Self-restraint has been noted in studies of self-injurious behavior (SIB) in mentally retarded individuals. These studies have investigated self-restraint in the context of SIB, and as a result it has been difficult to isolate the function of self-restraint within this complex clinical situation. This article summarizes and critically analyzes the literature on self-restraint, particularly its identification, etiology, modification, and relationship to SIB. At present there is some empirical evidence to suggest that self-restraint is negatively reinforced by the termination or avoidance of SIB. Other studies have demonstrated that self-restraint can develop stimulus control properties for the absence of SIB. Finally, it has been suggested that self-restraint may function independently of SIB and be maintained by escape from demands or by conditioned reinforcement. Clinical concerns include the need for a functional analysis of self-restraint, the undesirable effects on SIB of reducing self-restraint, and the necessity of replacing self-restraint topographies with ones which do not interfere with daily activities.