oscillations synchronized with steady visual stimuli were used as input for an interface. The proposed system, inspired by a brain-computer interface based on steady-state visual evoked potentials, does not require contact with the participant. However, the pupil oscillation mechanism limits the stimulus frequency to 2.5 Hz or less, making it hard to enhance the information transfer rate. Approach: Here, we compared multiple conditions for stimulation to increase the information transfer rate (ITR) of the pupil vibration-based interface, which were called monocular-single, monocular-superposed, and binocular-independent conditions. The binocular-independent condition stimulates each eye at different frequencies respectively and mixes them by using the visual stereoscopic perception of users. The monocular-superposed condition stimulates both eyes by a mixed signal of two different frequencies. We selected the shape of the stimulation signal, evaluated the amount of spectral leakage in the monocular-superposed and binocular-independent conditions, and compared the power spectrum density at the stimulation frequency. Moreover, 5, 10, and 15 patterns of stimuli were classified in each condition. Main results: A square wave, which causes an efficient pupil response, was used as the stimulus. Spectral leakage at the beat frequency was higher in the monocular-superposed condition than in the binocular-independent one. The power spectral density of stimulus frequencies was greatest in the monocular-single condition. Finally, we could classify the 15-stimulus pattern, with ITRs of 14.4 (binocular-independent, using 5 frequencies), 14.5 (monocular-superimposed, using 5 frequencies), and 23.7 bits/min (monocular-single, using 15 frequencies). There were no significant differences for the binocular-independent and monocular-superposed conditions. Significance: This paper shows a way to increase the number of stimuli that can be simultaneously displayed without decreasing ITR, even when only a small number of frequencies are available. This could lead to the provision of an interface based on pupil oscillation to a wider range of users.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.