Spatial Resolution of Visual Stimuli in SSVEP-based Brain-Computer Interface

Abstract

Communicating spatial coordinates plays a crucial role in human-robot interactions, where a given target, object, or location needs to be localized. The steady-state visual evoked potential (SSVEP) is one of the most robustly detectable signals in electroencephalography (EEG)-based brain-computer interfaces (BCIs). However, the spatial resolution of the visual stimuli in a SSVEP-based BCI needs to be characterized for localization applications. In this study, we demonstrate that the influence of an adjacent stimulus attenuates to the baseline level when it is outside the paracentral region of human field of view (FOV) based on data collected from five subjects. This conservatively defines the spatial resolution in SSVEP. A potential lateral inhibition phenomenon was also observed when the two stimuli were immediately next to each other, which may reflect the center-surround structure of the receptive fields in visual cortex. Moreover, different frequency setups appear to affect the robustness of the SSVEP-based BCI and suggest that adjacent stimuli should be with frequencies that are more distinguishable visually.