Saturday, May 22, 2010

GrammarTrainer and Verbal Behavior

I've just discovered that Grammar Trainer has been listed as a resource on the Verbal Behavior website. This is good news, not just for me, but also for addressing what I've always considered a weak point in Applied Behavioral Analysis (ABA), one of the standard therapies for autism, which includes Verbal Behavior as its approach to language teaching. 


In some ways, GrammarTrainer and ABA have always been very much on the same page.  Both teach by breaking things down into small pieces and teaching those pieces one at a time, sequentially and incrementally.  Both include initial modeling, gradually fading of prompts, and repetitive practice throughout.

What's different is in how the two therapeutic approaches conceive of grammar.  ABA/Verbal Behavior follows a behaviorist model of grammar that dates back to B.F. Skinner. GrammarTrainer follows the modern, structural approach that dates back to Skinner's rival, Noam Chomsky. The problem with the behaviorist model is that it fails to capture the structural complexity of grammar.  Verbal Behavior, traditionally, is all about categories, and not about hierarchical structure and transformations from deep structures to surface structures.   Questions, for example, are categorized into different types (who, what, where, etc.), but they aren't analyzed in terms of the grammar that underlies all questions: do-insertion and inversion of auxiliary verbs (as in Does he want a sandwich?). This is what makes question-formation particularly grammatically challenging, and where many children with autism go astray.

GrammarTrainer captures the grammar of question formation incrementally, first teaching auxiliary verbs and do-insertion in negative declarative sentences ("he doesn't want a sandwhich"), and later teaching do-insertion and auxiliary inversion in questions.  This incremental strategy is totally compatible with ABA's overall pedagogy. The fact that question formation isn't currently taught this way within ABA is most likely a reflection of the fact that few, if any, modern grammarian linguists have been involved in developing ABA's language curriculum. Perhaps times are changing, and I am happy to do my part to contribute to that change.

No comments: