Attentional biases to threat are considered central to anxiety disorders, however physiological evidence of their nature and time course is lacking. Event-related potentials (ERPs) characterized sensory and cognitive changes while 20 outpatients with panic disorder (PD), 20 with obsessive–compulsive disorder (OCD), and 20 healthy controls (HCs) responded to the color (emotional Stroop task) or meaning of threatening and neutral stimuli. ERPs indicated larger P1 amplitude and longer N1 latency in OCD, and shorter P1 latency in PD, to threatening (versus neutral) stimuli, across instructions to attend to, or ignore, threat content. Emotional Stroop interference correlated with phobic anxiety and was significant in PD. Participants with emotional Stroop interference had augmented P1 and P3 amplitudes to threat (versus neutral) stimuli when color-naming. The results suggest early attentional biases to threat in both disorders, with disorder-specific characteristics. ERPs supported preferential early attentional capture and cognitive elaboration hypotheses of emotional Stroop interference.
The clinical presentation of anxiety disorders includes cognitive, affective, somatic and behavioral changes (APA, 2000 and Zacharko et al., 1995). Cognitive alterations include a focus on thought content related to danger (Rachman, 2007 and Tata et al., 1996), and sensitivity to threat-related cues in the environment (MacLeod, 2004). Attentional biases to threat appear to be causally related to clinical anxiety (MacLeod, 2004). They predict, for example, cortisol reactivity to stress, which may present a vulnerability to anxiety disorders (Fox, Cahill, & Zougkou, 2010). Threat-related biases are therefore implicated as central to the etiology and maintenance of anxiety disorders (Williams, Mathews, & MacLeod, 1996).
Over three decades, many studies have investigated threat-related biases in clinical and healthy populations, mostly relying on indirect measures such as reaction time (RT) impairment in secondary tasks when threat stimuli are present. RT studies have limitations in that they only allow for the study of biases accompanied by measurable behavioral changes, whereas attentional biases can occur independently of behavioral interference (e.g. Thomas et al., 2007 and van den Heuvel et al., 2005). RT studies also provide little information about the relative involvement of sensory versus cognitive processes in threat-related biases. The nature and timing of anomalies is of theoretical and clinical importance (Bar-Haim, Lamy, Pergamin, Bakermans-Kranenburg, & van Ijzendoorn, 2007), as distinctions may guide treatments. Recently, neuroimaging studies have advanced our understanding of brain regions activated during threat processing in anxious and healthy participants, however neuroimaging studies also lack the temporal resolution to localize biases to specific information-processing stages. Additionally, RT and imaging studies indicate that brain and behavioral responses to phobic stimuli differ between obsessive–compulsive and other disorders (van den Heuvel et al., 2005), however these differences remain incompletely understood and controversial (Moritz et al., 2004).
