Abstract: Although orthographic neighborhood (N) size effects have been extensively studied in phonogram recognition, little research explored the processing of the orthographic neighbors in typically developing readers. Noted that, phonograms are the most common Chinese characters that children learn during the elementary school. Phonograms are comprised of a semantic radical and a phonetic radical. The phonetic and semantic radicals usually imply the character’s pronunciation and meaning to some extent, respectively. Existing evidence has indicated that semantic and phonetic radicals each play an important role in phonogram recognition. However, most of these researchers mainly focused on phonetic radical neighbors which ignoring the effect of semantic radical neighbors. Thus, this study attempted to investigate the comprehensive effect of semantic and phonetic neighbors in Chinese phonogram recognition among the developing readers. A two (grade: third vs. fifth vs. adult) × two (semantic neighbors: high vs. low) × two (phonetic neighbors: high vs. low) triple factors mixed experimental design was adopted. Four conditions were formed by crossing semantic neighbors and phonetic neighbors: H-H(phonogram comprised of a high semantic neighbor and a high phonetic neighbor), H-L(phonogram comprised of a high semantic neighbor and a low phonetic neighbor), L-H(phonogram comprised of a low semantic neighbor and a high phonetic neighbor), L-L(phonogram comprised of a low semantic neighbor and a low phonetic neighbor). The mean character frequency, stroke number and character familiarity were balanced across these four types of words. Eprime1.1 software was used for programming. Each trial began with a fixation cross in the center of the screen for 500 ms, which was then replaced by the target word presented for 150 ms. The blank appeared immediately after the target and stayed on the screen until participants responded, or for 1500 ms when no responses were made. Subsequently, the blank was replaced by a signal(~ ~)for 1500ms. The participants were required to judge whether the targets presented on the screen were real characters by pressing the corresponding keys as quickly and accurately as possible. Firstly, significant interaction between grade and semantic neighbor were found for reaction time. Further analysis revealed significant main effect of semantic neighbor was found in grade 3, the characters with high semantic neighbors (H-H, H-L) showing the faster reaction time than the characters with low semantic neighbors (L-H, L-L), but not in grade 5. Besides, significant interaction between semantic neighbors and phonetic neighbor were also found for reaction time. Further analysis showed that when the character was high phonetic neighbors (H-H, H-L), the characters with the high semantic neighbor (H-H) showing the fast reaction time than the characters with low semantic neighbor (H-L). These results suggest that, the effect of the neighborhood size of semantic radicals is modulated by the neighborhood size of phonetic radicals in Chinese phonogram character recognition among grade 3 and grade 5 children. Besides, the grade 3 readers rely on the semantic neighbors to identity a Chinese character more than the grade 5 readers. With the development of reading skills, the effect of phonetic neighbors is strengthened. The present results provided the evidence for the lexical tuning theory, in which researchers argued the transition from broadly tuned to finely tuned lexical representation in reading development.

Key words: phonogram, semantic neighbor, phonetic neighbor, children

摘要： 采用2（年级）×2（声旁家族大小）×2（形旁家族大小）三因素混合实验设计，在形声字识别中考察了三、五年级儿童的形旁和声旁家族效应。结果仅在三年级中发现了显著的形旁家族大小效应，而五年级中却没有发现。但两个年级中形旁和声旁家族的交互作用均显著。进一步分析发现在声旁处于大家族时，形旁也处于大家族时，三、五年级儿童的反应时更快。此结果说明声旁和形旁家族共同影响形声字识别，但在三年级儿童中，形旁家族效应可以不依赖于声旁家族单独发挥作用，五年级儿童中却没有独立的形旁家族效应。

关键词: 儿童 声旁家族 形旁家族 形声字识别