Stimulus-Driven Attention, Threat Bias, and Sad Bias in Youth with a History of an Anxiety Disorder or Depression

Abstract

Attention biases towards threatening and sad stimuli are associated with pediatric anxiety and depression, respectively. The basic cognitive mechanisms associated with attention biases in youth, however, remain unclear. Here, we tested the hypothesis that threat bias (selective attention for threatening versus neutral stimuli) but not sad bias relies on stimulus-driven attention. We collected measures of stimulus-driven attention, threat bias, sad bias, and current clinical symptoms in youth with a history of an anxiety disorder and/or depression (ANX/DEP; n = 40) as well as healthy controls (HC; n = 33). Stimulus-driven attention was measured with a non-emotional spatial orienting task, while threat bias and sad bias were measured at a short time interval (150 ms) with a spatial orienting task using emotional faces and at a longer time interval (500 ms) using a dot-probe task. In ANX/DEP but not HC, early attention bias towards threat was negatively correlated with later attention bias to threat, suggesting that early threat vigilance was associated with later threat avoidance. Across all subjects, stimulus-driven orienting was not correlated with early threat bias but was negatively correlated with later threat bias, indicating that rapid stimulus-driven orienting is linked to later threat avoidance. No parallel relationships were detected for sad bias. Current symptoms of depression but not anxiety were related to decreased stimulus-driven attention. Together, these results are consistent with the hypothesis that threat bias but not sad bias relies on stimulus-driven attention. These results inform the design of attention bias modification programs that aim to reverse threat biases and reduce symptoms associated with pediatric anxiety and depression.

Keywords: Adolescent; Anxiety; Attention bias; Depression; Sad bias; Stimulus-driven attention; Threat bias.

Figure 1

Attention tasks to measure stimulus-driven attention (G-SOT), threat bias and sad bias at 150ms (E-SOT), and threat bias and sad bias at 500ms (DPT). In each task, the trial begins with a central fixation cross followed by the presentation of a single box cue (G-SOT); a single face with a neutral, angry, sad expression (E-SOT); or a pair of faces with neutral and emotional expressions (DPT). Following the delay indicated in each panel in red, a target pair of dots appears randomly at the left or right location and the subjects’ task is to indicate with a button press whether the dots are oriented vertically or horizontally.

Figure 2

Threat avoidance 500ms after the appearance of emotional faces is associated with threat vigilance at 150ms and increased stimulus-driven attention in youth with a history of an anxiety disorder and/or depression (ANX/DEP). Among ANX/DEP, attention bias to threat at 500ms is negatively correlated with both attention bias to threat at 150ms and stimulus-driven attention. To graphically depict these negative correlations, ANX/DEP subjects are divided in the 50% of subjects with the highest (red) and lowest (blue) threat bias at 500ms. (A) ANX/DEP Subjects with the lowest threat bias (or highest threat avoidance) at 500ms had the highest threat bias at 150ms. (B) ANX/DEP subjects with the lowest threat bias (or highest threat avoidance) at 500ms had the highest measures of stimulus-driven attention to a generic, non-emotional stimulus.