Automatic recognition of hemodynamic responses to rare stimuli using functional near-infrared spectroscopy

Attention domain is of crucial importance for goal-directed behaviors and it has been widely studied through response analysis to rare stimuli using electroencephalography (EEG). More recent researches have explored the brain circuitry of attention by applying neuroimaging techniques, such as functional magnetic resonance. This paper investigates for the first time the feasibility of automatic recognition of responses to rare stimuli by using functional near-infrared spectroscopy (fNIRS). fNIRS is a portable brain imaging modality that optically measures the cortical hemodynamic activation and may prove useful in monitoring localized activity changes in frontal cortex related to attention processes. In this preliminary study, Fisher Linear Discriminant (FLD) is used to discriminate between average responses to rare task-relevant stimuli and responses to task-irrelevant stimuli.