关键词: 字母位置编码模型, 词汇识别, 转置效应, 模型对比

Abstract: Letter position information plays a crucial role in visual word recognition. Over the past decades, research on letter position encoding has significantly advanced various theoretical frameworks to explain experimental findings and their underlying cognitive mechanisms. This article systematically introduces six major computational models of letter position encoding, including the Overlap Model, the Open-bigram Model, the SERIOL Model, the Spatial Coding Model, the Bayesian Reader, and the Positional Ordering of N-Grams (PONG) Model. These models encompass various cognitive processing mechanisms, ranging from overlap encoding to sequence and spatial encoding, and represent important theoretical frameworks in the field of letter position encoding. We compare these models in terms of their structures, theoretical foundations, lexical recognition logic, cross-linguistic adaptability explanations, and explanations of some common effects. Additionally, this article highlights the limitations of these models by summarizing the impacts they failed to explain. Based on these comparisons, future model development could benefit from integrating more empirical research findings and diverse types of experimental data to enhance explanatory power. Furthermore, considering cross-linguistic factors and the findings from second language studies, further explorations of cross-linguistic consistency will be a valuable direction for future research.

Key words: computational models of letter position encoding, word recognition, transposition effect, model comparison