Mechanisms of vigilance loss in an online task: A generative modeling approach
Event Type
TimeWednesday, October 12th3:30pm - 4:30pm EDT
LocationPoster Gallery
DescriptionThe ability to detect rare signals among background noise declines with time on task, a phenomenon called the vigilance decrement. Current theories attribute the vigilance decrement to three alternative mechanisms: bias shifts, sensitivity losses, and attentional lapses. This experiment examined the contributions of each mechanism in an online vigilance task. One hundred ninety-four participants completed an online visual signal detection task, judging whether the gap between two probe circles exceeded a criterion value. Data were fit with a generative cognitive model with parameters representing response bias, sensitivity, and attentional lapse rate. Change in each parameter was compared across the first and last four minutes of the task in a hierarchical Bayesian analysis. Data gave evidence that the decrement was driven by a conservative shift in response bias and an increased frequency of attentional lapses. Although data trended in the direction of a sensitivity loss, the effect was not statistically credible.