(PS3-3) Adaptation and Evaluation of a Child Version of the Attention Bias Questionnaire

Abstract Text: Extensive research documents a link between anxiety and attention bias to threat. Threat-related attention biases are typically measured using computer-administered behavioral tasks, which can be difficult to implement outside of research laboratories and fail to capture the subjective experience of attention bias. A self-report measure, the Attention Bias Questionnaire (ABQ), was recently developed to measure the subjective experience of attention bias towards threat in adults, and promising support was obtained for its reliability and validity. No such measure exists for youth, despite ample research from behavioral tasks demonstrating the existence of threat-related attention bias in anxious youth. The present research adapted the ABQ for use with children (ABQ-C) and examined its factor structure, internal consistency, and validity. A team of four clinical child and developmental psychologists first adapted the nine items from the adult ABQ using developmentally appropriate language. In study 1, the resulting ABQ-C was administered to N = 188 community-based youth ages 10-14 years (mean = 12.2 years, SD = 1.1 years) to examine factor structure, internal consistency, and concurrent validity. Confirmatory factor analysis (CFA) supported the theorized two-factor structure of the ABQ-C, with fit indices demonstrating adequate fit (RMSEA = 0.08, CFI = 0.95, SRMR = 0.05). The first factor consists of five items measuring Difficulty to Disengage from Threat, while the second factor consists of four items measuring heightened initial Engagement with Threat. Internal consistency was acceptable, with Cronbach’s alphas of 0.88, 0.84, and 0.78 for the total ABQ-C, Difficulty to Disengage from Threat, and Engagement with Threat factors, respectively. Concurrent validity was supported via expected correlations with measures of anxiety symptoms, executive functioning, and error sensitivity (Pearson r’s ranged from .39 to .42). In study 2, the ABQ-C was administered to N = 78 clinically referred socially anxious youth ages 10-14 years (mean = 11.8 years, SD = 1.6 years). Youth also completed the Dwell Task, an established attention bias measure that records attention using eye tracking while participants view matrices displaying happy (non-threatening) and angry (threatening) faces. Convergent validity was supported via expected correlations between ABQ-C scores and speed of orienting to threatening stimuli and longer initial dwell time on threatening stimuli, respectively (Pearson r’s of -.27 and .27). Overall, findings from these two studies support the theorized factor structure, internal consistency, and validity of scores on the ABQ-C as a self-report measure of attention bias in youth. Future research is encouraged to examine retest reliability.