Schneiderman, E.,Kotsopoulos, Angélique.2009-03-192009-03-1919981998Source: Dissertation Abstracts International, Volume: 59-07, Section: A, page: 2444.9780612283527http://hdl.handle.net/10393/4081http://dx.doi.org/10.20381/ruor-13567Objectives. There are three major goals to the present study: (1) to study the nature of errors in recalling sentences; (2) to investigate how specific errors in recalling sentences relate to expressive grammatical deficits and reading impairment; and (3) to make suggestions for the design of a pilot screening test that could predict processing and reading deficits. Method. Subjects were 45 children of normal intelligence, ages 8 to 13 years. The children presented with psychiatric disorders, and had been admitted to a day treatment and school program. The most prevalent diagnosis was attention deficit with hyperactivity and aggressive behaviour. A smaller proportion had adjustment and anxiety disorder. Excluded from the sample were severe cases of autism, schizophrenia and physical handicaps. Most of the children exhibited academic impairment and language deficits. Assessments. The recalling sentences, formulated sentences, sentence assembly, and sentence structure subtests of the Clinical Evaluation of Language Fundamentals (CELF-R) were administered, as were the reading decoding and reading comprehension subtests of the Kaufman Test of Educational Achievement (K-TEA). To investigate relationships between errors in recalling sentences and grammatical deficits and reading impairment, the following statistical analyses were used: correlation, multiple regression, chi-square and kappa analyses. Results. In recalling sentences, errors involving content words occurred more frequently than errors involving grammatical words. Words with a large number of associatives, unfamiliar words, and less frequently occurring words were liable to omissions and substitutions. Content omissions affected mostly syntactically unsupported elements (e.g. adjuncts). Most content and grammatical omissions were words that do not assign thematic roles or have thematic roles assigned to them. Content substitutions affected mostly arguments (e.g., subjects or complements). Most of the substituted words were synonyms of the target word. Deficits in processing and timing, in particular, were identified as possible contributors to content omissions and substitutions. Grammatical errors occurred in syntactically complex sentences (mostly relative clauses) and affected functive words and expanded auxiliaries. Grammatical omissions were more often associated with processing errors, while grammatical substitutions were associated with syntactic deficits. Two types of errors were associated with reading deficits: content omissions and grammatical substitutions. They accounted for 44% of variance in reading comprehension and 31% of variance in reading decoding. Conclusion. The findings of the present study, together with an extensive literature review, indicated that sentence recall is a complex task involving both processing and verbal memory, and that it can be used to predict processing deficits and reading disability. Recommendations are made as to the type of sentences to include in a pilot screening test that could predict processing and reading deficits.706 p.Education, Educational Psychology.Thesis