Word-reading development, the double-deficit hypothesis, and the diagnosis of dyslexia

Abstract

In this study, it was hypothesised that with age, word-readingspeedgets better explained by reading-related word-holistic or lexicalvariables than by word-analyticorsub lexicalvariables. At an operational level these conceptual groups of variables were measured by continuous-naming tasks, and tasks which require phonemic manipulations, respectively.The tasks were administered to randomly sampled readers at cross-sectional age levels of 8, 10, 12 and 14 years of age. After data reduction, and at each age level, the word-reading variable was regressed on phonemic performance and alphanumeric naming (numbers and letters). As predicted, with age, alphanumeric naming speed increasingly contributed unique variance to reading speed, whereas unique phonemic contributions declined.The second part of the study applied Wolf and Bowers’ (1999) double deficit hypothesis. For each of the variables word-reading, alphanumeric naming, and phonemic performance, and at each age level, four normative proficiency levels (level 1 = poor; 2 = below average; 3 = above average; 4 = high) were distinguished, and combinations of naming and phonemic performance levels were defined as double, single, or no-deficit categories. At various age and reading proficiency levels, reading scores were compared as a function of deficit categories. A double cognitive deficit (DD) was highly predictive for (very) poor reading performance. It is proposed to consider DD readers as ‘classic dyslexics’. Discussion about the remaining profiles, theoretical and practical issues, and implications for remedial support and policy is provided.