Evaluation of long-range temporal correlations during overt and covert attention in a steady state visual evoked potential based brain-computer interface

Abstract

Gaze control is required for successful brain-computer interface (BCI) operation in different paradigms. It has been shown that the performance of a steady-state visual evoked potential-based BCI is lower in covert attention when the participants attend to the stimuli covertly, without the need to move their eyes. Some studies in the literature have tried to find the brain regions that are affected by covert attention. Moreover, it has been shown that the signal-to-noise (SNR) ratio is smaller in covert attention than in overt attention. Based on the fact that brain oscillations exhibit long-range temporal correlations (LRTCs), which can be measured by the Hurst exponent, and estimated using the detrended fluctuation analysis (DFA), this is the first study focusing on the DFA differences in overt and covert attention in an SSVEP-based BCI experiment. The main hypothesis is that there should be differences between DFA exponents of EEG in overt and covert attention, as there are differences in SNR between these attentional states. Gender differences between overt and covert attention were also evaluated using DFA. The results revealed significant differences in LRTCs depending on the gender and the attentional state. These results could be taken into account for an efficient SSVEP-based BCI design.