ABSTRACT
We developed and tested a neuroergonomic smartphone application called Mind Metrics that can be used to evaluate vigilance and working memory under naturalistic conditions. The application met a requirement to the field of neuroergonomics because the cognitive tasks were made for a smartphone platform, allowing the ability to make predictions about the neural processes that impact human performance during naturalistic work related activities (i.e. ubiquitous computing). However, if naturalistic tasks are to be developed that are sensitive to cognitive processes, these tasks must be tested and evaluated for validity by comparing performance to data obtained in controlled laboratory environments. In this study, we developed tasks that measure working memory and vigilance, two processes that are well known to affect human performance at work. We then tested participants on these tasks using both a smartphone and a desktop computer platform. The tasks we used to measure vigilance included a vigilance task called the Psychomotor Vigilance Task (PVT) and a vigilance task called the Spatial Discrimination Vigilance Task (SDVT). To measure working memory, we used a Color N-back Task (CNB) and a Spatial N-back Task (SNB). Using a mixed group design, participants were assigned to a desktop or smartphone condition and completed all four tasks. As predicted, there was a vigilance decrement for both the PVT and the SDVT, which was demonstrated by an overall slowing of responses as the tasks progressed. This decrement occurred for both the smartphone and the 240desktop tasks. Another interesting finding related to improvement over time for the N-back tasks-- participants performed faster on the n-back tasks as the task progressed. This indicates that task learning is an important factor to consider when developing neuroergonomic tasks aimed at detecting cognitive functioning in the wild. In previous research it was found that increased resource demands exacerbate the vigilance decrement. These findings suggest that learning can play a role in attenuating the vigilance decrement effect in tasks with high resource demands. If vigilance tasks developed on the smartphone can be administered in naturalistic environments this platform will provide a method of easy-to-obtain samples of repeated task performance, thereby reducing the impact of learning effects that can mask the vigilance decrement. The possible implications of this research are a more sensitive measure of the vigilance decrement for detecting vigilance in the wild.
