Background <p>The present study represents the first attempt to identify potential psychological predictors of change in interpretation bias and state worry following a single session of cognitive bias modification of interpretations (CBM-I), which is a computerised training to modify interpretation bias, using two samples of worry-prone young adults.</p> Methods <p>High worriers with a history of anxiety and/or depression (Study 1; <i>N</i> = 83) and worry-prone individuals (Study 2; <i>N</i> = 146) completed a single session of CBM-I training. Pre-training, they completed a battery of self-report measures and tasks that assessed key moderators of CBM responses: attentional control, cognitive flexibility, sensitivity to reward, and imagery ability. Levels of interpretation bias and state worry were also assessed pre and post training to index CBM-related changes.</p> Results <p>In study 1 a greater ability to imagine positive events and lower levels of cognitive flexibility at baseline were associated with a greater increase in positive interpretation bias. Lower levels of cognitive flexibility pre training were associated with greater reduction in state worry post training. In study 2 higher levels of cognitive flexibility and lower levels of responses to positive affect at baseline had greater increase in positive interpretation bias, but not reductions in worry post training.</p> Conclusions <p>In both studies, attentional control was not a significant predictor of change in interpretation bias or state worry following a single session of CBM-I training. There were differences in the role of cognitive flexibility, emotion-focused rumination and positive mental imagery in the two samples. Given non-replications, individual differences that predict change in near and far transfer outcomes require further research. Nevertheless, the present findings provide insights to improve the outcome of CBM-I. For instance, incorporating a longer imagery training or cognitive flexibility training may be helpful.</p>