Attentional bias to threat in the general population is contingent on target competition, not on attentional control settings

Abstract

Dot-probe studies usually find an attentional bias towards threatening stimuli only in anxious participants. Here, we investigated under what conditions such a bias occurs in unselected samples. According to contingent-capture theory, an irrelevant cue only captures attention if it matches an attentional control setting. Therefore, we first tested the hypothesis that an attentional control setting tuned to threat must be activated in (non-anxious) individuals. In Experiment 1, we used a dot-probe task with a manipulation of attentional control settings ('threat' - set vs. control set). Surprisingly, we found an (anxiety-independent) attentional bias to angry faces that was not moderated by attentional control settings. Since we presented two stimuli (i.e., a target and a distractor) on the target screen in Experiment 1 (a necessity to realise the test of contingent capture), but most dot-probe studies only employ a single target, we conducted Experiment 2 to test the hypothesis that attentional bias in the general population is contingent on target competition. Participants performed a dot-probe task, involving presentation of a stand-alone target or a target competing with a distractor. We found an (anxiety-independent) attentional bias towards angry faces in the latter but not the former condition. This suggests that attentional bias towards angry faces in unselected samples is not contingent on attentional control settings but on target competition.

It occurs after all: Attentional bias towards happy faces in the dot-probe task

“…The results of Experiments 1, 2a, and 2b suggest that an attentional bias towards happy faces can occur in the dot-probe task under specific conditions. However, this bias was not contingent on the presentation of distractor stimuli during the target displayunlike attentional bias towards angry faces as we have previously shown (Wirth & Wentura, 2018a). So why did a bias towards happy faces occur in Experiment 1 while other dot-probe studies did not find such a bias?…”

“…Participants again performed a dotprobe task where they had to classify either socially meaningful target stimuli (schematic faces) or socially meaningless target stimuli (scrambled schematic faces). If we focus on the double-line target condition (which uses the same target-distractor assignment as previous experiments by Wirth & Wentura, 2018a, 2019, and the present Experiments 1, 2a, and 2b), participants showed a significant attentional bias towards happy face cues. Unlike attentional bias towards angry faces (Wirth & Wentura, 2019), however, this bias was not significantly moderated by the social character of the target stimuli.…”

“…To this end, we conducted five experiments that used a short CTOA of 100 ms and perceptually non-confounded happy face cues (without exposed teeth) in order to increase the likelihood of detecting a potential bias. In previous studies (Wirth & Wentura, 2018a, 2019, we showed that (non-anxious) participants show an attentional bias towards angry faces in the dotprobe task only if (1) target stimuli have to compete for attention with simultaneously presented distractor stimuli (i.e., if participants are not in an onset-singleton search mode) and (2) a social-processing mode is activated due to current task demands. Consequently, we additionally investigated in the present experiments whether attentional bias towards happy faces is contingent on similar top-down processes.…”

“…Interestingly, while other paradigms reliably find attentional biases towards threatening stimuli in the general population (see Yiend, 2010, for a review), dot-probe studies usually find such a bias only in anxious, but not in non-anxious participants (for meta-analyses see Bar-Haim et al, 2007;Frewen et al, 2008). Consequently, it has been argued that attentional bias towards negative emotional stimuli in the dot-probe task might not be unconditional but contingent on top-down processes that are affected by participants' trait anxiety (Puls & Rothermund, 2018) or by current task demands (Wirth & Wentura, 2018a, 2019. These considerations should be taken into account when investigating whether an attentional bias towards happy faces (i.e., positive emotional stimuli) can be reliably shown in the dot-probe task.…”

Furious snarling: Teeth-exposure and anxiety-related attentional bias towards angry faces

“…For example, Everaert et al [ 16 ] found a significant bias towards threatening scenes in an unselected sample when participants had to simultaneously perform a secondary task that required affective categorization, but not when the secondary task required semantic categorization. Furthermore, Wirth and Wentura [ 63 ] found an attentional bias towards angry faces in an unselected sample when target categorization required social processing of the stimuli, but not when social processing was irrelevant for target categorization.…”