A Comparison of Artificial Neural Networks, Logistic Regressions, and Classification Trees for Modeling Mental Workload in Real-Time

Abstract

The use of eye metrics to predict the state of one's mental workload involves reliable and accurate modeling techniques. This study assessed the workload classification accuracy of three data mining techniques; artificial neural network (ANN), logistic regression, and classification tree. The results showed that the selection of model technique and the interaction between model type and time segmentation have significant effects on the ability to predict an individual's mental workload during a recall task. The ANN and classification tree both performed much better than logistic regression with 1-s incremented data. The classification tree also performed much better with data averaged over the full recall task. In addition, the transparency of the classification tree showed that pupil diameter and divergence are significantly more important predictors than fixation when modeling 1-s incremented data.

Assessing Cognitive Workload Using Cardiovascular Measures and Voice

“…Various pattern recognition methods have been applied to the task of classifying cognitive workload using psychophysiological measures. It has been pointed out that artificial neural networks are opaque and hard to interpret in terms of how individual variables interact to predict workload [11]. Classification methods have been used such as discriminant analysis and support vector machines [28], as well as logistic regression and classification trees [11].…”

“…It has been pointed out that artificial neural networks are opaque and hard to interpret in terms of how individual variables interact to predict workload [11]. Classification methods have been used such as discriminant analysis and support vector machines [28], as well as logistic regression and classification trees [11]. There is no indication that other classification methods can provide better results for cognitive workload monitoring [11,28].…”

“…Classification methods have been used such as discriminant analysis and support vector machines [28], as well as logistic regression and classification trees [11]. There is no indication that other classification methods can provide better results for cognitive workload monitoring [11,28]. Recently, however, trinary cognitive workload level classification with cardiovascular signals has been demonstrated with promising results [29].…”

“…Increased mental demand is highly correlated with increased cardiovascular reactivity [7,8]. Various classifier methods have also been used to recognize different cognitive workload states (reliably high and low workload) based on combined psychophysiological signal sources [9][10][11]. Despite this, going beyond a binary high/low workload classification has proved to be difficult.…”

Foundations of Augmented Cognition

“…Three supervised-learning regression algorithms were evaluated for their ability to infer workload from physiological data: linear regression, model trees, and the multi-layer perceptron (also known as an artificial neural network). Due to space limitations, and because it had the best performance of the algorithms evaluated, only the model tree algorithm (Quinlan 1992, Wang and Witten 1997, Fong 2010 are discussed further in this section, although the performance of all three algorithms is presented in Sect. 4.…”