Gaze-and-brain-controlled interfaces for human-computer and human-robot interaction

• Published in: Psychology in Russia : state of the art 2017-01, Vol.10 (3), p.120-137
• Main Authors: Shishkin, Sergei L., Zhao, Darisii G., Isachenko, Andrei V., Velichkovsky, Boris M.
• Format: Article
• Language: English
• Subjects: attention; brain output pathways; brain-computer interface (BCI); electroencephalography (EEG); expectancy wave (E-wave); eye movements; eye-brain-computer interface (EBCI); eye-to-eye contact; human-robot interaction; Robots
• ISSN: 2074-6857; 2307-2202
• DOI: 10.11621/pir.2017.0308

Background. Human-machine interaction technology has greatly evolved during the last decades, but manual and speech modalities remain single output channels with their typical constraints imposed by the motor system’s information transfer limits. Will brain-computer interfaces (BCIs) and gaze-based control be able to convey human commands or even intentions to machines in the near future? We provide an overview of basic approaches in this new area of applied cognitive research. Objective. We test the hypothesis that the use of communication paradigms and a combination of eye tracking with unobtrusive forms of registering brain activity can improve human-machine interaction. Methods and Results. Three groups of ongoing experiments at the Kurchatov Institute are reported. First, we discuss the communicative nature of human-robot interaction, and approaches to building a more e cient technology. Specifically, “communicative” patterns of interaction can be based on joint attention paradigms from developmental psychology, including a mutual “eye-to-eye” exchange of looks between human and robot. Further, we provide an example of “eye mouse” superiority over the computer mouse, here in emulating the task of selecting a moving robot from a swarm. Finally, we demonstrate a passive, noninvasive BCI that uses EEG correlates of expectation. This may become an important lter to separate intentional gaze dwells from non-intentional ones. Conclusion. The current noninvasive BCIs are not well suited for human-robot interaction, and their performance, when they are employed by healthy users, is critically dependent on the impact of the gaze on selection of spatial locations. The new approaches discussed show a high potential for creating alternative output pathways for the human brain. When support from passive BCIs becomes mature, the hybrid technology of the eye-brain-computer (EBCI) interface will have a chance to enable natural, fluent, and the effortless interaction with machines in various fields of application.