Systems and methods are described for performing an action related to an identifier for a recommended media asset presented to a user, based on the user's detected emotional indicator. An emotional indicator of the user is detected based on an image of the user captured while an identifier for an initial recommended media asset is generated for presentation, and an action related to such identifier is performed based on the detected emotional indicator. Systems and methods are also described to enable determining a first recommended media asset and a second recommended media asset based on respective user profiles of a first user and the second user, each being identified by facial recognition. Respective identifiers selectable to access the first or second recommended media asset are generated for display, and a recommended media asset associated with a selected identifier is generated for display.