Nonword repetition in bilinguals

We completed this study a couple of years ago but it always takes some time to get things written up and then submitted and so on. Sometimes it’s a bit frustrating that the process takes so long. There are lots of papers I see at conferences that I might wait a couple of years before I see the full paper in print. It’s the same with our stuff of course, sometimes what we present at a conference is only one analysis or is with fewer participants than the final paper. The revision process usually helps to focus and strengthen the paper, but that also takes time.

Anyway, this is a long introduction to writing about a paper by Summers, Bohman, Gillam, Peña, & Bedore titled “Bilingual performance on nonword repetition in Spanish and English” in the International Journal of Language and Communication Disorders. In this paper we tested bilingual children on nonword repetition tasks in both their languages. We also tested some kids on the monolingual end of the continuum on both sets of words. The words we used are those developed by Dollaghan & Campbell (for English) and Janet Calderon (for Spanish).

There were a couple of interesting findings. One is that the children generally were more accurate on Spanish nonwords than English nonwords. Why would this be? Well, the phonotactics of Spanish and English are different. Spanish has fewer vowels and consonants, and many words have CVCV patterns. I think this makes it easier to produce longer words. English on the other hand has more consonants, more final consonants in words, more blends, and more vowels. This makes it complex at the level of the syllable– consequently more English words are shorter in comparison to the length of Spanish words. This brings us to another finding, children were more accurate with the longest Spanish nonwords than the longest English nonwords. Here our findings are consistent with Kohnert and colleagues’ suggestions that nonwords are linguistically influenced. The nonwords we used are the more nonwordlike words rather than the wordlike words. Probably a comparison of both types in bilinguals would tell us more about how children use their knowledge of their language(s) to do this task.

Performance on nonwords  was related to language experience. We also found relationships between nonword repetition and morphosyntax. The relationship with morphosyntax is interesting because this relationship is not typically studied when looking at nonword repetition performance in children. It could be that children’s ability to manipulate morphemes as indexed by our morphosyntax task is related to remembering syllables. But, this needs to be explored further. For now however, this study allows us to begin to understand the differences in Spanish and English cross-linguistically and to understand how children perform on highly similar tasks in two languages.


, , , ,

  1. #1 by Dr Alison Crutchley on October 16, 2009 - 9:35 am

    Interesting! So nonword rep is not the ‘language-independent’ process that some have suggested. Do you think it would be possible to arrive at a set of nonwords that are equally difficult for everyone?

  2. #2 by Elizabeth D. Peña on October 16, 2009 - 11:59 am

    An interesting question. Kathy Kohnert has certainly suggested that nonwords are not as language independent as one would like, see Kohnert, Windsor, and Yim (2006) I think that nonwordness (is that a word??) might be relative to the language the children know rather than completely independent. Perhaps it could be possible to construct some words that are balanced (as a set) with respect to how related they are to language A vs. language B.

  3. #3 by Kathryn Kohnert on October 17, 2009 - 6:39 am

    As Liz notes, we have found that while using nonwords can certainly reduce the role of previous experience, they do not eliminate it. That is, sets of nonwords are typically constructed using the phonotactic constraints of a particular language and children who have more experience (&/or greater ability in that language) seem to do better. This was an issue we took up empirically in an article titled “Do language-based processing tasks separate children with primary language impairment from typical bilinguals?” published in 2006 in Journal of Learning Disabilities Research & Practice. This doesn’t mean the task is not helpful in linguistically diverse learners with and without LI, just that performance cannot be completely separately from previous experience. For our 2006 sample, we found that NWR performance could be informative for ‘ruling out’ (vs. ruling in) LI.

  4. #4 by Kathryn Kohnert on October 17, 2009 - 6:52 am

    Oops . . . I inadvertently hit the submit…. so consider this a continuation of comment 3:

    More recently we have used Spanish, English and Hmong nonword repetition tasks, administering all three to 6-10 year old bilingual Spanish-English and Hmong-English children as well as to English-only speaking children. Imagine how challenging it may be for English-only speaking children (and adults!) to repeat combinations of sounds overlaid with one of several different tones, designed to be consistent with the Hmong language? Anyway, results from the first part of this study are currently under review — comparing >200 children monolingual English and bilingual Spanish-English speakers, with and with LI. Results clearly indicate that both integrity/efficiency of the underlying language processing system as well as previous language experience affect Nonword repetition performance. Within each bilingual and monolingual group, there are strong positive associations between children’s performance in English and Spanish (so we could predict English-only speakers ability to repeat words in Spanish based on their English nonword repetition ability and the same was true for bilingual children). At the same time, bilingual children outperformed monolingual children in Spanish and English-only children were better at repeating English nonwords than the bilingual children. So, dual-influence of both experience in a lanugage and the efficiency of the underlying language processing system. These results sound very consistent with those talked about by Liz in the original post. Hooray for converging evidence!!.

  5. #5 by Ms. Chavez on January 2, 2016 - 5:09 pm

    Where can I find a NWR task in Spanish to use? Please advise at

    Thank you!

  6. #6 by Elizabeth D. Peña on January 27, 2016 - 10:54 pm

    check out this resource– there’s a chapter on NWR

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: