“…A final line of research has examined effects for self-reinforcement and similar procedures when major parameters of external reinforcement are violated. For example, certain types of "self-punishment" procedures can be as effective as or more effective than self-reinforcement procedures even when the effects should be in the opposite direction (Castro, Perez, Albanchez, & Ponce de Leon, 1983;Castro & Rachlin, 1980), that delivering a "consequence" in these procedures before the behavior is more effective than delivering one after it (Nelson, Hayes, Spong, Jarrett, & McKnight, 1983), or that deprivation of the supposed reinforcer has no effect on the procedure . This type of research has greatly limited the useful scope of the concept of self-reinforcement in a theoretical sense, but with the exception of Rachlin's model (to be discussed later) has not suggested an alternative model.…”