In this presentation we present the design of a serious game for young learners that was inspired by VanPatten (2002)'s Processing Instruction (PI) model, which we believe is compatible with the DUB approach. The role of grammar in foreign language teaching has long been controversial, from the so-called "grammar-translation method" modelled after the teaching of classical languages to current task-based approaches. However, discussions of the place of grammar in teaching usually focus on productive skills. By contrast, input-based theories of L2 acquisition stress the importance of processing grammatical elements for meaning when receiving input in order for acquisition to take place. VanPatten's PI model states that because morphosyntactic elements are redundant and/or not salient, learners do not notice them and must therefore be taught to pay attention to them and process them for meaning. This presentation focuses on the use of the Processing Instruction model in the design of a serious game for 6-year-old beginners in English. Most studies using the PI model have used older learners and studied the acquisition of inflections (eg, Kasprowitz & Marsden 2017). With young beginners, the target feature chosen was the syntax of noun phrases (five shirts, green snake), in which the meaning of at least two lexical elements is integrated into a coherent whole. Previous studies have shown that, when listening, French elementary school students tend to focus on one known lexical item and ignore the others (Audin 2003). Van Patten's model consists of a first phase in which learners receive explicit instruction about the target feature. Because we believe that grammar patterns are essentially form-meaning pairings, just like lexical items, we contend that explicit instruction should mainly be about the provision of meaning rather than abstract rules. Since we are dealing with a simple integration operation that also exists in L1 French, the first phase was replaced with the establishment of form-meaning links for lexical items. The second phase uses referential activities that require processing of the target feature. For this central phase, we used multiple choice questions with oral stimuli associated with graphic representations (requiring learners to integrate the meaning of several lexical items). The last phase consists of activities exposing learners to more instances of the target feature in different contexts. In our game, children were exposed to the target structure in the narrative and dialogues that accompanied the activities. An experiment was conducted with 690 first graders and showed that the experimental group outperformed the control group in the comprehension of individual words as well as word combinations. They were also able to produce a few word combinations. Audin, L. (2003). L'apprentissage d'une langue étrangère à l'école primaire : Quel(s) enseignement(s) en tirer ? Les Langues modernes, 3, 10‑19. Kasprowicz, R., & Marsden, E. (2017). Gaming Grammar : Designing and Evaluating a Digital Game for Learning L2 French Morphosyntax. EuroCALL 2017. VanPatten, B. (2002). Processing Instruction : An Update. Language Learning, 52(4), 755‑803