Social anxiety disorder (SAD) and major depression have high levels of comorbidity (Kessler, 1995 and Pini et al., 1997) and share several common features such as social avoidance and self-critical cognitions (Gotlib & Meltzer, 1987; Heimberg et al., 1989; Wallace & Alden, 1991). Cognitive models postulate that both SAD and depression are associated with biased attentional processes such that individuals with these conditions selectively attend to mood-relevant information (Beck, 1987; Clark & Wells, 1995; Rapee & Heimberg, 1997). These models suggest that this bias only occurs when material matches the specific vulnerability of each disorder (Beck, 1987; Clark & Wells, 1995; Rapee & Heimberg, 1997). Within the cognitive framework, anxiety-relevant material is characterized by vulnerability and danger whereas depressive-relevant material is characterized by negative views of the self, world, and future (Beck, Rush, Shaw & Emery, 1979; Beck, Emery, & Greenberg, 1985). Thus, individuals with SAD would be expected to over-attend to cues pertaining to negative evaluation (e.g., “rejection”), and individuals with depression would be expected to over-attend to cues about hopelessness and negativity.
Several studies have examined the attentional biases postulated to be associated with SAD and depression (Beck, 1987; Clark & Wells, 1995; Rapee & Heimberg, 1997). Many of these studies have used the Emotional Stroop Task (EST) to measure attentional interference (e.g., Hope, Rapee, Heimberg, & Dombeck, 1990; Gotlib & Cane, 1987). In this paradigm, emotionally threatening words are presented in different colors and participants name the color of the words as quickly as possible while ignoring their meaning. Cognitive bias is revealed when individuals display longer latencies to naming mood-relevant words than neutral words. Several studies have found that individuals with SAD display Stroop interference (i.e., longer color-naming latencies) for socially threatening stimuli compared to control words (e.g., Hope et al., 1990 and Mattia et al., 1993). However, with respect to the depression literature, some studies have found that individuals with depression take longer to color name depressive threatening stimuli (e.g., Gotlib & Cane, 1987; Gotlib & McCann, 1984), whereas others have not (e.g., Hill & Knowles, 1991; Mogg, Bradley, Williams, & Mathews, 1993). A recent study by Gotlib et al. (2004), which compared individuals with SAD, individuals with depression, and disorder-free control individuals using the EST, found no evidence of Stroop interference for either SAD or depression. However, the lack of a group of individuals with both SAD and depression may limit the conclusions that can be drawn from the study, because it has been suggested that the discrepant findings in this literature may be the result of ignoring comorbid conditions (Mogg et al., 1993).
To address this concern, studies have compared the attentional processes of individuals with depression, individuals with generalized anxiety (some of which also were experiencing depression), and disorder-free control individuals using the EST (Bradley, Mogg, Miller, & White, 1995; Mogg et al., 1993). Results of these studies found that the Anxiety group displayed longer response latencies for anxiety threat words compared to control and depressive words, but no differential latencies were found in the depressed group between the different stimuli. Two recent studies, using the framework of the tripartite model (e.g., Clark & Watson, 1991), found some support for general distress symptoms of anxiety to be associated with Stroop interference, whereas no support was found for general distress symptoms of depression to be associated with an attentional bias (Yovel and Mineka, 2004 and Yovel and Mineka, 2005). It is salient to note that some have suggested that when generalized anxiety and depression are comorbid, depression seems to counteract the attentional bias associated with anxiety (Mogg et al., 1993 and Bradley et al., 1995). Although these data are important in documenting how depressive symptoms can influence generalized anxiety, research is needed to examine how comorbidity impacts attentional processes between social anxiety and dysphoria. The current study addresses this gap in the literature by administering the EST to undergraduates chosen to represent a Social Anxiety group, a Dysphoric group, and a Social Anxiety/Dysphoric group. It was hypothesized that the Social Anxiety group would display an attentional bias to socially threatening stimuli compared to depressive threat, neutral, and positive stimuli. It also was hypothesized that neither the Dysphoric nor the Social Anxiety/Dysphoric group would display attentional biases to any of these threat stimuli.
