中文摘要：

本研究探讨了色词与颜色重组训练能否让被试习得与真实情境相似的色词与颜色范畴联结，并引起相应的偏侧化颜色范畴知觉，从而进一步揭示语言编码对颜色范畴知觉的影响。采用六种蓝色作为材料。B1与B2为先前研究的两种蓝色。B11和B12，B21和B22分别为使B1，B2变浅和变深所得，四者是连续的蓝色。训练被试分别用“duan”和“ken”命名B1和B2。由于被试可能习得色词与颜色范畴的联结，因此B12与B21为可能变范畴间颜色，B11和B12，B21和B22为可能变范畴内颜色。训练前后分别让被试完成一个视觉搜索测验。结果发现：前测中可能变范畴间颜色的辨别优于可能变范畴内颜色，且在左视野呈现时更显著；后测中可能变范畴问颜色的辨别同样优于可能变范畴内颜色，但在右视野呈现时更显著。这提示：（1）人们在知觉同一语言范畴的不同颜色时可能会区分深浅范畴，并因此引发右半球颜色范畴知觉；（2）短期习得的语言范畴能引起偏左半球颜色范畴知觉，且使右半球颜色范畴知觉转为左半球颜色范畴知觉；（3）被试在训练中习得了色词与颜色范畴的联结；（4）范畴学习可以在仅学习一个样例的条件下自动发生。

英文摘要：

Zhou et al. （2010） demonstrated that newly trained lexical categories produced lateralized categorical perception （CP） of color, suggesting that color CP can be a result of learned categories. However, since the way in which participants learned the name of a color in their study different from the one in actual life, the ecological validity of this study can be challenged. In actual life, a color name is connected with a color category. For example, the word ＂blue＂ refers to all kinds of blue. Can the training used by Zhou et al. （2010） induce participants to learn such a connection between the color name and a color and therefore produce color CP？ In order to reveal the relationship between lexical categories and color CP more exactly, the present study probed into this question. Moreover, the present study also aimed at investigating the cognitive mechanism in acquired category effects. Six colors were used in the present study. B 1 and B2 were light blue and dark blue. By making B 1 and B2 lighter and darker, we obtained B 11 and B 12, B21 and B22 respectively. B 11,B 12, B21 and B22 formed a graded series from light to dark blue. Participants were trained to name B1 and B2 with two meaningless syllables du~n and k6n, respectively, in six training phases. Given that participants may connect the color name to a color category after training, we expected BI2B21 to be between-category color, but BllB12 and B21B22 to be within-category color pairs. Participants also completed a visual search task before and after the training, in which they were asked to search a target color from a ring of 12 colored squares surrounding a fixation marker, and indicated whether the target was on the left or right side of the circle by making button-press responses. The data showed that, in the visual search task before training, reaction times to targets were faster for between-category color pair candidate than for within-category color pair candidates. Moreover, this effect was more significant when the targ