By connecting objects across space and time our visual system is able to perceive objects in motion and to structure our environment. Studies with apparent motion displays in adults have shown that spatio-temporal as well as feature information is used to solve the correspondence problem, i.e., the integration of different object instances into one percept . The aim of our study was to examine how this integration process develops, specifically, whether children solve correspondence in the same way as adults. We used the Ternus Display, an ambiguous apparent motion display, in which two objects are presented in a first frame, followed after a variable inter-stimulus interval (ISI) by two objects shifted by one position to the side. This display is ambiguous, as it can be perceived as both objects moving together (group motion) or one object moving across the other (element motion), depending on how correspondence has been solved. We manipulated the ISI and used a form-based feature bias, such that the form of the objects across frames was either compatible with group or with element motion. We asked children between the age of 5 and 7 to report their motion percept. Our results showed an influence of the ISI and feature information. In contrast to our adult control group, however, the feature effect was limited to the group bias condition. Our results suggest that children solve correspondence based on ISI and feature information, but not in the same way as adults do.