|
|
Experimental studies of infant learning show that babies can learn patterns that reflect both word units and grammatical patterns with only minutes of exposure to novel language input. Moreover, this input-driven learning generalizes to untrained examples. In contrast, treatment studies for specific language impairment can go on for months, and overall gains can still be modest. This raises the critical question: Could we produce better treatment outcomes by making treatment protocols adhere to the conditions known to facilitate rapid learning by infants? A series of experimental studies demonstrated that both children and adults with impaired language skills can demonstrate rapid learning (i.e., generalization that occurs within a single session) when input is structured in ways that are known to facilitate input-based learning. Most recently, we have begun to translate this experimental work to treatment. In a series of feasibility and early efficacy studies, we show that application of learning principles can boost treatment effects for a well-studied clinical procedure (conversational recast treatment). This work was supported by NIDCD grants R01DC004726-S1, R21DC014203, and R01DC011276. |
|