Working memory enables people to maintain task-relevant information in a highly active state. Although everyday life is filled with a great deal of visual information, our visual working memory can maintain representations of only three to four objects at a time 
. However, visual working memory capacity is not necessarily constant, but varies across individuals 
. Different working memory capacities among individuals affect several cognitive abilities, such as fluid intelligence 
or stereotype 
. Working memory performance is also severely disrupted in people with psychopathological traits, such as anxiety.
Anxiety consumes the working memory capacity available for superior performance 
. Depleted working memory capacity during worry is especially observed in highly anxious people 
. Considering this, many researchers believe that anxious people have diminished working memory capacity because anxiety occupies their working memory. In fact, some studies have shown that reduced working memory capacity is associated with anxiety 
. Amir and Bomyea 
used the operation span paradigm (OSPAN) 
, in which participants remember sequentially presented words with simple math equations. They have shown that individuals with social anxiety disorders remember fewer words than non-anxious individuals do. However, few previous studies have investigated visuospatial working memory. Considering that the important role of working memory is to maintain representations spatially and simultaneously 
, it is necessary to examine the visual working memory capacity at one time in anxiety.
In attentional control theory 
, Eysenck mentioned that when people perceive themselves to be under threat and experience anxiety, it is advantageous to allocate visual attention more widely in order to detect threatening stimuli. Based on this hypothesis, it is possible that individuals with anxiety attend to and hold many stimuli at a time. Actually, previous studies have shown the possibility of high attentional resources, but not visual working memory capacity, in high trait anxiety 
. Bishop 
used a perceptual load task 
in which a target was present in the center and a task-irrelevant distractor was present in the peripheral field. The stimuli were not emotionally laden and consisted of simple letters. Participants were instructed to detect the target without processing the task-irrelevant distractor. However, individuals with high trait anxiety, but not state anxiety, detected both the target and the task-irrelevant distractors. According to the perceptual load theory 
, attention has a limited capacity and the processing of task-irrelevant distractors depends on attentional resources. Individuals who had few attentional resources devoted all of them to the targets and could not detect task-irrelevant stimuli. However, individuals who have sufficient attentional resources can devote some to the target, but spare resources remain. The remaining resources are allocated to peripheral task-irrelevant stimuli. Therefore, it is possible that the processing of task-irrelevant distractors in individuals with trait anxiety was observed because they have more attentional resources than those with low trait anxiety. Considering that visual working memory and visual attention are intimately related 
, we assume that individuals with high trait anxiety also have high working memory capacity compared to those with low trait anxiety.
However, the processing of task-irrelevant distractors in high trait anxiety in previous studies also indicates the possibility that individuals with high trait anxiety have low visual working memory capacity. Based on a recent cognitive model in anxiety 
, top-down control, which includes inhibition of task-irrelevant stimuli, is impaired in trait anxiety but not in state anxiety. It is possible that individuals with high trait anxiety could not inhibit the task-irrelevant distractors because of impaired top-down control. Therefore, the processing of task-irrelevant distractors was observed with high trait anxiety. In fact, individuals with high trait anxiety showed reduced activity in the prefrontal cortex, which is associated with top-down control 
. This relationship between high trait anxiety and low top-down control was observed even when controlling for state anxiety. Because the role of working memory is to maintain task-relevant information and ignore task-irrelevant stimuli, these previous studies predict that even with visual information, trait anxiety, but not state anxiety, might be associated with low working memory and low visual working memory capacity. However, these previous studies investigated attention, and the relationship between anxiety and visual working memory capacity is not yet clear.
The present study investigated the effects of social anxiety on visual working memory capacity. We focused on trait social anxiety, which is related to anxiety in social situations (e.g., meeting new people, public speaking, and going to parties) and fear of negative evaluation from others 
. Individuals with social anxiety are too sensitive to evaluation from others and fearful of negative evaluation 
. They are very vigilant to social information in social situations, such as facial expressions and gaze direction of others 
. A recent study also showed that individuals with social anxiety are not only sensitive to social or emotional stimuli but also to non-emotional visual stimuli, such as bright stimuli 
. High sensitivity to visual stimuli might be due to fear that is activated by visual images in socially anxious individuals 
. General anxiety, however, consists mainly of thought and is not necessarily associated with concrete visual images 
. Therefore, we assume that visual working memory is more associated with trait social anxiety. We also measured the degree of state anxiety. According to recent studies and models 
that show impaired top-down attention and processing of task-irrelevant stimuli, trait anxiety, but not state anxiety, might be associated with visual working memory capacity. We used a change-detection task in this study, a method that is commonly used and is an established task for measuring visual working memory capacity 
. In this task, participants are shown an array of visual stimuli to encode. A test array is presented after a short retention interval, and participants are required to answer whether the test array is identical to or different from the memory array. The accuracy of the task is used to estimate visual working memory capacity 
. Based on the high attentional resources and the wide visual attentional allocation in trait anxiety 
, visual working memory capacity might be positively correlated with trait social anxiety but not state anxiety. However, according to the theory of impaired top-down control in trait anxiety, visual working memory might be negatively correlated with trait social anxiety but not state anxiety.