Abadir, C. M., DeBar, R. M., Vladescu, J. C., Reeve, S. A., & Kupferman, D. M. (2021). Effects of video modeling on abduction‐prevention skills by individuals with autism spectrum disorder. Journal of Applied Behavior Analysis, 54(3), 1139-1156. https://doi.org/10.1002/jaba.822

 

Reviewed by Nicole Radzilowicz, MEd, BCBA and Mary Jane Weiss, PhD, BCBA-D
Endicott College

Why research this topic?

Procedures to teach specific skills for individuals with autismAccording to the National Center for Missing and Exploited Children (NCMEC), in 2019, over 1,300 instances of attempted abduction occurred in the United States (NCMEC as cited in Abadir et al., 2021). Abduction attempts occurred most often during transitions between locations (e.g., as children transition between home and school). Verbal lures, such as phrases asking a child for assistance or informing a child that a family member sent them, are one strategy used in abductions and attempted abductions of elementary and middle-school children. There is evidence that a large number of child abductions involve family members. For example, in 2016, 60% of the AMBER alerts occurred due to abduction by family members (NCMEC as cited in Abadire et al., 2021). Limited information exists on strategies to prevent child abduction from familiar individuals. This study addressed potential abduction scenarios from both familiar and unfamiliar people.

There are many commonly seen characteristics of autism that increase the risk of abduction (e.g., difficulties in reading social cues and comprehending danger). Autistic individuals may not engage in pertinent behaviors during unsafe situations such as accurately identifying danger, being aware of their immediate environment, struggling with social interactions, advocating for themselves, and responding to social cues. Furthermore. autistic individuals may engage in bolting or wandering away from caregivers, which also places them at increased risk of getting lost in the community. Due to the prevalence of child abduction and the increased risk for individuals with disabilities, it is necessary to teach autistic individuals how to respond during potentially dangerous situations to help decrease the likelihood of child abduction.

What did the researchers do?

The researchers explored the effects of video models to teach autistic individuals how to respond to verbal lures from familiar individuals and strangers. The participants in the study were four twelve-year-old autistic males who attend a self-contained educational classroom within a public school. To be included in the study, each participant was required to have the following pre-requisite skills: discrimination between known individuals and strangers, following verbal directions as a listener, following directions to “wait” in an area for at least one minute, have the ability to wear a camera, and strong imitation skills. Each of these skills was assessed by the researcher. Discrimination was assessed by directing the participants to sort images of familiar and unfamiliar adults. Following verbal directions was assessed by giving the participant instructions to perform an action and say a word when they hear a specific word. Waiting was assessed by the experimenter giving the direction to wait to the participant, and then moving away from the participant. The ability to tolerate wearing a camera was assessed by having participants wear a camera on a lanyard for 5-minute intervals.

During the training phase, the researchers showed participants short video clips (ranging from 7 to 16 seconds) of individuals of a similar age engaging in the target skills. Each peer in the video model was within 2 years of age of the participants. The videos demonstrated different situations, such as the peer being approached by a familiar person or a stranger. The researchers used a variety of lure phrases within the videos, such as “Your mother wants me to take you home,” or “Your teacher says it is time to go.” If the peer was approached by a stranger, the peer replied, “No,” moved away from the stranger, approached the experimenter, said “Help,” and pointed towards the stranger. If the peer was approached by a familiar person, the familiar person would either provide the correct code word or an incorrect code word. The peer then responded correctly depending on the situation. If the correct code word was provided, the peer replied, “Okay,” and remained in close proximity to the familiar person. If the incorrect code word was provided, the peer replied, “No,” moved away from the individual, approached the experimenter, said “Help,” and pointed towards the familiar individual. Following the video model, the researchers took each participant to practice a situation similar to the one seen in the video. After the participants met master criteria to responding to lures, the video models were removed in the post-training phase. Similar situations were tested without prior exposure to the video models. To assess generalization, researchers also conducted trials in locations that were not used during training.

