The current study succeeded in replicating and extending the results reported by Vassilopoulos et al. (2009). Compared to pre-training assessments, participants (94 primary school children between 10 and 12 years old) trained to make benign interpretations in both the imagery and verbal instruction conditions interpreted ambiguous events less negatively and anticipated less negative consequences of these events. The study also found that instructions to focus on the verbal meaning of the descriptions led to significantly greater decreases in negative interpretation and judgmental biases than did instructions to imagine the same hypothetical events. In addition, children in the verbal-instructions condition reported a significant decrease in social anxiety and the tendency to discount positive events compared with children in the imagery condition. Notably, the training conditions had differential effects on social anxiety, but not depressive symptoms. Thus, these findings cannot be explained by a general response bias. The findings suggest that interpretation-training in children can be effective with verbal instruction; however, further investigation is required to determine how to optimize the effectiveness of interpretation-training for children. In a benign interpretation-training paradigm descriptions of ambiguous hypothetical events were presented in a form requiring participants to endorse the more benign of two interpretations. The children were asked to either imagine these hypothetical events or to read the same descriptions while thinking about their verbal meaning. 1 table, 1 figure, and 41 references