An experiment with 150 children aged 5-7 and 10-12 revealed that children's reactions to violent TV shows differed based on when they received negatively valenced, evaluative mediation. Results also showed that the effect of mediation's timing consistently depended on children's age. Older children responded better to pre-exposure mediation than they did to postexposure mediation. In some cases, older children exhibited reactance to postexposure mediation. For younger children, on the other hand, mediation given at any time was more beneficial than no mediation at all. Therefore, the timing of active mediation may be an important factor in interpreting research and in the design and implementation of mediation interventions.