PNNL

Abstract

The focus of this paper is on the critical challenge of bridging the gap between psychophysiological sensor data and the inferred cognitive states of users. It is argued that a more robust behavioral data collection foundation will facilitate accurate inferences about the state of the user so that an appropriate mitigation strategy, if needed, can be applied. The argument for such a foundation is based on two premises: (1) To realize the envisioned impact of augmented cognition systems, the technology should be applied to a broad, and more cognitively complex, range of real-world problems. (2) To support identifying cognitive states for more complex, real-world tasks, more sophisticated instrumentation will be needed for behavioral data collection. It is argued that such instrumentation would enable inferences to be made about higher-level semantic aspects of performance. The paper describes how instrumentation software developed to support information analysis R&D may serve as an integration environment that can provide additional behavioral data, in context, to facilitate inferences of cognitive state that will enable the successful augmenting of cognitive performance.