Children with Specific Language Impairment (SLI) are slower to learn new words than their peers, placing them at risk for academic failure. In this study, we are improving a storybook reading treatment to help Kindergarten children with SLI learn new words. In this study, we compare three versions of book reading that vary in how often children are tested on, meaning asked to talk about, the words they are learning in the book: low vs. mid vs. high testing. We then examine which version of the treatment leads to better learning of the words during treatment and remembering of the words after treatment. We also seek to understand individual differences in treatment outcomes by examining pre-treatment predictors as well as progress during and after treatment.

Children with Specific Language Impairment (SLI) are slower to learn new words than their peers, placing them at risk for academic failure. Our long-term goal is to develop an effective word learning treatment for kindergarten children with SLI, thereby improving their academic and vocational outcomes. During the prior funding period, we successfully taught new words to children with SLI via interactive book reading, a treatment involving an adult reading a storybook to a child and deviating from the text to teach new words. We identified the adequate intensity of the treatment and showed that children with SLI learn an appropriate number of words by the end of 8-weeks of treatment, approximating the number of words learned by typically developing children in prior studies. However, this successful support of short-term word learning revealed new challenges that must be overcome in this renewal to continue to understand and improve long-term word learning by children with SLI. Thus, a second preliminary clinical trial involving 60 kindergarten children with SLI is proposed. Aim 1 addresses the challenge that newly learned words were forgotten once treatment was withdrawn. We attempt to buffer forgetting by comparing different amounts of testing during interactive book reading (low vs. mid vs high testing). Incorporating testing into training is a well-established and highly replicated means of reducing forgetting by adults and typically developing children. Aim 1 will determine whether testing can be harnessed to buffer forgetting by children with SLI under real world conditions. Aims 2 and 3 address the challenge that not all children benefitted equally from interactive book reading. In Aim 2, we identify pre-treatment characteristics of children with SLI that are associated with the slope of learning during treatment or the slope of forgetting post-treatment. Moreover, we select a pre-treatment battery that samples a wide array of skills likely to be associated with learning (language processing, working memory, and episodic memory) or forgetting (overall learning during treatment, decay rate). Aim 2 will provide a foundation for predicting which children will benefit from interactive book reading and will identify which skills are major barriers to long-term word learning by children with SLI. In Aim 3, we classify each child's response at the end of treatment (learner vs. non-learner) and at the end of post-treatment monitoring (rememberer vs. forgetter). Then, we examine earlier performance to determine when treatment and post-treatment outcomes can be predicted. This yields empirically based benchmarks for progress that can be used to tailor the treatment to individual children and establishes the stability of learning and forgetting over time. Overall, this research advances a promising treatment to effectively overcome the significant word learning challenges faced by children with SLI and reveals the contribution of learning and forgetting to language normalization by children with SLI. The results will have impact beyond word learning and SLI because all treatments require boosting learning and buffering forgetting. Thus, the knowledge gained will further catalyze clinical research.

Inclusion Criteria:

Eligible for kindergarten enrollment Age 5-6 years Normal hearing Normal nonverbal IQ -- Nonverbal IQ of 85 or higher Language impairment as documented by language score \

De-identified pre-treatment test scores and pre-, during, and post-treatment outcome measures will be included as supplemental material to publications. All study information will be shared as supplemental material to publication or upon request.

Data will become available as findings are published (in conjunction with publication)

Data will be open access

All participants during the 15 treatment sessions (8 weeks) and during the retention period up to 12 weeks post treatment.

We excluded the data from 8 children from analyses whose testing schedules were disrupted due to the 2020 pandemic. These children finished the clinical trial, however, because the primary questions of this study focused on learning rates and forgetting rates, and the schedule for these children differed substantially from the other children, we elected to exclude their data as their treatment schedules would not capture learning and retention rates on the same schedule as other children.

Pre- and post-outcome assessments will be performed by an individual blind to arm assignment.