Video Modeling

Video Modeling to Improve Academic and Social Skills for Children with Autism
Children with autism spectrum disorder (ASD) often require systematic and intensive interventions in order to develop appropriate social and academic behaviors. One intervention that has been used with some success on improving a variety of behaviors is video modeling (Acar & Diken, 2012; Wang, Cui, & Parrila, 2011; Wilson, 2013). Video modeling involves the process of creating a video of a person or persons exhibiting a desired behavior and subsequently showing the video in a planned, systematic manner to the individual in need of intervention with the intent to measure imitation of the desired behavior by the viewer (McCoy & Hermansen, 2007). This intervention has

yielded promising results among a variety of participants with ASD on different targeted areas (Bellini & Akullian, 2007; Gul, & Vuran, 2010; Shukla-Mehta, Miller, & Callahan, 2010).

The majority of studies on video modeling to date have focused on the development of social skills. Play skills are a common focus. Taylor, Levin, and Jasper (1999) found that video modeling improved the spontaneous play comments in the two young participants in their study. Nikopoulos and Keenan (2003) conducted a study measuring social play initiation and reciprocal play skills among seven participants with ASD. After treatment with video modeling, four of the seven participants demonstrated growth in the targeted areas. The same authors conducted a subsequent study in which three participants were treated with video modeling. An increase in both the initiation of social play and reciprocal play skills were observed in each of the three participants (Nikopoulos & Keenan, 2004). Nikopoulos and Keenen partnered together again

