Reading aloud pseudo-characters by individuals with acquired dyslexia: Evidence for lexically mediated processes in reading Chinese

Abstract

This study examined a hypothesis discussed in Bi, Han, Weekes, and Shu (2007) of a non-lexical system consisting of correspondence rules between character subcomponents and phonology in reading Chinese. It contrasts with the lexical view stating that reading aloud must be lexically mediated as assumed in all current models of character naming. To contrast these accounts, we asked two Cantonese brain-injured anomic individuals with largely preserved reading abilities to read aloud non-existing phonetic compounds containing a free-standing or a non-free-standing phonetic radical. The lexical view predicts that both subjects should be able to provide plausible responses to pseudo-characters, whereas such a prediction is not made by the non-lexical account. We found that both participants could produce legitimate responses to pseudo-characters, similar to the normal individuals. Furthermore, their responses to complex pseudo-characters were strongly influenced by the presence of free-standing phonetic radicals, lending further support to the lexical view.

Keywords:

• Acquired dyslexia
• Reading Chinese
• Pseudo-character
• Lexically mediated processes

Acknowledgements

The work reported here was supported by a grant from Small Project Funding at the University of Hong Kong. We are grateful to YKM, YYW, and their normal controls for participation in this study.

Notes

¹This is true for Cantonese. In Mandarin, the character is a retroflex produced in co-articulation with the preceding syllable.

²In this paper, phonetic transcriptions of Chinese characters are given in jyutping, a romanisation system developed by the Linguistics Society of Hong Kong. The number in the transcription represents the tone.

³Lee, Tsai, Chiu, Tzeng, and Hung (2006) reported an event-related potentials (ERPs) study in which adult participants were simultaneously presented, on each trial, with a pseudo-character containing a phonetic radical from consistent phonetic compounds and a syllable that might or might not be compatible with the pronunciation associated with the phonetic compound family. Different ERPs were elicited in the two conditions. The data were interpreted as indicative of the presence of phonological information generated by the pseudo-character stimuli.

⁴The frequency count of all characters is based on Ho (1992).

⁵A phonetic radical neighbourhood must consist of at least four phonetic compounds listed in Ho (1992).

⁶As one of the anonymous reviewers pointed out, reading pseudo-characters with consistent phonetic radicals does not necessarily differentiate the two hypotheses. Their inclusion may nevertheless reflect the use of some strategy on the part the dyslexic participants, i.e., reading aloud the phonetic radicals. This was discussed toward the end of General Discussion.

⁷Although the division was made largely arbitrarily, we took into consideration the relative numbers of items in the two sets and the numbers of phonetic radicals being the dominant pronunciations in each set.