Background and Purpose: The provision of feedback is important for effective skill learning. The purpose of this study was to examine the nature of feedback provided during the practice of everyday activities (such as standing up, walking, and reaching and grasping objects) during stroke rehabilitation, both when the therapist was present and when the patient was practicing alone. Method: Design: A cross-sectional observational study of the feedback received during rehabilitation by people who had had a stroke was conducted. Forty unique patient-therapist dyads were observed during 30minutes of actual practice of everyday activities with data collected through behavioural mapping. The following was recorded: the activity practiced, whether the therapist was present, whether feedback was provided verbally or by equipment, and the content of feedback. Participants: A sample of all therapists providing rehabilitation within one Australian health service and their patients who had had a stroke. Measures: Quantity, frequency, mode (verbal or equipment) and content (information feedback, motivational statements, unrelated or none) of feedback during the practice of everyday activities were determined. Results: For 68% of the time that patients were practicing activities, they received ≥1 occasion of feedback/minute. When the therapist was present, the frequency of motivational statements was more than four times greater, at 1.32 (SD 0.6) occasions/minute, than information feedback. For 25% of the time, the therapist was not present, and no feedback was provided. Conclusion: Given the importance of specific content for learning, therapists could replace some motivational statements with information feedback. When practicing alone, information feedback could be provided by commercially available biofeedback or customized equipment.