The sensitivity of children to the production of false memories from a Deese-Roediger-McDermott (DRM) paradigm appears to be low. Activation association theory suggests that the associative network plays a central role in the production of DRM false memories. However, relatively few studies have created child-specific lists. One hundred thirteen children with an average age of 9 years and 7 months were solicited to create DRM lists via a lexical association task. These lists were then tested on a sample of 42 children with an average age of 9 years and 6 months. Depending on the lists used, the children produced a proportion of false memories ranging from .04 to .46. The use of lists specifically created by and for children seems to favour the emergence of false memories. The results are consistent with activation explanatory theories of DRM false memories. However, confidence in their response is higher for correct answers than for errors. Public Significance Statement We can create false memories, that is, the memory of an event that never happened, in participants using word lists constructed in such a way that all the words in the list converge towards the same associated concept. To do so, these lists need to be designed specifically for the population in which we want to create these false memories. This study presents the creation of lists for children aged 8-10.