Reference values of nonword repetition test for Brazilian Portuguese-speaking children

ABSTRACT Evaluation of the phonological working memory (PWM) through repetition of nonwords can provide important information on the linguistic abilities of children, thus differentiating those with and without communication disorders. Objective: The aim of this study was to obtain reference values in the Nonword Repetition Test (NWRT) in order to investigate the performance of children without language disorders concerning this type of memory. Material and Methods: The study was conducted on 480 normal children of both genders aged 4 years to 8 years and 11 months, attending preschool and elementary school. The NWRT consisted of repeating 20 (children up to 4 years) or 40 (for children aged 5 years or more) invented words with 2 to 5 syllables. The results were subjected to descriptive statistical analysis. Comparison between ages and between the number of syllables in nonwords was performed by the Tukey's multiple-comparison test and one-way analysis of variance, at a significance value of p<0.05. Results: There was statistically significant difference (p<0.05) in performance between children of different age groups, except between 7- and 8-year-olds. The analysis also showed statistically significant difference (p<0.05) in the number of syllables between the different age groups. Conclusions: The reference values obtained indicated an improvement in performance with the increase of age of children, indicating an improvement in the storage of verbal material in the PWM. The performance was worsened with the increase in the number of syllables in words, demonstrating that the greater the number of syllables, the greater is the difficulty in storing verbal material.


INTRODUCTION
The Psycholinguistic Model (PLM) has decisively influenced the way to assess and treat language disorders in the last decade 8 . This model has been proven efficient because it explains how human beings process information coming to their senses, access the words stored in their lexicon and use the mental representations that encode information, thus understanding the nature of language disorders 6,9,19,23 . This model considers all processes involved in the act of communicating, from the primary level, involving the input and output of verbal information, up to the third, which corresponds to the level of cognitive operations of more complex language 17 . These processes include the working memory, which plays a significant role in the maintenance of thinking and learning, verbal comprehension and lexicon access 11,14,16 . It is a system for processing and storing information on a short-term basis, organized into four components, namely the central executive, two work subsystems -the phonological and visuospatial loop -, and the episodic buffer 2,5,8 . The phonological loop stores and manipulates material based on speech www.scielo.br/jaos and has two components: the phonological storage, which receives information through direct (auditory presentation) and indirect ways (visual presentation); and the reverberation process or subvocal test, which occurs serially in real time and acts to restrain the natural decay of phonological storage. One of the primary functions of the phonological loop or phonological working memory (PWM) is to store unfamiliar sound patterns, until a record of more permanent memory becomes consistent 1,4 .
The PWM has a fundamental role in acquiring language skills in children 8,13 and its deficit has been suggested as the origin of linguistic difficulties in children with specific language impairment 1,15,16,18,24,25 .
In the clinical context, the PWM is evaluated by two procedures: digit span (repeating sequences of numbers in direct and inverse order) and repetition of words or nonwords (NW). The repetition of NW is indicated as a more reliable test for the PWM, because the verbal material input is unknown and hence not subject to lexical influences 3,4,10,21 .
Thus, considering the lack of instruments based on the Portuguese language for assessment of the PWM, the objective of this study was to obtain reference values for the Nonword Repetition Test

MATERIAL AND METhODS
The study was conducted on 537 children  Intonation -all words are paroxytones. An adequate repetition is considered when emitted in an identical manner as the examiner. However, it may be considered correct in case of replacement of the vowel "e" by "i" in the end of words, or also closed vowels "e, o" by open vowels "é, ó". If a phonological disorder is observed, the processes should be recorded. In these cases, the replacement or omission of a phoneme during the repetition will not be considered a mistake. Instructions: "I will say some words that do not exist. You should pay attention because you will repeat as I said. I will say it once and you will repeat it. It may be slightly strange, but it won't take long. Attention, let's go!" For children aged 3

RESULTS
The results showed statistically significant