Perceptual decoupling in the sustained attention to response task is indeed unlikely: a reply to Shelat and Geisbrecht (in press).

Abstract

Shelat and Geisbrecht (in press) challenge Bedi et al.'s (Exp Brain Rese 242(8):2033-2040 2024b) position that perceptual decoupling in the Sustained Attention to Response Task (SART) is unlikely. Instead they argue perceptual decoupling is likely in the SART and advocate for the SART's continued use in perceptual decoupling research. Shelat and Geisbrecht, however, are overlooking the extensive behavioral evidence that perceptual decoupling in the SART is indeed unlikely, including research by the researchers who originally developed the task demonstrating nearly 100% awareness of the task stimuli. The SART was developed to be a very short replacement for the long duration low Go, high No-Go target detection tasks used by sustained attention or vigilance researchers. While altering the response format in the SART to a high Go, low No-Go task indeed resulted in errors occurring reliably in a very short duration, the resulting SART has a substantial speed-accuracy trade-off. This causes immense confusion when interpreting performance in the SART. Furthermore, Shelat and Geisbrecht suggest DeBettencourt et al. (Nat Hum Behav 3(8):808-816, 2019) as a method improvement on the original SART, but ignore the entire point of the SART which was to be a short duration replacement for traditional vigilance tasks. The task utilized by DeBettencourt et al. (Nat Hum Behav 3(8):808-816, 2019) is as long in duration or longer than traditional vigilance tasks, but still is contaminated with a speed-accuracy trade-off, which makes untangling the underlying processes involved challenging. If researchers want to study sustained attention- perceptual decoupling, vigilance researchers have already figured out how to do this and the way to do this is not the SART.