| Anxious adults show a robust attentional bias toward negative information. However, little is known about the role of biased attention in late-life anxiety or about the neural substrates of any such bias. In Study 1, 60 older adults (age 60+) completed the emotional Stroop (eStroop) task to assess attentional bias. Participants were stratified into high- (n=20) and low-worry (n=20) groups on the basis of a self-report measure, the Penn State Worry Questionnaire. During a 4-color eStroop task, as predicted, anxious older adults exhibited a reaction time (RT) bias towards negative words, while non-anxious participants showed a bias away from negative words. Furthermore, the non-anxious group showed a bias towards positive words, consistent with an established positivity effect in older adults; while the anxious group showed a bias away from positive words. In Study 2, treatment-seeking late life generalized anxiety disorder (GAD) patients (n=16) and non-anxious age-matched controls (NACs; n=12) completed the eStroop task in a magnetic resonance imaging (MRI) scanner. Functional MRI analyses were conducted to test the predictions of a neurocognitive model of attentional bias emphasizing deficiencies in top-down attentional control regions (prefrontal cortex [PFC]). As in Study 1, GAD participants showed a bias towards, while NACs showed a bias away from, negative words. Functional MRI data revealed that during trials of negative (in comparison to neutral) words, NACs showed increases in dorsomedial and ventrolateral PFC activity, coupled with decreases in task-irrelevant bottom-up emotional processing regions (amygdala, hippocampus). By contrast, GAD participants showed dorsolateral PFC decreases during negative words, and no differences in amygdala activity across word types. Across all participants, RT bias toward negative words was correlated with decreases in PFC regions. A positive correlation between RT bias and amygdala activation was also present, but this relationship was mediated by the PFC. During positive words, no RT bias was found in either group, and fMRI data revealed no PFC deficits in GAD. These results are most consistent with models of anxious attentional bias that emphasize deficiencies in top-down control over negative emotional material. Strategies to enhance top-down attentional control may be particularly relevant in late-life GAD treatment. |
