归类优势与基本水平效应的再探讨

Re-examining the Classifying Advantage and Basic Level Effect

类别特异性分数是预测归类优势的有效指标。实验1考察该分数能否预测基本水平归类优势,结果:分数较高的表现归类优势。实验2检验有归类优势的下位类别的分数,结果:它们都有较高分数,同时发现下位类别的特异性可以影响分数。实验3探讨相同的基本水平与特异性不同的下位组合的归类表现,结果:下位特异性低就表现基本水平归类优势,高则相反。据此提出"经验说",分数高低与人们在经验中对各层次类别形成的表征特异性程度相关。

People are generally faster and more accurate to name or categorize objects at the basic level （e.g., dog）, which is known as basic level effect. Previous researchers interpreted basic level effect as the comprehensive effects of steep rise of within-category similarity and slow rise of between-category similarity. But such an explanation can’t explain why People are sometimes faster and more accurate to name or categorize objects at the subordinate （e.g., pigeon）. An operational method of computing the comprehensive effects was put forward in this study, which was called Scores of Category Specificity and defined as the difference between inferior between-category similarity and superior between-category similarity. We suggested that the category whose score was relatively high in its category series would show classifying advantages. Weather the scores could predict people’s classifying performance at basic-level categories was tested in Experiment 1. The results showed that subjects performed faster and more accurately at basic level in classifying tasks when the scores of basic-level categories were relatively high and that on the contrary was not true. In Experiment 2, we computing the scores of subordinate categories which had been proved that people were faster and more accurate to name or categorize objects at and referred to as special cases in previous studies. The results showed that people performed faster and more accurately at these subordinate categories just because the scores of these categories were relatively high, rather than interpreted as special cases. We found at the same time that specificity of subordinate categories had an impact on the scores. In Experiment 3, we further explored whether people had different performance when the same basic level category was combined with subordinate categories of different specificity. The results showed that people were faster and more accurate at basic level categories if they were combined with subordinate categories with low specificity and that people were faster and more accurate at subordinate categories if basic level categories were combined with subordinate categories with high specificity. Scores of Category Specificity were an effective indicator to predict at which level of categories people performed better in classifying tasks. The reason why scores of levels of categories varied was that people formed different degree of representational specificity according to various levels of taxonomical series in daily experience, which was known as Experience Theory.