Discrimination of Mental Workload Levels in Human Subjects With Functional Near-Infrared Spectroscopy

Abstract

We have applied functional near-infrared spectroscopy (fNIRS) to the human forehead to distinguish different levels of mental workload on the basis of hemodynamic changes occurring in the prefrontal cortex. We report data on 3 subjects from a protocol involving 3 mental workload levels based on to working memory tasks. To quantify the potential of fNIRS for mental workload discrimination, we have applied a 3-nearest neighbor classification algorithm based on the amplitude of oxyhemoglobin (HbO 2 ) and deoxyhemoglobin (HbR) concentration changes associated with the working memory tasks. We have found classification success rates in the range of 44%-72%, which are significantly higher than the corresponding chance level (for random data) of 19.1%. This work shows the potential of fNIRS for mental workload classification, especially when more parameters (rather than just the amplitude of concentration changes used here) and more sophisticated classification algorithms (rather than the simple 3-nearest neighbor algorithm used here) are considered and optimized for this application.

Real-time Prediction of User Performance based on Pupillary Assessment via Eye-Tracking

“…Scholars have found physiological markers of mental effort in EEG (Fairclough, Venable, & Tattersall, 2005;Wang, Gwizdka, & Chaovalitwongse, 2016;Buettner, 2017b), fMRI (Lim et al, 2010;Gwizdka, 2013b), fNIR (Sassaroli et al, 2008;Herff et al, 2014), EDA (Wilson, 2002;Boucsein, 2012;Buettner, 2017b), HR (Vogt, Hagemann, & Kastner, 2006;Brookhuis & de Waard, 2010), facial action (Buettner, 2017b(Buettner, , 2018, fEMG (Stone & Wei, 2011;Ekman, Friesen, & Hager, 2002), PET (Kramer, 1990;Just, Carpenter, & Miyake, 2003), MEG (Tanaka, Ishii, & Watanabe, 2015;Ishii, Tanaka, & Watanabe, 2016), and various eye-tracking measures (e.g., Rayner, 1998;Buettner, 2013Buettner, , 2017bGwizdka, 2016). In particular, they have found that mental effort is associated with a lot of eye-related characteristics, such as a user's pupil diameter (Hess & Polt, 1964;Beatty, 1982;Buettner, 2017b), eye-blink duration and blink rate (Fairclough et al, 2005;Marshall, 2007), eye saccade speed (Porter et al, 2010;Buettner, 2013Buettner, , 2017b, and the number of eye gaze fixations (Rayner, 1998;Buettner, 2013Buettner, , 2017b.…”

Dynamic difficulty using brain metrics of workload

“…Predictive models have been used to differentiate the fNIRS signal between levels of workload [4,20,28,34], verbal and spatial working memory [19], and game difficulty levels [17]. Furthermore, it has been used to determine periods of cognitive multitasking [36,37], levels of expertise [7], preference [25,29], and emotion [40].…”

Brain-based target expansion

“…Relying on simple and cheap fundamental technology, fNIRS is safe, comfortable, easy-to-setup, and has the potential for portability. The data from the device has been used to differentiate between levels of workload [3,25,37,42], verbal and spatial working memory [24], game difficulty levels [16], and cognitive multitasking. Putting this data to practical use, predictive models have been applied in real time to adapt user interfaces to these physiological states [45,46].…”