Citation

Abdelfattah SM, Merrick KE, Abbass HA. Proc. Hum. Factors Ergon. Soc. Annu. Meet. 2016; 60(1): 21-25.

Copyright

(Copyright © 2016, Human Factors and Ergonomics Society, Publisher SAGE Publishing)

DOI

10.1177/1541931213601005

PMID

unavailable

Abstract

Visual tracking is a vital task in various control systems such as air traffic control. Analysing mental processing indicators during such tasks can lead to a better understanding of changes in the cognitive state of human operators. In this paper, we conducted an electroencephalography experiment for simulated visual tracking tasks using 10 subjects. A descriptive model based on decision trees is adopted for electroencephalography feature selection. Our analysis methodology involved four steps. First, we generated a large pool of features. Second, we formed a descriptive decision tree for each subject. Third, we contrasted the features found among all subjects' decision trees to identify those commonly used features, which we then ranked. Fourth, we compared the classification accuracy when using the selected features against the whole dataset and unselected features.

RESULTS showed that Theta-Beta Ratio for different channels contributed to approximately 90% of the highly rated selected features. These features also yielded better generalization and classification accuracy. This finding indicates that Theta-Beta Ratios can be used as an effective cognitive indicator for visual tracking tasks.

Language: en