a Department of Cognitive Science , Johns Hopkins University , Baltimore , MD , USA.
Cogn Neuropsychol. 2013;30(6):396-428. doi: 10.1080/02643294.2014.880675.
Visual word recognition requires information about the positions as well as the identities of the letters in a word. This study addresses representation of letter position at prelexical levels of the word recognition process. We present evidence from an acquired dyslexic patient, L.H.D., who perseverates letters in single-word reading tasks: Far more often than expected by chance, L.H.D.'s reading responses include letters from preceding responses (e.g., SAILOR read as SAILOG immediately after FLAG was read correctly). Analyses carried out over two large data sets compared the positions of perseverated letters (e.g., the G in SAILOG) with the positions of the corresponding "source" letters (e.g., the G in FLAG). The analyses assessed the extent to which the perseverations preserved source position as defined by various theories of letter position representation. The results provided strong evidence for graded both-edges position representations, in which the position of each letter is encoded coarsely relative to both the beginning and the end of the word. Alternative position representation schemes, including letter-context and orthosyllabic schemes, were not supported.
视觉单词识别需要有关单词中字母位置和身份的信息。本研究探讨了单词识别过程中预词汇水平上的字母位置表示。我们提供了一位获得性阅读障碍患者 L.H.D. 的证据,他在单个单词阅读任务中坚持字母:L.H.D. 的阅读反应中出现前一个反应的字母的情况比预期的随机情况要多得多(例如,在正确阅读 FLAG 后,立即将 SAILOR 读为 SAILOG)。在两个大型数据集上进行的分析比较了坚持的字母(例如 SAILOG 中的 G)和相应的“源”字母(例如 FLAG 中的 G)的位置。这些分析评估了坚持程度是否保留了源位置,这是由字母位置表示的各种理论定义的。结果为分级双边边缘位置表示提供了有力的证据,其中每个字母的位置相对于单词的开头和结尾都以粗略的方式进行编码。替代位置表示方案,包括字母上下文和正字法音节方案,均不支持。