A critical examination of the research and theoretical underpinnings discussed in Thomson, Besner, and Smilek (2016).

Abstract

Thomson, Besner, and Smilek (2016) propose that performance decrements associated with sustained attention are not consistently the result of a decline in perceptual sensitivity. Thomson et al. (2016) present empirical evidence using a novel, nontraditional vigilance task to support their assumptions. However, in the present rebuttal, we argue that the authors have not only have misinterpreted previous research in sustained attention, but also have misapplied those interpretations to their study. Thomson et al. have also neglected key elements of the literature in their argument, including research on expectancy theory and individual differences on vigilance performance. Furthermore, Thomson and colleagues implement an experimental paradigm that is not appropriate for evaluating sensitivity and bias changes in vigilance tasks. Finally, their analyses do not capture the manner in which changes in response bias and sensitivity can manifest in signal detection theory. We discuss the theoretical and experimental issues contained in Thomson et al. (2016) and propose suggestions for future vigilance research in this area. (PsycINFO Database Record

The ANTI-Vea task: analyzing the executive and arousal vigilance decrements while measuring the three attentional networks

“…It is worth noting that, recently, the study of Thomson et al (2016) received several critics by Fraulini, Hancock, Neigel, Claypoole, & Szalma (2017). These authors questioned the analytical methods carried out by Thomson et al to obtain SDT metrics.…”

“…This distinction is arbitrary in the ANTI-Vea and only made a posteriori for the purpose of analysis. Therefore, despite the objection stated by Fraulini et al (2017), the computation of corrected SDT metrics seems to be an adequate approach to analyze the EV decrement in signal detection tasks, at least in tasks like the one embedded in the ANTI-Vea.…”

“…Moreover, Fraulini et al (2017) pointed out that in Thomson et al (2016) the vigilance task was performed online and outside the lab, without controlling the experimental context. Notwithstanding, Thomson et al outcomes were replicated later in a study conducted under the typical lab conditions (Claypoole et al, 2018).…”

“…For instance, in the standard version, we observed a higher percentage of hits, a linear decrement on hits, and a more pronounced increment in the response bias, in comparison with the lure task. Furthermore, the standard version does not suffer from Fraulini et al (2017)' criticism about the three categories of events, as discussed above. Therefore, although the lure task was tested here as an alternative version of the ANTI-Vea, the standard ANTI-Vea was revealed as the most adequate version of the task.…”

Independent Coactors May Improve Performance and Lower Workload: Viewing Vigilance Under Social Facilitation

“…Vigilance, or the ability to maintain attention over extended periods and to respond to critical events when they occur (Davies & Parasuraman, 1982), has been at the forefront of human factors research for almost 70 years (Fraulini, Hancock, Neigel, Claypoole, & Szalma, 2017). Vigilance tasks can be cognitively demanding and also boring (Scerbo, 2001), and as such, performance tends to decline with time on watch (Mackworth, 1948, 1950).…”