in 2007 to conduct a third study measuring the effect of video modeling on play skills. Two experiments were conducted as part of their third study, examining video modeling used with a total of four participants. Results from the 2007 study suggest that video modeling was successful in improving social initiation, reciprocal play, object engagement, and imitative response among all four participants (Nikopoulos & Keenan, 2007). D’Ateno, Mangiapanello, and Taylor (2003) measured the effect of video modeling on the complex play sequences performed by one participant. After treatment, both motor and verbal responses increased in the participant. In another study, reciprocal pretend play skills were measured in four participants grouped in pairs; each pair consisted of one child with ASD and a neurotypical peer. Play skills increased in each of the four participants after video modeling was introduced (MacDonald, et al., 2005).
Social communication skills are another common set of behaviors measured in previous research on video modeling. Charlop and Milstein (1989) observed an increase in the number of verbal responses and questions asked by their three participants after the introduction of video modeling. Maione and Mirenda (2006) indicated that the combined treatment of video modeling with video feedback and prompted improved the social language skills produced by the participant in their study. In another study, video modeling was combined with Social Stories™ as a treatment package for two participants. The number of greetings by the participants directed to school staff increased after the treatment was introduced (Kagohara, et al., 2013). A similar treatment package was used in Scattone’s study (2008). Social Stories™ and video modeling were combined as an intervention on the conversation skills of one participant. In this study, eye contact and initiation of conversations improved after the intervention was introduced. Another conversation skill—smiling—did not improve. Apple, Billingsley, and Schwartz (2005) conducted a study in which two separate experiments measured the number of compliments given by a total of five participants. In the first experiment, two participants were treated with video modeling combined with a reinforcement schedule. The second experiment involved the use of video modeling combined with self-management among three participants. These combined treatment packages yielded an increase in the number of compliments given by each of the five participants. Charlop, Gilmore, and Chang (2008) found that the conversational responses were more varied in their two participants after video modeling had been introduced. Sherer, et al., (2001) found that three out of their five participants demonstrated improved conversation skills after video modeling had been introduced. The mixed results were attributed to the visual learning ability of the participants. Tetrealt and Lerman (2010) found that scripted responses and eye contact improved in their three participants after being treated with a combined package of video modeling and reinforcement.
Other social skills have been measured in research on video modeling. Reeve, et al. (2007) studied the effect of video modeling combined with reinforcement and prompts on the helping behaviors of four participants. Results from this study indicate a positive correlation between the intervention and target behavior. Simpson, Langone, and Ayres (2004) looked at the effect of video modeling embedded in computer based instruction on a variety of social skills. Among four participants, sharing behaviors, complying with teacher directions, and social greeting skills improved after implementation of the intervention.
Independent functioning skills have been another area of focus in the literature about video modeling. Daily living skills (i.e., setting a table, mailing a letter, pet care, and squeezing orange juice) were measured in three participants in one study. After the introduction of video modeling, these skills improved in each of the participants (Shipley-Benamou, I.utzker, & Taubman, 2002). Self-help skills such as putting on a watch, zipping a jacket, and cleaning a pair of sunglasses improved in three participants who were treated with video modeling in another study (Norman, Collins, & Schuster, 2001). Haring, et al. (1987) conducted a study in which video modeling was combined with a reinforcement system as an intervention on the purchasing skills of three participants. Each of the participants demonstrated growth in their purchasing skills in the community setting after implementation of the intervention. Alcantara (1994) combined video modeling with prompting and reinforcement as an intervention on the purchasing skills of three participants; each participant demonstrated an improvement in his or her ability to independently purchase items from a store.
In one study, a variety of behaviors were targeted among five participants, with each participant being treated for one or two unique behaviors. One participant demonstrated a need for development in the expressive labeling of emotions; another participant exhibited a need for intervention in independent play skills. A third participant demonstrated a need for growth in oral comprehension and spontaneous greeting. Two other participants exhibited a need for intervention in conversation and play skills, and self-help and play skills, respectively. After being treated with both video modeling and in vivo (face-to-face) modeling, results indicate that video modeling was more effective in the generalization of the targeted behaviors by each of the participants (Charlop-Christy, Le, & Freeman, 2000).
When creating the videos to be used in the implementation of video modeling, one may elect to use adults or peers as the models for the appropriate behavior. Other forms of video modeling including self-modeling, in which the individual in need of intervention is filmed displaying the desired behavior and point-of-view modeling, in which the camera is positioned at a subjective viewpoint so that the resulting product is viewed at the eye level of the viewer and the viewer experiences the video in the same perspective in which he or she would experience the situation. Mixed models are also sometimes used, combining two or more of these types (McCoy & Hermansen, 2007).
The most common model type used in the literature reviewed in this study involved the use of adults as models. Charlop-Christy, Le, and Freeman (2000) indicated that they deliberately chose to use adults over peers, as previous research (Ihrig & Wolchik, 1988, as cited in Charlop-Christy, et al., 2000) suggests that participants with ASD demonstrate similar outcomes regardless of model type and that young models are more difficult to train than adults. The study by Sherer, et al. (2001) supports this hypothesis that model type has little effect on outcomes on participants with ASD. Other studies that used adults as models include those by Alcantara (1994), Charlop, et al. (2008), Charlop and Milstein (1989), D'Ateno, et al. (2003), MacDonald, et al. (2005), Maione and Mirenda (2006), Reeve, et al. (2007), and Scattone (2008). Four studies reviewed in this paper used neurotypical peers as models: Apple, et al. (2005), Haring, et al. (1987), Nikopoulos and Keenan (2007), and Simpson, et al. (2004). Three of the reviewed studies used point-of-view modeling, in which the camera faces away from the model so that the viewer experiences the video in the same way in which he or she would experience the targeted situation. In these studies, the videos were recorded at the child’s eye level from a subjective viewpoint (Norman, et al., 2001; Shipley-Benamou, et al., 2002; Tetreault & Lerman, 2010).
Several of the studies used a combined approach when selecting models for the videos.

Nicopoulos and Keenan (2003) used familiar adults, unfamiliar adults, and peers in their study. The same two authors used neurotypical peers paired with adults in their 2004 study (Nicopoulos & Keenan, 2004). Sherer, et al. (2001) compared the use of self-modeling to video modeling with peers as modeling. They found that both types of models yielded similar outcomes. In the study by Taylor, Levin, and Jasper (1999), siblings of the participants were paired with adult models in the videos. In the study by Kagohara, et al. (2013), humans were not used as models in the videos; the authors created videos using cartoon models. This method yielded positive outcomes among the

participants.
The purpose of this study is to examine the effect of video modeling with adult models on the social and academic skills of one participant. Combining these two behaviors will contribute new information to the current body of research. This author will seek to answer the following question: What is the effect of video modeling on the conversational turn-taking and on-task behaviors of a second-grade student with autism spectrum disorder?