Citation

Spence ML, Mattingley JB, Dux PE. J. Exp. Psychol. Hum. Percept. Perform. 2018; 44(12): 1981-1994.

Affiliation

School of Psychology.

Copyright

(Copyright © 2018, American Psychological Association)

DOI

10.1037/xhp0000584

PMID

30475052

Abstract

Humans intuitively evaluate their decisions with different levels of confidence. Although confidence and sensitivity are highly correlated, recent evidence has shown that confidence is disproportionately impacted by signal variability (e.g., de Gardelle & Mamassian, 2015; Spence, Dux, & Arnold, 2016; Zylberberg, Roelfsema, & Sigman, 2014). Previous studies investigating this issue have involved manipulations of variability in the components of the decision stimulus itself. Here, in 3 experiments, we investigated whether discrete variability, from a visual stimulus that does not require a response, can affect confidence in a secondary visual task. Participants made brightness or direction judgments about the dots in motion kinematograms with different ranges of motion around the global direction. Specifically, after viewing pairs of dot-motion displays, participants received a postcue at the end of each trial asking them to report either the relative brightness or the direction of the second display relative to the first. Importantly, the range of motion directions was irrelevant to the task when individuals were required to judge the brightness of the dots. We found that a larger range of motion directions reduced participants' confidence in their brightness judgments but with no corresponding change in performance. These findings suggest that confidence is impacted not only by variability directly relevant to the decision stimulus itself but also by variability in the context in which the decision evidence was encoded. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Language: en