Advantages in diverse aspects of cognitive functioning have been reported in early bilinguals (Bialystok, 2011) as well as in children frequenting an early bilingual immersion school program (Nicolay and Poncelet, 2015). However, during the last decade, some studies failed to replicate these advantages. Currently, the presence of cognitive benefits in children frequenting an immersion program remains debated. The lack of consistency between the studies could come from the fact that time spent by children within the immersion program is variable from (...) one study to the other and that studies used different tasks to assess the same cognitive function. The main aim of the present study was to determine how time spent in immersion affects the emergence of cognitive advantages along the primary schooling. We compared 196 immersed Dutch-speaking children since they were 5 years old and 195 non-immersed French-speaking children, from different grades of the primary schooling (i.e., at 6, 7, 8, and 12 years old) by using the same attentional and executive tasks as those used in previous studies having shown a bilingual advantage. Furthermore, these groups were matched on a set of variables known to influence cognitive functioning. After 1, 2, and 3 years of enrolment in this program, performances of immersed compared to non-immersed children did not differ for any task. However, after 6 years, immersed children outperformed non-immersed children on the cognitive flexibility and the working memory tasks. These results show that, in French-speaking children immersed in Dutch, cognitive advantages could depend on the length of time spent in immersion since they are not present at the beginning (after 1, 2, and 3 years) but seem to emerge at the end of it (after 6 years). In contrast, in previous studies conducted in English immersion, advantages appear at the beginning of the primary schooling but are absent at the end of it. Furthermore, these results suggest that the emergence of cognitive advantages may vary depending on the second language learned. The results are discussed in terms of linguistic characteristics and status of the languages at stake. (shrink)