Bilingualism and Disfluency in Children: Do Gestures Help?

• In Event: Poster Session 01
  In Poster Session: PS 01 Topic - Language, Communication

Abstract

When people speak, they are likely to pause, repeat words, or use fillers. These temporary disruptions in speech are called disfluency (Zebrowski & Kelly, 2002). It has been suggested that gestures facilitate lexical access and help speech production (Morsella & Krauss, 2004). Moreover, using gestures complements toddlers’ words, and thus, enhances early language development (Özçalışkan & Goldin-Meadow, 2005). Previous research on bilingual children’s language development suggests that, compared to monolinguals, bilinguals may have delays with some linguistic skills such as vocabulary knowledge (Bialystok et al., 2010). Disfluent speech might reflect children’s difficulties in speech production and give insight on typical language development (Byrd et al., 2015). In this study, we investigated (1) disfluency and gesture rates of bilingual and monolingual children, and (2) whether using gestures was associated with the fluency of speech during narrative production.
We tested 5- and 7-year-old Turkish monolingual (N=61, Mage=80.52, SD=11.10) and Turkish-English bilingual children (N=43, Mage=80.43, SD=12.36). We used Turkish expressive and receptive vocabulary test (TIFALDI) (Kazak-Berument & Güven, 2013) to measure children’s language competence. The Frog Story (Mayer, 1969) was used to obtain Turkish narrative retellings from the participants. We used ELAN software (Version 5.4) to transcribe speech and code disfluencies (filled pauses, repetitions, and repairs) and gestures (iconic, deictic, beat, and emblem gestures). We calculated gesture frequency and disfluency rate by dividing the total number of gestures and the total number of disfluencies, respectively, by the total word count. We also coded simple and complex clauses to obtain children’s linguistic complexity. These scores were calculated by dividing the total number of complex clauses by the total number of clauses.
We conducted a hierarchical linear regression analysis to predict the total disfluency rate in children’s narratives. The predictor variables were language group (monolingual or bilingual), age group (5 or 7 years), overall gesture frequency, TIFALDI score, and linguistic complexity score. The regression equation was significant, F(5,95)=5.34, p<.001, with an R2 of .219. Only language group (β=.189, p=.047) and age group (β=.532, p<.001) significantly predicted disfluency rates. Gesture frequency, TIFALDI score, or linguistic complexity score did not significantly contribute to the model (βs<|.089|, ps>.05). Overall, bilinguals (M=.10, SD=.06) were more disfluent than monolinguals (M=.08, SD=.05), and 7-year-old children (M=.11, SD=.05) were more disfluent than 5-year-olds (M=.06, SD=.04). Moreover, an independent samples t-test showed that, regardless of the language group, 7-year-olds’ speech (M=.51, SD=.15) was significantly more complex than 5-year-old’s speech (M=.41, SD=.22), t(109)=-2.76, p=.07.
These findings suggest that bilingual children’s higher disfluency rates might signal difficulties in speech production, particularly in finding the right words. On the other hand, 7-year-old children’s higher disfluency rates might be associated with starting primary school and, thus, trying to express more ideas and concepts. Although gestures are suggested to support language development, the way they enhance and complement speech might change as children age and reach a certain competency level. In a more developed language system, gestures might facilitate children’s speech in a more complex way than complementing words or resolving disfluencies.

Authors

• Burcu Arslan, Koç Üniversitesi
  Presenting Author

• Asli Aktan-Erciyes
  Non-Presenting Author

• Can Avcı, Koç University
  Non-Presenting Author

• Selin Tezel, Koç University
  Non-Presenting Author

• Tilbe Göksun, Koç University
  Non-Presenting Author