Augmented Cognition: Developing and Testing a Physiology-Based Task Adaptation System

Abstract

The purpose of this study was to examine if an individual's performance could be improved by integrating physiological arousal data with a computer task, specifically a video game, which increased or decreased in difficulty based on the person's physiological status. Ninety-nine participants (80 males; 19 females) were included in the data analyses. The mean age of the included participants was 21.9 years (SD = 2.6). Using cardiac inter-beat interval (IBI) data each individual's respiratory sinus arrhythmia (RSA) was derived and used to objectively assess the parasympathetic activity of the autonomic nervous system. The RSA data were continuously converted into a standardized arousal score. In the condition of interest, Auto, the difficulty of the computer game was continuously manipulated based on the real-time arousal score of the participant. Subjective arousal data were also collected verbally throughout each experimental session. It was hypothesized that manipulating the difficulty of the computer game such that arousal was maintained at an optimal level would result in better performance compared to yoked and manual control of game difficulty. Results showed that automated task difficulty manipulation actually resulted in lower performance scores compared to yoked assistance manipulation, and tended to result in lower scores compared to manual assistance manipulation.