Background
Current tools to assess cognitive load using physiological data are largely constrained with respect to the task space in which they are applied and cannot be used to estimate cognitive performance or to assess brain health in a meaningful way. This PDIR addresses this gap by functionalizing key identified measure(s) of load to a fieldable implementation, enabling us to gather objective measures of load in relevant settings. The system was used to examine the effect of physical load on cognitive performance.
Approach
A system was created for the collection of psychophysiological measurements during cognitive loading tasks. Using this system, cognitive performance data was gathered from 12 participants before and after a cognitive loading task. During data collection, a subject’s cognitive baseline was established during cognitive loading tasks (Reading Working Memory [RWM] and Rotational Working Memory assessments). Subjects pedaled on a stationary bike to physically load the subject until they were unable to continue. Cognitive performance was compared before and after the physical loading task to quantify the effect of physical fatigue on cognitive performance.
Figure 1: Stationary bike used for physical loading task.
Figure 2: Candidate components for fieldable system.
Accomplishments
We found that generally, cognitive performance decreased following the physical loading task. There were notable exceptions, where cognitive performance increased following the physical loading tasks. These individuals generally had the shortest duration of physical loading, likely choosing to stop before they experienced significant physical load. From these results, a small amount of physical loading promotes increased cognitive performance, while a significant amount of physical load promotes decreased cognitive performance on RWM tasks. Cognitive baseline data was also used to train a model to predict cognitive performance from psychophysiological data. The predictive power of this model will be compared to that of a similar model trained on data gathered from a lab-based system. This work is ongoing.