Kaori Suefusa1, Toshihisa Tanaka. 1. Department of Electronic and Information Engineering, Tokyo University of Agriculture and Technology, 2-24-16 Nakacho, Koganei-shi, Tokyo, 184-8588, Japan.
Abstract
OBJECTIVE: Brain-computer interfacing (BCI) based on visual stimuli detects the target on a screen on which a user is focusing. The detection of the gazing target can be achieved by tracking gaze positions with a video camera, which is called eye-tracking or eye-tracking interfaces (ETIs). The two types of interface have been developed in different communities. Thus, little work on a comprehensive comparison between these two types of interface has been reported. This paper quantitatively compares the performance of these two interfaces on the same experimental platform. Specifically, our study is focused on two major paradigms of BCI and ETI: steady-state visual evoked potential-based BCIs and dwelling-based ETIs. APPROACH: Recognition accuracy and the information transfer rate were measured by giving subjects the task of selecting one of four targets by gazing at it. The targets were displayed in three different sizes (with sides 20, 40 and 60 mm long) to evaluate performance with respect to the target size. MAIN RESULTS: The experimental results showed that the BCI was comparable to the ETI in terms of accuracy and the information transfer rate. In particular, when the size of a target was relatively small, the BCI had significantly better performance than the ETI. SIGNIFICANCE: The results on which of the two interfaces works better in different situations would not only enable us to improve the design of the interfaces but would also allow for the appropriate choice of interface based on the situation. Specifically, one can choose an interface based on the size of the screen that displays the targets.
OBJECTIVE: Brain-computer interfacing (BCI) based on visual stimuli detects the target on a screen on which a user is focusing. The detection of the gazing target can be achieved by tracking gaze positions with a video camera, which is called eye-tracking or eye-tracking interfaces (ETIs). The two types of interface have been developed in different communities. Thus, little work on a comprehensive comparison between these two types of interface has been reported. This paper quantitatively compares the performance of these two interfaces on the same experimental platform. Specifically, our study is focused on two major paradigms of BCI and ETI: steady-state visual evoked potential-based BCIs and dwelling-based ETIs. APPROACH: Recognition accuracy and the information transfer rate were measured by giving subjects the task of selecting one of four targets by gazing at it. The targets were displayed in three different sizes (with sides 20, 40 and 60 mm long) to evaluate performance with respect to the target size. MAIN RESULTS: The experimental results showed that the BCI was comparable to the ETI in terms of accuracy and the information transfer rate. In particular, when the size of a target was relatively small, the BCI had significantly better performance than the ETI. SIGNIFICANCE: The results on which of the two interfaces works better in different situations would not only enable us to improve the design of the interfaces but would also allow for the appropriate choice of interface based on the situation. Specifically, one can choose an interface based on the size of the screen that displays the targets.
Authors: Alberto J Molina-Cantero; Jaime Guerrero-Cubero; Isabel M Gómez-González; Manuel Merino-Monge; Juan I Silva-Silva Journal: Sensors (Basel) Date: 2017-06-29 Impact factor: 3.576