This experiment tested the effect of a dual-task on time reproduction in 5- and 8-year-olds. Children had to reproduce a stimulus duration lasting for 6 or 12 s, during which they either did or did not perform a concurrent non-temporal task (i.e. picture naming) both in low (LA) and high (HA) attentional demand conditions. The results showed that children reproduced shorter durations in the dual-task than in the single-task condition, whatever the duration value used. However, this shortening effect was greater in the 5-year-olds than in the 8-year-olds. Furthermore, in the 5-year-olds, temporal reproductions were significantly shorter in both dual-tasks (LA or HA) than in the single-task, whereas, in the 8-year-olds, differences reached significance only between the HA dual-task and the single-task. In the non-temporal task, the proportion of naming errors was also greater in the dual-task than in the single-task, especially under high attentional demand, but it did not significantly differ between the two age groups tested.