What are the pre-requisites?
Visual prompts are not effective for everyone or every task.
Posted Mar 29, 2010
In following up on a previous blog entry, "Do all children learn with the same teaching procedures?" (http://www.psychologytoday.com/blog/radical-behaviorist/201002/do-all-children-learn-the-same-teaching-procedures) I'm going to discuss some nuances of effective instruction. While many children, especially children with ASDs, learn efficiently through visual prompts, they are not effective for everyone. First I'll give a little background informantion.
Video modeling is a teaching method involving visual prompting, that has been shown to be an effective procedure for teaching a variety of skills to children with autism, including; daily living skills (e.g., Shipley-Benamou, Lutzker, & Taubman, 2002), communication (Charlop-Christy, Lee, & Freeman, 2000) and play skills (e.g., MacDonald, Sacramone, Mansfield, Wiltz & Ahearn, 2009; Roberts, MacDonald, & Ahearn, 2007). Video modeling typically involves presenting a videotaped sample of models engaged in a specific series of scripted actions and/or vocalizations. The videotaped model is shown 2 or 3 times and then the child is given an opportunity to perform the modeled response. The usefulness of this form of prompting has been well established but not all children learn through video modeling.
Weiss and Harris (2001) suggested several prerequisite skills for video modeling which include one-step imitation, attending and an interest in videos. At NECC, we have been examining the prerequisites for video modeling. Towards this end we developed an assessment of tests to measure levels of performance on a variety of skills that could influence learning using video modeling and, initially found that poor responding on delayed match-to-sample tasks was associated with failure to learn using video modeling (Tereshko, MacDonald, & Ahearn, 2010). We've more recently expanded our skills assessments to focus more acutely on delayed imitation which also tends to be absent in children who do not learn with video modeling. Thus far our data suggest that memory, or remembering, is an important prerequisite for learning through video modeling. Though this is not a particularly surprising finding, it does establish a specific goal in instruction that might be easily obscured in focusing on the specific responses.
Delayed matching-to-sample (DMTS) is a widely used procedure to measure short term memory (Constantine & Sidman, 1975). Matching-to-sample is jargon for techniques that involve teaching learners that stimuli are related to each other. For example, a spoken word (e.g., "car) is related to a specific object (a car) and/or a written word (c-a-r). Constantine and Sidman found greater decreases in accuracy on DMTS in individuals with developmental disabilities than typically developing individuals. Though problems with remembering are not unique to autism it is likely the case that assessing memory skills and designing teaching procedures for individuals with developmental disabilities to promote them is necessary. Given this findings, it appears that memory plays an important role in learning using video modeling. In video modeling the child observes the video and then imitates the model. This requires that the child remember the observed actions for the duration of the video and the period of time between the video and the task presentation.
Remembering occurs naturally in many learning contexts. The control exerted by a history of consequences on responding is apparent in everyone's behavior but some gap in time between cues that signal what consequence will follow behavior often leads to a breakdown in behavior. Such prerequisites are uncovered in many skill areas. Another example is with picture activity schedules. Picture activity schedules are commonly used to cue children diagnosed with autism to perform tasks
independently (McClannahan, MacDuff, & Krantz, 2002). Activity schedules usually consist of binders with one picture per page that children are taught to open, turn the pages, look at the pictures, and engage in the corresponding task (MacDuff, Krantz, & McClannahan, 1993; McClannahan & Krantz, 1999). When children start learning to read, it may be developmentally appropriate to replace the pictures with printed words.
In another recently conducted study at NECC, we showed that children who worked effectively with picture activity schedule (they could look at pictures that represented activities and then gather the materials and engage with them) and read words, did not respond effectively to written words when they replaced pictures until they were taught the relations between the pictures and correponding written words (Miguel, Yang, Finn, & Ahearn, 2009). In this study, matching-to-sample training was used to teach the relations and once they mastered these, both of the children then responded to the written words as they did the pictures.
Persons with autism are often referred to as visual learners but that doesn't mean that every visual prompting technique is going to be effective for each person with autism. Perhaps the most important take home point from this post is that effective instruction requires that educators finely analyze the skills they are teaching. While many learners may be taught effectively through specific techniques, we may learn more about the skills embedded within more complex behavior when learning does not occur.