An Examination of Sustained Attention During Complex Multitasking Scenarios

This dissertation project examined the time-course of vigilance decrements that occur when operators perform demanding multitasking activities. For this aim, we conducted three experiments which implemented a novel paradigm we developed that measured performance during two sessions: a Single-Task session in which participants performed a go-no-go target detection task in the absence of any other task for approximately 12 minutes; and a Multi-Task session in which participants performed the detection task simultaneously with a driving-based tracking task for the same duration. A total of 183 participants from the University of South Carolina Department of Psychology took part in this study. Growth curve analyses revealed quadratic trajectories across accuracy and error measures, and linear trajectories across the tracking and RT measures, and that the inclusion of late verbal tasks (Experiment 3) negatively affected all detection task measures during more difficult conditions, but did not affect the tracking task measure. Further, vigilance decrements during both sessions and across all measures had higher intensity and variability during the more difficult conditions, and practice reduced these effects, but only to an extent. Together, these findings support the cognitive overload and opportunity cost models of vigilance performance. Insight from this work can significantly inform the design and development of complex operator-system interfaces, and further develop theoretical understanding of attention and multitasking performance.