One of the most replicated effects in the contemporary word recognition literature is the transposed-letter effect (TL-effect): pseudowords created by the transposition of two letters (e.g., MOHTER) are often misread as the real word. This effect ruled out those accounts that assume that letter position is encoded accurately and led to more flexible coding schemes. Here, we examined whether reading skill modulates this effect. The relationship between reading skill and the TL-effect magnitude is a contentious issue both empirically and theoretically. The present lexical decision experiment was designed to shed some light on the relationship between reading skill and the TL-effect magnitude with a large sample of Grade 6 children. To that end, we conducted both multiple regression and path analyses. Results showed that a specific aspect of reading skills (pseudoword reading) negatively correlates with the TL-effect's magnitude in the error data (i.e., MOHTER is less wordlike for better readers). This finding highlights the need for a comprehensive visual word recognition model that includes individual variability and the multidimensional character of reading in school-age children.

当代单词识别文献中复制最多的效应之一是字母转置效应（TL 效应）：由两个字母（例如 MOHTER）的换位产生的伪词经常被误读为真实单词。这种效应排除了那些假设字母位置被准确编码并导致更灵活的编码方案的帐户。在这里，我们检查了阅读技能是否会调节这种效果。阅读技能和 TL 效应量级之间的关系在经验和理论上都是一个有争议的问题。本词汇决策实验旨在通过 6 年级儿童的大样本来阐明阅读技能和 TL 效应量级之间的关系。为此，我们进行了多元回归和路径分析。结果表明，阅读技巧的一个特定方面（伪词阅读）与错误数据中 TL 效应的大小呈负相关（即，对于更好的读者，MOHTER 不太像单词）。这一发现凸显了对综合视觉单词识别模型的需求，该模型包括学龄儿童的个体差异和阅读的多维特征。