Abstract: Spelling difficulty (SD) is a kind of learning disorders. Children with SD have deficits in graphic memory. Because of the complexity in graphic representation, Chinese characters are difficult to be memorized only by visual encoding. Other encoding ways are needed to improve graphic memory. The aim of the present study is to investigate the influence of motor encoding and visual encoding to the graphic memory of children with SD.
In the first experiment, effects of visual encoding and visual-motor encoding on graphic memory of children with SD and children without spelling difficulties (CNSD) were compared. First, participants were asked to learn a series of simple psudo-characters either in visual condition (only seeing the characters) or in visual-motor condition (tracing the character with finger while seeing the character). After a mental arithmetic task, their performances were tested by using a free recalling task.
As a follow-up experiment, in experiment 2 we are going to investigate whether stroke phonetic encoding could help improve the effect of visual-motor encoding on graphic memory of children with SD. The procedure was the same as in Experiment 1, except that visual condition was changed into visual-motor-phoneme condition. In the new condition, participants were asked to learn the psudo-character together through seeing the character, tracing with finger and speaking out the pronunciation of every stroke.
In experiment 1, 1) main effect of encoding mode and group were significant. Psudo-character memory performance of all participants under visual-motor condition was higher than under visual condition; and global performance of children with SD was poorer than children without SD. 2) Interaction between encoding mode and group was prominent. Visual-motor encoding can significantly improve recalling performance of children with SD, but have no significant effect for children without SD.
In experiment 2, 1) main effect of group was not significant, and that of encoding mode was only marginal significant. The correct rate of graphic recalling under visual-motor-phoneme condition was slightly higher than under visual-motor condition. 2) Interaction between encoding mode and group was significant. Pronouncing the strokes didn’t significantly facilitate graphic recalling performance of children with SD, but prominently improved the performance of children without SD.
In experiment 1, visual-motor encoding could improve the graphic recalling performance of children with SD. Motor tracing might induce children with SD to elaborately encode each character’s grapheme. Children without SD have intact visual encoding abilities; they could directly encode the subtle properties of grapheme, and form graphic representation of high quality. Hence, motor tracing had no significant facilitating effect for them.
In experiment 2, phonological encoding of strokes can improve the graphic recalling performance of children without SD, but have no significant effect on that of children with SD. There might be three possible reasons. First, visual-motor-phoneme encoding task required participants to simultaneously encode the graphic and phonetic properties of strokes. However, the poor graphic-phonetic connections of children with SD blocked the facilitating effect of phonetic process on graphic process. Second, when participants were enforced to encode the graphic and phonetic characteristics simultaneously, children with SD would tend to ignore and inhibit the phonetic characteristics. This process needs mental resource, and may in turn interrupt graphic encoding process. Lastly, according to the phonological deficit hypothesis of spelling difficulties, children with SD may have deficit in phonological processing.

Key words: spelling difficulties, phonological encoding, motor encoding