Successful self-regulated learning in a personalized learning environment (PLE) requires self-monitoring of the learner and reflection of learning behaviour. We introduce a tool called CAMera for monitoring and reporting on learning behaviour and thus for supporting learning reflection. The tool collects usage metadata from diverse application programs, stores these metadata as Contextualized Attention Metadata (CAM) and makes them accessible to the learner for recapitulating her learning activities. Usage metadata can be captured both locally on the user's computer and remotely from a server. We introduce two ways of exploiting CAM, namely the analysis of email-messages stored locally on a user's computer and the derivation of patterns and trends in the usage of the MACE system for architectural learning.1