What did the researchers find?

The researchers found that during the baseline condition, without any training, individuals engaged in low rates of correct responding. During baseline sessions, the experimenter directed the participant to wait in an area. The experimenter then moved away and pretended to complete a task. The participant was approached by a confederate stranger, who used a verbal lure to try and get the participant to leave the area with them. If the participant did not respond to the lure for 5 seconds or attempted to leave with the stranger, the stranger terminated the session. If the participant asked follow-up questions, the stranger repeated the lure up to two more times and then terminated the session. The low baseline data indicate a serious deficit in participants’ safety skills. The use of video models increased correct responses across all participants and situations. Once the video model was removed, three out of the four participants continued to respond correctly. One participant required additional training. The additional training helped to temporarily increase his correct responses again; however, the skill again decreased. For two of the four participants, an additional direction, “Do what the person in the video did,” was necessary for the participant to replicate the behavior in the video during trials. Video modeling was effective in increasing correct responding to verbal lures from familiar individuals and strangers.

What are the strengths and limitations of the study?

The strengths of the study include the ease of the intervention to teach important safety skills. Video models can be created quickly and implemented across a variety of environments and skills. The study focused on generalization of the skill by conducting trials in areas not targeted during teaching. This demonstrated the ability of video models to be effectively used to teach individuals to respond in a variety of situations. The study also demonstrated maintenance of skills in 3 out of the 4 participants across one-week, three-week, or a four-week probe. Given that generalization and maintenance are difficult for individuals with autism, the inclusion of these measures is notable. The study was conducted as a concurrent multiple probe design, which allowed training to occur across participants without a large passage of time in between the training of different participants. Social validity measures were conducted to identify the social acceptability of goals, procedures, and outcomes. Compared to the baseline, the social validity measures following mastery of the skill were rated much higher, demonstrating that the participant responded correctly during the presented situations. High social validity ratings were also noted for the importance of the target skills.

Some limitations of the study include the use of role-playing scenarios, as these create an assessment environment that is not naturally occurring. However, given the content of the training, this was necessary. Still, there are inherent limitations to analog training. Although there were high levels of correct responding to the role-play scenarios, that may not be mirrored in more naturally occurring situations when individuals are not aware that their behavior is being observed and assessed. The presence of an experimenter and the camera worn by participants may have influenced their responses. Instead of the participants responding solely to the stranger or familiar person and the verbal lures, the presence of the experimenter and camera may have suggested certain responses. An additional limitation for the study is that all participants were required to have discrimination skills to discriminate between familiar individuals and strangers. This level of discrimination is complex and limits the extent to which these findings would be relevant for individuals who do not have this skill. Another limitation is that most confederates who provided the verbal lures were female, as the researchers struggled to recruit male confederates. The inclusionary criteria for participation included some skills that many autistic children do not have. This limits the external validity of the study. Individuals without these prerequisite skills would require more robust intervention to target the needed discrimination skills. Since the skill taught is important and a life-long skill, it would be helpful to see if the skill maintained past the three-week and four-week probes conducted in the study.

What do the results mean?

These results demonstrate that it is possible to teach autistic individuals safety skills through the use of video modeling. Autistic individuals successfully attended to short video clips of similar-aged peers demonstrating safety skills. They imitated the responses during role-play scenarios and demonstrated the skills in novel locations. Video modeling was shown to be an important tool to teach skills that are challenging to recreate in the natural environment, such as abduction prevention skills.

Citation for this article:

Radzilowicz, N., & Weiss, M. J. (2023). Research Synopsis: Effects of video modeling on abduction‐prevention skills by individuals with autism spectrum disorder. Science in Autism Treatment, 20(11).

Related ASAT Research Synopses: ​

Other Related ASAT Articles: ​

Related Media Watch Letters

#BehaviorAnalysts #Community #Preschoolers #Safety #SchoolAge

Print Friendly, PDF & Email