Swan D, Goswami U
Department of Experimental Psychology, University of Cambridge, England.
J Exp Child Psychol. 1997 Jul;66(1):18-41. doi: 10.1006/jecp.1997.2375.
The claim that the well-documented difficulties shown by dyslexic children in phonological awareness tasks may arise from deficits in the accuracy and the segmental organization of the phonological representations of words in their mental lexicons is receiving increasing interest from researchers. In this experiment, two versions of the phonological representations hypothesis were investigated by using a picture naming task and a battery of phonological measures at three linguistic levels (syllable, onset-rime, phoneme). The picture naming task was used to identify words for which dyslexic and control children had accurate vs inaccurate phonological representations, and performance in the phonological awareness tasks was then compared for the words which had precise vs imprecise representations. Findings indicated that frequency effects in the phonological awareness tasks at all levels disappeared for dyslexic and control children once representational quality was taken into account, and that the availability of sublexical units for analysis appeared to differ according to (1) the accuracy and retrieval of the phonological representation and (2) the linguistic level tapped by the phonological awareness task.
阅读障碍儿童在语音意识任务中表现出的有充分记录的困难,可能源于其心理词典中单词的语音表征在准确性和音段组织方面的缺陷,这一观点正受到越来越多研究者的关注。在本实验中,通过使用图片命名任务以及三个语言层面(音节、首音-韵脚、音素)的一系列语音测量方法,对语音表征假说的两个版本进行了研究。图片命名任务用于确定阅读障碍儿童和对照儿童对哪些单词具有准确或不准确的语音表征,然后对具有精确或不精确表征的单词在语音意识任务中的表现进行比较。研究结果表明,一旦考虑到表征质量,阅读障碍儿童和对照儿童在各级语音意识任务中的频率效应就会消失,并且用于分析的次词汇单元的可用性似乎因以下因素而异:(1)语音表征的准确性和检索情况;(2)语音意识任务所涉及的语言层面。