Modality switching plays an important role in children’s language processing. During everyday life, especially during school hours, spoken language is often supplemented by visual information, so that switching between auditory, visual, vocal and manual processing becomes necessary. Previous studies with adults have shown that modality switching is influenced by modality compatibility. Specifically, switching between relatively incompatible combinations (i.e., auditory-manual and visual-vocal) takes more time, is more error prone, and leads to higher switch costs compared to switching between relatively compatible combinations (i.e., auditory-vocal and visual-manual). So far, however, data are limited to adults and underlying cognitive mechanisms are still under debate. The present study contributes to a better understanding by providing the first data from children (first to fourth grade). In two experiments, children switched between compatible and incompatible modality combinations while deciding whether the presented pictures and tones (Experiment 1; N=32) or gestures and spoken words (Experiment 2; N=32) represent an animal or not by pressing a yes- or no- button or saying “yes” or “no”. Both experiments revealed longer reaction times, higher error rates, and higher modality switch costs for switching between incompatible combinations compared to compatible switching. The results thus show for the first time that modality-specific effects on cognitive control processes already exist in childhood.