Background: Attention bias modification (ABM) aims to reduce attentional bias for threat (AB), thereby diminishing anxiety symptoms. However, recent metaanalyses indicated mixed effects. Recent works suggest that the presence of AB prior to ABM can be considered as a critical moderating factor that may account for these mixed results. Methods: We assessed AB among highly trait-anxious individuals (n = 77) using both a face-version and a word-version of the dot-probe task at multiple time points: two weeks before ABM (t1), just prior to ABM (t2), and after ABM (t3). All participants were submitted to an ABM procedure including facial expressions. Analyses focused on 2 components of AB prior to ABM: a stable component, representing variance shared between the two baseline points (t1 and t2), and a dynamic component, representing variance that is specific to that point (t1 or t2). Results: The stable component of AB at baseline predicted the intensity of AB after ABM (t3)
b s t r a c tBackground: Attention bias modification (ABM) aims to reduce attentional bias for threat (AB), thereby diminishing anxiety symptoms. However, recent meta-analyses indicated mixed effects. Recent works suggest that the presence of AB prior to ABM can be considered as a critical moderating factor that may account for these mixed results.
Methods:We assessed AB among highly trait-anxious individuals (n ¼ 77) using both a face-version and a word-version of the dot-probe task at multiple time points: two weeks before ABM (t 1 ), just prior to ABM (t 2 ), and after ABM (t 3 ). All participants were submitted to an ABM procedure including facial expressions. Analyses focused on 2 components of AB prior to ABM: a stable component, representing variance shared between the two baseline points (t 1 and t 2) , and a dynamic component, representing variance that is specific to that point (t 1 or t 2 ).Results: The stable component of AB at baseline predicted the intensity of AB after ABM (t 3 ) while the dynamic component did not. The dynamic component of AB at baseline positively predicts performance improvement during ABM procedure, while the stable component negatively predicted it. Limitations: The findings depicted above only appear with the face-version of the dot-probe task.
Conclusions:The present results highlight the contribution of both the stable individual differences and dynamic components of preexistent AB. They also show the importance of moving the conceptualization of AB beyond the group-based analysis by integrating the notion and the assessment of within-person variability.