An electronic device monitors audible output for media content being presented in its vicinity. It detects a volume change event within the monitored audible output. It obtains identification of the media content. Based on the identification, the device obtains a list of expected volume change events within the media content. It determines whether the detected volume change event correlates with one of the expected volume change events. In accordance with a determination that there is no correlation, the device determines that the detected volume change event comprises a volume change event from a user of the device. It analyzes the detected volume change event to determine user interest in the presented media content. The determined level of interest is sent to a server system. The device receives from the server system one or more recommendations based on the determined level of interest, and presents to the user the recommendations.