Poor language skills undermine academic success, which eventually impacts socio-economic outcomes and quality of life. When deficient language skills are first noticed in young children, there is relatively little time available to close the gap before they are faced with the increased language demands of formal education as well as the potential for academic failure. For the 8-13% of preschool children with impaired language skills, language treatments that are faster and more effective are urgently needed. Yet current treatments are notoriously protracted and expensive, and the effects of treatment can be weak. There is a growing call among scholars to step back from the business-as-usual approach to treatment research in favor of a systematic approach that integrates promising theoretical frameworks with experimental manipulations designed to isolate and enhance the effective components of treatment approaches. This grant proposes to leverage insights from the statistical learning perspective on language acquisition, which explains rapid, unguided learning sometimes even in the presence of impaired language. The grant proposes six treatment studies that target two groups of children with poor language skills. “Late Talkers” are children (ages 2-3 years) who are identified by their limited lexicons. Preschool children with specific language impairment (ages 4-5 years) show marked deficits in the use of grammatical morphemes. Parallel sets of studies with these two populations will determine the extent to which treatment variables enhance or detract from treatment efficacy across language domains. The goal of this work will be to identify specific treatment methods, derived from general learning principles, that clinicians can employ to enhance learning outcomes for children with impaired language skills.