This paper comes from a research entitled "xCALE" (eXplaining Competency and Autonomy development in Learning Environments) and aims to analyze how learners self-regulate during an online training. In a context where lifelong learning and online training, researchers went to understand how students learn in these formations. Self-regulated learning is a key to understanding success in an online environment (Kizilcec et al., 2017). We present an empirical study on self-regulation strategies in adult training using online resources. Self-regulation is defined as “the degree that they are metacognitively, motivationally, and behaviorally active participants in their own learning process” (Zimmerman, 1989, p.329). We examine two self-regulated learning strategies: seeking information and seeking social assistance (Zimmerman and Pons, 1986). The objective is to identify and describe the practices of learners asking for help during the use of online training. Our questions are: which helpers are mobilized when learners ask for help? Why do they choose online support or social assistance? On what do learners rely when choosing to seek for help? The study is based on 23 interviews with adults in digital training. The analysis underlines a diversity of practices of help requests and emphasizes that there are favorable conditions for requesting help. Results show the existence of rules of preference to determine the source of help the most adapted, it depends on their need and to the context of the request for help.