James W Montgomery1, Jennifer Windsor. 1. School of Hearing, Speech & Language Sciences, Grover Center W231, Ohio University, Athens, OH 45701-2979, USA. E-mail: montgoj1@ohio.edu
Abstract
PURPOSE: This study investigated the effects of processing speed and phonological short-term memory (PSTM) on children's language performance. METHOD: Forty-eight school-age children with specific language impairment (SLI) and age peers completed auditory detection reaction time (RT) and nonword repetition tasks, the Clinical Evaluation of Language Fundamentals-Revised (CELF-R; E. Semel, E. Wiig, & W. Secord, 1987), and a word recognition RT task. Correlation and regression were used to determine unique and shared contributions to variance among measures. RESULTS: Children with SLI were outperformed by age peers on each task. Auditory detection RT was correlated with nonword repetition (NWR) in each group. However, both variables covaried with age, and auditory detection RT did not contribute unique variance to NWR in either group. For the SLI group, NWR predicted unique variance in CELF-R performance (about 15%); auditory detection RT predicted a smaller amount of unique variance in the word recognition RT task (about 9%). CONCLUSION: Processing speed and PSTM measures covaried with chronological age. Processing speed was associated with offline language performance only through association with PSTM. Processing speed contributed to online language performance, suggesting that speed is associated with processing more familiar language material (i.e., lexical content and structure) than less familiar material (e.g., various content on the CELF-R).
PURPOSE: This study investigated the effects of processing speed and phonological short-term memory (PSTM) on children's language performance. METHOD: Forty-eight school-age children with specific language impairment (SLI) and age peers completed auditory detection reaction time (RT) and nonword repetition tasks, the Clinical Evaluation of Language Fundamentals-Revised (CELF-R; E. Semel, E. Wiig, & W. Secord, 1987), and a word recognition RT task. Correlation and regression were used to determine unique and shared contributions to variance among measures. RESULTS:Children with SLI were outperformed by age peers on each task. Auditory detection RT was correlated with nonword repetition (NWR) in each group. However, both variables covaried with age, and auditory detection RT did not contribute unique variance to NWR in either group. For the SLI group, NWR predicted unique variance in CELF-R performance (about 15%); auditory detection RT predicted a smaller amount of unique variance in the word recognition RT task (about 9%). CONCLUSION: Processing speed and PSTM measures covaried with chronological age. Processing speed was associated with offline language performance only through association with PSTM. Processing speed contributed to online language performance, suggesting that speed is associated with processing more familiar language material (i.e., lexical content and structure) than less familiar material (e.g., various content on the CELF-R).