Drift-Diffusion Modeling of Attentional Shifting During Frustration: Associations With State Frustration and Trait Irritability

Abstract

Irritability, a prevalent and impairing symptom in many mood and anxiety disorders, is characterized by aberrant responses to frustrative nonreward. Past research investigating irritability have used a cued-attention task with rigged feedback, the affective Posner task (AP), to induce frustrative nonreward. Previous studies have not been successful in linking differences in self-reported irritability to traditional AP metrics (i.e., reaction time and accuracy). Computational modeling, via the estimation of parameters reflecting latent cognitive processes, may provide insight into the cognitive mechanisms of irritability and reveal potential targets for mechanism-based interventions. This study applied the drift-diffusion model (DDM) to the AP to determine if DDM parameters are associated with individual differences in irritability. Young adults (N = 152, M_age = 20.93 ± 1.98) completed the AP and self-reported state frustration and trait irritability. Multiple linear regressions were used to evaluate whether DDM parameters better predict state frustration and trait irritability over traditional AP metrics. Higher state frustration was predicted by lower decision threshold during the frustration block and larger decrease in this parameter between nonfrustration and frustration blocks, over traditional AP metrics. These findings demonstrate the potential of applying the DDM to study frustrative nonreward in healthy adult populations. The utility of DDM awaits validation in populations with clinical levels of irritability.