Use of video modeling to increase generalization of social play by children with autism

Journal of Speech-Language Pathology and Applied Behavior Analysis, The, Summer, 2007 by Christos K. Nikopoulos

Thus, video modeling was used as an antecedent strategy (Cuvo & Davis, 1998; Heflin & Alberto, 2001) which exerted stimulus control over children's performances. It might be a case that the relevant stimuli--toys, model, & researcher--had been captured close enough together in terms of the two-dimensional TV screen, enhancing the acquisition of the stimulus control of the successful imitative responding (Charlop-Christy et al., 2000). However, after a short video modeling intervention children's behavior changed even in the absence of any video display (i.e., all generalization sessions & follow-ups) providing evidence that children's performance was not solely under stimulus control of the video display, but perhaps also by the toy(s) or the researcher or peers. That the frequency of responding remained high in the presence of physically different toys during generalization is perhaps indicative of a functional equivalence class (Masia & Chase, 1997; McGuigan & Keenan, 2002).

Following the above assumption that the toy(s) or the researcher/peers could act as discriminative stimuli for responding, the video modeling intervention could be explained within the paradigm of establishing operations. The term 'establishing operation' was first introduced by Keller and Schoenfeld (1950) and further explored by Michael (1993, 2000). It refers to those events or conditions that affect an organism by momentarily altering the reinforcing effectiveness of other events and also the frequency of occurrence of that part of the organism's repertoire relevant to those events as consequences. During the baseline sessions in these studies, children did not use the toys in interactive play behavior with the experimenter. However, that behavior did occur after video modeling had been introduced. Consequently, it seemed that the video display altered the reinforcing effectiveness of the toys (e.g., Koegel, Camarata, Valdez-Menchaca, & Koegel, 1998; Stahmer & Schreibman, 1992) and thus, the occurrence of social play remained in the children's repertoire at high rates.

Participants' performance during the video modeling procedures could have been influenced by their reinforcement histories with respect to the same or similar stimulus materials (e.g., toys) or to the modeled responses (e.g., social interaction, reciprocal play). Thus, a history of reinforcement delivered for some behaviors could function to maintain an entire class of imitations (Baer, Peterson, & Sherman, 1967). This is consistent with the suggestion that an imitative behavior cannot be explained just as a separate stimulus-response relationship but rather as an operant class of responses. That is, as soon as an imitative response of a person is reinforced, then that person will tend to imitate other behaviors, even if they contain no apparent elements in common with the imitative behavior that was reinforced (Martin & Pear, 2006). Even, a history of intermittent reinforcement for imitation can be responsible for a learner's imitation of novel behaviors. In fact, Skinner (1953) proposed that a behavior can become stable or resistant to extinction when it has been reinforced intermittently. From that perspective, targets of the video modeling procedures described in the aforementioned studies could be viewed as explicit examples of generalized imitation. This occurs when an individual imitates a new response without direct reinforcement (Baer & Deguchi, 1985); so did occur in both studies.