Attentional resources in social anxiety and the effects of perceptual load

Abstract

According to the attentional control theory, anxious people should process task-irrelevant distractors more than non-anxious people. This could be due to them having more attentional resources to allocate to the processing of task-irrelevant distractors. The current study was designed to assess the processing of task-irrelevant stimuli in low- and high-socially-anxious people. We conducted four experiments using perceptual load tasks. In the high-perceptual-load condition, interference effects from task-irrelevant distractors were not observed in low-socially-anxious people. This finding is consistent with the view of exhausting attentional resources to process task-relevant stimuli. However, interference effects were observed in high-socially-anxious people in the case of high perceptual load (Experiment 1). These effects were observed when presenting the distractor at a fixation (Experiment 2) or attracting attention involuntarily to the target location by spatial cueing (Experiment 3). However, when distractors were masked to decrease their visibility, distractor processing was not observed (Experiment 4). These results suggested that people with high social anxiety may have more attentional resources than people with low social anxiety, and this might partially derive from enhanced stimulus-driven attention.

Keywords:

• Social anxiety
• Attentional resources
• Perceptual load
• Goal-directed attention
• Stimulus-driven attention

Notes

¹Some studies divide participants into high- and low-anxiety groups by a quartile or tertile (Bradley et al., 1997; Bradley et al., 1998; Mansell, Clark, Ehlers, & Chen, 1999). On the other hand, many studies use a median split to divide participants (Bishop, 2009; Keogh & French, 2001; Mogg, Bradley, de Bono, & Painter, 1997; Mogg et al., 2000; van Honk, Tuiten, de Haan, van den Hout, & Stam, 2001). We use a median split to increase the size of groups.

²In the present research, we used a median split to divide the participants equally, and the criteria for high and low social anxiety were different in each experiment. We divided the participants into high- and low-social-anxiety groups by using other criteria in Experiments 1, 2, 3, and 4. In all the experiments, the lowest scores of social anxiety in high-socially-anxious people varied between 41 and 46. On the basis of these scores, we divided the participants into different high-social-anxiety groups, by the score of above 41, 42, 43, 44, 45, and 46 respectively, and reanalysed the RTs in each experiment. The results were similar to those described in this manuscript. For the compatibility effects, the three-way interactions between social anxiety, perceptual load, and compatibility in Experiments 1, 2, and 3 were also statistically or marginally significant. The follow-up t-tests for the simple effects in Experiments 1, 2, and 3 and post hoc further analysis in Experiment 4 also showed similar statistical results in every experiment.

³This procedure corresponds to that of Henderson (1991, Experiment 5). In this study, the target appeared at each of the four locations with equal probability, regardless of the location of the appearance of the cue. Even if the cue was completely uninformative, its validity effect was observed.