中文摘要：

本研究基于消费者性别角色的视角，探讨性别对消费者决策过程的影响，对认知心理学领域的结构匹配模型进行补充和延伸。本研究以学生群体为研究对象，采用情境模拟实验方法，具体考察了消费者对比信息处理过程时的性别差异。实验研究的结果显示，不同性别消费者的对比决策过程截然不同。具体来说，男性消费者在对比决策时比女性消费者更多地关注可以直接比较的属性信息，这种性别差异在高购买自信水平（和低购买自信水平相比）的情况下更为显著。同时，女性消费者在对比决策时比男性消费者更多地关注独特的差异化属性信息，这种性别差异在低购买自信水平（与高购买自信水平相比）的情况下更为显著。

英文摘要：

With the adventure of new communication media, consumers are able to compare between options and make tradeoffs before arriving at their final decision. However, there is a major void in the literature in terms of the effect of gender on consumers＂ comparative decision-making process. As such, understanding how female and male consumers cope with decisions that involve difficult trade-offs between choice alternatives has crucial significance for both marketing practice and academic research. Two experiments were proposed to examine the moderating roles of consumer characteristic（i.e., gender） and situational factor（i.e., confidence level） on the relationship between attribute alignability and consumer responses. Specifically, we hypothesize that female participants（compared to male participants） with low level of confidence emphasized nonalignable attributes and discounted alignable attributes, and low confidence led participants to rely more on nonalignable attributes. The experimental design for study 1 is a 2（gender： male vs. female） x2（brand： alignable-better brand A vs. nonalignable-better brand B） mixed design, with consumer gender being a between-subjects factor. Data analysis results showed the significant interaction effect between brand and consumer gender. Subsequent planned contrasts revealed that male participants＂ purchase likelihood ratings for alignable-better brand A were greater than for nonalignable-better brand B; whereas the reverse was true for female participants. Thus, both H1 and H2 were supported. Study 2 is a 2（confidence： low vs. high）x2（gender： male vs. female）x2（brand： alignable-better brand A vs. nonalignable-better brand B） mixed design. Participants＂ preferences （maximum 100 points） towards the two brands serve as the dependent variable. Consistent with our proposed H3, points allocated to the alignable-better option were higher in the high confidence condition than in the low confidence condition; while the mean preference all