Quasi-Universal Non-word Repetition Task: Extension to a new population

Abstract

The current study investigated the potential use of the Non-word Repetition Task (NWRT) as a diagnostic tool of language impairment in a new population: newly arrived migrant pupils (NAMS). NAMS form a highly varying group when it concerns their cultural, migrational and language backgrounds. It is therefore important that (language) evaluation tools applied in this population are cross-linguistically and cross-culturally deployable, and subsequently minimally biased. The NWRT is considered unbiased as it uses non-words. The task is often used in language impairment (LI) research and diagnostics as it mimics novel word-learning processes thought attenuated in LI patients. However, typically developing bilinguals’ NWRT performance can resemble that of monolingual LI patients (Boerma et al., 2015) because the phonotactic and (supra)segmental characteristics of the items are target language specific. To avoid misdiagnosis in bilinguals, Chait (2015) has developed a quasi-universal NWRT with items as linguistically universal as possible. In their study on the Dutch specific and the quasi-universal NWRT, Boerma and colleagues (2015) found that the quasi-universal NWRT was indeed the better predictor of LI in bilinguals. To extend their research, this explorative study administered the same quasi-universal and Dutch specific NWRTs to a group of typically developing NAMS living in the Netherlands, and compared their performance to the typically developing bilinguals’ in Boerma’s study. The purpose of this study was two-fold: 1) to (further) examine the universalness of the quasi-universal NWRT; 2) to investigate whether the task can be used as a tool to evaluate LI in NAMS. The results showed that NAMS indeed performed equally well on the quasi-universal task as the typically developing bilinguals. More surprisingly, NAMS also performed similarly to the bilinguals on the language specific NWRT. The discussion of these results showed that there are several explanations for this (partly) unexpected outcome, and that it is likely that the NAMS’ quasi-universal performance can be considered reliable. This study therefore tentatively concluded that the quasi-universal NWRT is in fact universal enough to have a future as a diagnostic tool of language impairment in NAMS.