We investigated the use of a multimodal functional neuroimaging system in quantifying mental workload of healthy human volunteers. We recorded behavioral performance measures as well as electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) simultaneously from subjects performing n-back tasks. The EEG and fNIRS signals were used in feature generation and classification offline using support vector machines. We examined the classification accuracy of three distinct systems: EEG based; fNIRS based; and Hybrid, which contained features from the first two systems as based on their interactions. The classification accuracy of the Hybrid system was observed to be greater than that of either system, indicating the synergistic role played by multimodal signals and by neurovascular coupling in quantifying mental workload.