Agreement between the 3-minute Psychomotor Vigilance Task (PVT) Embedded in a Wrist-worn Device and the Laptop-based PVT

Abstract

The study assesses the agreement between the 3-minute version of the Psychomotor Vigilance Task (PVT) with an interstimulus interval (ISI) of 2 to 10 seconds and the validated 3-minute laptop-based PVT (ISI=1-4 seconds). The experiment utilized a randomized, within-subject, repeated-measures design with three factors (PVT device type, the backlight feature of the wrist-worn device, ambient lighting). Results show the differences in reaction times (RT) between devices are incrementally associated with the magnitude of the RTs. These differences tend to be in opposing directions when the backlight feature in the wrist-worn device is on. That is, RTs in the wrist-worn device tend to be faster compared to the laptop for (on average) faster individuals, whereas (on average) slower individuals tend to do better in the laptop compared to the wrist-worn device. The proportional bias introduced by the wrist-worn device compared to the laptop makes it difficult to translate individual RTs between different devices. The proportional bias, however, may work in favor for detecting differences between slow and fast RTs.