Quality of phonological representations: A window into the lexicon?

Abstract

Background: There is a great deal of evidence to support the robust relationship between phonological awareness and literacy development. Researchers are beginning to understand the relationship between the accuracy and distinctiveness of stored phonological representations and performance on phonological awareness tasks. However, many of the tasks currently used to assess the integrity of underlying representations are confounded by requiring spoken output. Aims: This paper describes the development of the Quality of Phonological Representations (QPR) task, a task that does not require speech output, and its evaluation in the context of a larger study examining predictors of literacy outcomes in Western Australia. Methods & Procedures: The QPR task was given as part of a larger task battery to a cohort of 235 mainstream children in the last term of their Preprimary year (average age = 5;5) and to 179 children at follow-up at the end of Year 2 (average age = 7;9).Outcomes & Results: Normative data for both accuracy and reaction time are presented in percentile tables (appendix B). In their Preprimary year, children were able to identify correct productions of multi-syllabic words (hits) on average 87.5% of the time, rising to an average of 93.8% in Year 2. As expected, children became quicker at making these judgements, reaction time shifting from an average of 1.1 s in Preprimary to 0.83 s in Year 2. A similar pattern was observed with the data for correct rejections. To make these judgements, the children had to identify a pseudo-word as an incorrect pronunciation by ‘Katie the computer’. In the Preprimary year, children were able to reject correctly the pseudo-words on average 68.5% of the time, rising to an average of 81.7% in Year 2. As expected, children became quicker at making these judgements, reaction time shortening from an average of 1.4 s in Preprimary to 0.81 s in Year 2. The QPR task was shown to have moderate reliability and concurrent validity. Conclusions: The QPR task appears to be a useful and cost-effective addition to task batteries aiming to identify at-risk children in the early stages of schooling. The ability to profile children's phonological awareness skills and gain insight into their underlying phonological representation skills allows more informed goal setting and intervention planning.