Applying Real Time Physiological Measures of Cognitive Load to Improve Training

This paper discusses how the fields of augmented cognition and neuroergonomics can be expanded into training. Several classification algorithms based upon EEG data and occular data are discussed in terms of their ability to classify operator state in real time. These indices have been shown to enhance operator performance within adaptive automation paradigms. Learning is different from performing a task that one is familiar with. According to cognitive load theory (CLT), learning is essentially the act of organizing information from working memory into long term memory. However, our working memory system has a bottleneck in this process, such that when training exceeds working memory capacity, learning is hindered. This paper discusses how CLT can be combined with multiple resource theory to create a model of adaptive training. This new paradigm hypothesizes that a system that can monitor working memory capacity in real time and adjust training difficulty can