Functional processing units in writing Chinese : a develomental study

Abstract

Background Studies that investigate the writing process mostly involve observing the writing errors produced by dysgraphic patients suffered from strokes who have premorbid normal writing abilities. Through analyses of substitution, addition, deletion and transposition errors, previous studies suggested that Chinese people use strokes, logographemes and radicals as processing units in Chinese character writing (e.g. Law & Leung, 2000; Han, Zhang, Shu & Bi, 2007). What remains unclear is how children develop the use of these processing units with different grain sizes in writing because they are not formally introduced in education. Replicating the experiments on children is difficult since the errors produced by children will reduce as they grow up, which allows less items for analyses. Using Android tablets, the current study observed the processing units used by Chinese children at different age in copying. Method Participants. Altogether 62 children, 31 grade 1 (mean age= 7;08) and 31 grade 5 (mean age= 11;2) were recruited from an ordinary mainstream primary school in Hong Kong. All children were native Cantonese speakers and received education in Hong Kong since kindergarten. They were all assessed to have normal non-verbal intelligence, visual memory, visual spatial perceptual, visual motor coordination and Chinese reading abilities. Stimuli. A total of 36 pseudo-characters, each containing two radicals with two constituent logographemes, were constructed. All radicals are of high frequencies based on the Hong Kong Corpus of Primary School Characters (Leung & Lee, 2002). Task. Each participant was tested individually in a quiet room. In each trial, each participant was instructed to copy a target pseudo-character using a stylus pen on a 7-inch tablet (Quad-core with 2.20 GHz processors, resolution of 1820x1200, refresh rate of 60 Hz) running Android 4.1.1. The elapsed time and the coordinates each time the stylus pen left / touched the tablet screen were recorded. The durations and distance travelled by the stylus between successive strokes located at the logographeme boundaries, radical boundaries and non-boundaries are calculated accordingly. Results Results revealed a significant grade effect that grade 5 children spend less time between each strokes compared to grade 1 children (p<.05). Radical effect was significant such that durations located at radical boundaries are longer than durations not located at the boundaries after controlled for distance traveled (p<.001). Significant interaction effect was also observed that in grade 5, the durations located at logographeme boundaries are longer than durations not located at the boundaries after controlled for distance traveled (p<.001), while such difference is not significant in grade 1. Discussion Results showed that grade 5 children are more competent overall in copying. The grade difference is observed not only quantitatively but also qualitatively. We suggest that any significant radical / logographeme effect is the result of the retrieval and planning of the successive functional writing units in the writing process. The significant logographeme effect observed exceptionally in grade 5 implies that grade 5 children use logographemes as functional writing units while grade 1 children do not. Our results reflect a sequential pattern in the development of orthographic knowledge. Future studies using other writing tasks such as delayed-copying and writing-to-dictation, are recommended to confirm this developmental pattern.