中文單字辨識中的鄰群數目效果：促進？抑制？

Abstract

在中文單字辨識的領域，近幾年湧現出大量有關鄰群效果的研究。對於中文單字辨識中鄰群數目呈現促進還是抑制的效果，研究者們觀點不一，且此議題始終處於熱議中(Bi et al., 2006; Zhang & Jiang, 2008; Bi & Weng, 2008; Zhang & Wang, 2010; Li et al., 2010; Li et al., 2011). 為了釐清此問題，本研究設計了兩個實驗，這兩個實驗的刺激材料分別為不規則不一致的形聲字（實驗1），規則一致的形聲字（實驗2）. 研究者操弄了目標字的鄰群數目和更高頻鄰居的數量，并控制住了字頻。實驗結果顯示，無論目標字的規則性和一致性如何，研究者都觀察到了抑制的鄰群數目效果。而且無論目標字是否有更高頻鄰居，依然觀察到抑制的鄰群數目效果。研究者據此提出了中文單字辨識中鄰群數目效果背後的可能機制，并針對中文單字和雙字詞中的鄰群效果勾勒出整體的理論框架。

In recent years, there has been an increase of research investigating neighborhood effect in Chinese character recognition. As for whether orthographic neighborhood size (NS) effect reveals facilitative or inhibitory in Chinese character recognition, opinions vary and the question is still under debate (Bi, Hu & Weng, 2006; Zhang & Jiang, 2008; Bi & Weng, 2008; Zhang & Wang, 2010; Li, Bi & Zhang, 2010; Li, Bi, Wei & Chen, 2011). To clarify this topic, the present study implemented two experiments, the materials of which respectively were irregular & inconsistent phonograms in Experiment 1, and regular & consistent phonograms in Experiment 2. Researchers manipulated the orthographic neighborhood size and the number of higher frequency neighbors (HFNs) of the target characters, with the character frequency controlled. Results indicated that, regardless of the regularity and consistency of the stimuli, an inhibitory NS effect was observed. And both in the conditions of targets without HFNs and with HFNs, the NS effect was stably inhibitory. Researchers raised the possible underlying mechanism of the orthographic neighborhood size effect in Chinese character recognition, and exploited a global theoretical framework involving the neighborhood effect of characters and words in Chinese.