A key affordance of both task-based language teaching (TBLT) and telecollaboration lies in possibilities created for synchronous oral interaction. Young beginners need opportunities to use the second language (L2) in meaningful contexts, and listening and speaking in such contexts has been shown to contribute to L2 acquisition (Dooly & Vinagre 2022). However, designing appropriate tasks which young learners can accomplish independently is challenging. Indeed, much previous research has focused on pedagogical rather than acquisitional considerations (Pennock-Speck & Clavel-Arroitia 2023). The present study is part of ongoing research which seeks to fill this gap by investigating learner autonomy and participant roles in task-as-process as opposed to task-as-workplan (Breen 1989), and extends our focus to learner-learner interaction in the absence of direct teacher supervision. Our paper offers a fine-grained analysis of learner-learner interaction in a synchronous task conducted via videoconference between two French primary school classes. Analysis focuses on a Who’s who information-gap task (Calvez et al., 2022), conducted in English as a lingua franca (CEFR, A1 level). Data include video and audio recordings of hors-champ interaction among four learners on one side of the exchange, pedagogical resources (learner ID cards and a listening grid), and post-task teacher and learner interviews. Our research questions are: 1.How do learners marshall a range of resources to open, manage, and close task iterations in the absence of direct teacher supervision? 2.How are L2 competences (listening, speaking, interacting) demonstrated in exchanges among local and remote learners? Close analysis of multimodal transcriptions highlights the importance of paper-based pedagogical resources and interlearner cooperation in the immediate hors-champ environment, allowing learners to successfully negotiate task phases towards a final goal of identifying remote learners. Our results demonstrate how learners orchestrate gaze, gesture, L1 resources, and webcam framing to negotiate and scaffold interactions and reach task outcomes.