Two probes and better than one: Development of a psychometrically reliable variant of the attentional probe task

Abstract

Cognitive models contend that attentional bias to negative information contributes to elevated anxiety vulnerability and dysfunction. The most common approach for assessing such bias is the attentional probe task. In this task, participants are presented with stimulus pairs, usually comprising a negative and benign member. A single visual probe subsequently appears in the locus where either stimulus was displayed, and remains on-screen until the participant identifies it. Attentional distribution is inferred from relative speed to identify probes in the location of each stimulus. Recently, investigators have raised concerns about the psychometric reliability of this attentional probe task as a measure of attentional bias, and have called for the development of new tasks with the capacity to more reliably assess variation in biased attentional responding to target stimuli. In response to this call, we report the development of a new dual probe attentional assessment approach, in which two probes are briefly presented on each trial, and attentional distribution is inferred from relative accuracy to identify probes appearing in each screen location. Across four studies, we show that this dual probe approach demonstrates much higher psychometric reliability than the single probe task, and can sensitively detect anxiety-linked attentional bias to negative information.