Abstract: This study probes into the cognitive processing mechanism of shape code during inputting Chinese characters. The study adopts three factors within-subject design: 2(Stroke number: More strokes, less strokes)× 2(Structure features: symmetrical, dissymmetrical)× 4 (Frequency : level 1-4). 29 students who are proficient in the CKC Chinese input system as subjects. 80 Chinese characters single words as experimental materials. The study finds that: 1)The cognitive processing of transforming the perceived component or strokes into Form code has not show strokes effect and frequency effect. 2)There are differences among the code number of Form code in the reaction time, and reaction time of less Code number is significantly faster than more code number. The series processing is very obvious. 3)The average reaction time of error code identification is significantly slower, and the average accuracy rate is lower.

摘要： 采用2(笔画数:多笔画、少笔画)×2(结构特征:构型对称、构型非对称)×4(频级:4级)三因素的被试内设计,通过分析29名有效被试的平均反应时和平均正确率来探讨纵横码认知加工的规律,从而揭示汉字输入中形码的认知加工模型,结果发现:(1)形码的认知加工没有显现出笔画效应、频率效应,说明其基本加工单元与汉字知觉的基本加工单元不同。(2)在反应时上存在形码码数的差异,码数少的反应时显著快于码数多的,系列加工的方式非常明显。(3)结构效应显著,且错误码识别的平均反应时显著更慢,而平均正确率更低,符合多层次格式塔双向加工模型的假设。