Attention Distribution and Integration of Non-native Segments and Tones by Early Multilingual Speakers

Yi Liu, Jinghong Ning


It is well acknowledged that tone language speakers are attentive to both segmental and tonal dimensions when processing tonal speech, while less-experienced second language (L2) learners whose first language (L1) has no lexical tones are unlikely to distribute their attention to tonal information. However, it is unclear whether ignorance of the tonal dimension is also prevalent among early and highly experienced multilingual speakers. In the current study, native Cantonese speakers as well as early multilingual speakers whose L1 is Urdu (non-tonal) participated in a series of attention distribution and integration tasks in Cantonese. The multilinguals acquired Urdu first and started learning Cantonese (tonal) and English (non-tonal) simultaneously at an early age. The results showed that although the multilingual speakers could phonologically process Cantonese tones, they were unable to distribute and integrate their attention in an exact Cantonese-like way. The multilingual speakers retained their L1 attentional strategy to the processing of L2 tones in highly cognitively-demanding tasks. The results suggested that a language overlap and well-developed L1 and L2 systems both existed for multilingual speakers.


 DOI: 10.21437/SpeechProsody.2020-66

Cite as: Liu, Y., Ning, J. (2020) Attention Distribution and Integration of Non-native Segments and Tones by Early Multilingual Speakers. Proc. 10th International Conference on Speech Prosody 2020, 324-328, DOI: 10.21437/SpeechProsody.2020-66.


@inproceedings{Liu2020,
  author={Yi Liu and Jinghong Ning},
  title={{Attention Distribution and Integration of Non-native Segments and Tones by Early Multilingual Speakers}},
  year=2020,
  booktitle={Proc. 10th International Conference on Speech Prosody 2020},
  pages={324--328},
  doi={10.21437/SpeechProsody.2020-66},
  url={http://dx.doi.org/10.21437/SpeechProsody.2020-66}
}