Abstract: Relational analogical reasoning is the process of deriving a relation from one situation (the source) and applying it to another (the target). Based on the theories of the proactive brain (Bar, 2007, 2009), this study tested the influences of surface similarity and presentation mode (simultaneous versus sequential) on analogical relationship mapping. The study further compared the rate of relationship mapping when an identical match for the key object in the source was also present in the target (Experiment 1) and when an identical match was not present (Experiment 2).
The experiments adopted a 2 (presentation mode: simultaneous presentation, sequential presentation) × 2 (surface similarity: high similarity, low similarity) between-subject design. The picture mapping paradigm was a paper-and-pencil test, in which subjects viewed pairs of black-and-white sketches illustrating 30 themes (e.g., hanging an item). One item illustrating a relationship was circled in the source picture (e.g., a hat rack) and participants needed to circle the corresponding item in the target picture (e.g., a doorknob). The perceptual similarity between items in the source and target pictures was manipulated in different experimental conditions. High and low similarity items in the target pictures had the same relationship structure to the circled item in the source picture, but with different degrees of surface similarity. For example, a hat rack in the source image might map onto a bag hook (high similarity) or a doorknob (low similarity) in the target image. As for presentation mode, simultaneous presentation meant that the source picture and the target picture were presented at the top and bottom of a single page, so that participants could easily look back and forth between pictures. Sequential presentation meant that the source picture and the target picture were presented on the front and back sides of a single sheet, so that they could not be viewed at the same time. In Experiment 1, the target pictures always included identical matches for the critical items from the source images (e.g., a hat rack in the source image also appeared in the target image, even though the analogical relationship was to a different object). In Experiment 2, the identical matches were not present: the critical items from the source images were deleted in the target images. The sample sizes for Experiments 1 and 2 were 187 and 183, respectively. The primary dependent measure was the proportion of trials on which items were chosen based on relationship mapping.
Experiment 1 with identical matches found that the participants were more likely to choose items with the same relationship structure in the target pictures in the low-surface-similarity condition. Therefore, low surface similarity was more conducive to relational reasoning than high surface similarity. In contrast, Experiment 2 without identical matches found that the participants were more likely to choose items with the same relationship structure in the target pictures in the high-surface-similarity condition. Therefore, in this experiment, high surface similarity was more conducive to relational reasoning than low surface similarity, contrary to the results of Experiment 1. In both experiments, simultaneous presentation was more conducive to relational reasoning than sequential presentation.
The comparative analysis of the two experiments reveals an effect, which is called the "match effect" in this study. When an identical match for the critical object in the source image was present, low surface similarity promoted relational reasoning. However, when an identical match was absent, high surface similarity promoted relational reasoning. This effect indicates that the presence or absence of an identical match for critical objects across situations is one of the important conditions for stimulating the proactive brain to explore novel relationships. In addition, both experiments found that relational reasoning performance in simultaneous presentation was better than that in sequential presentation, which indicates that relational reasoning process relies on working memory since sequential presentation requires the participants to hold one image in working memory to compare it with the other. The match effect of relational reasoning can inform education. When people are presented with repetitive and similar ideas, their proactive brains are more likely to function and come up with more novel solutions through in-depth thinking.

Key words: relational reasoning, surface similarity, presentation mode, working memory, proactive brain

摘要： 关系类比推理是将一种情境中的某种关系推导至另一种情境中的过程,被试由“源问题”中隐含的某种结构关系映射到“靶问题”中也存在这种关系,即进行了关系类比推理。本研究采用图片匹配范式,通过两个实验考察了表面相似性与呈现方式对关系类比推理的影响,其中实验1探讨“源问题”和“靶问题”存在相同匹配物的情景,实验2探讨无相同匹配物的情景。实验1结果发现,同时呈现方式更有利于关系推理,低表面相似性更有利于关系推理;实验2发现,虽然同时呈现方式仍利于关系推理,但高表面相似性反而更利于关系推理。两个实验的对比分析揭示了一种“匹配物效应”：有相同匹配物情景下,低表面相似性比高表面相似性做出了更多的关系推理,而无相同匹配物情景下则出现了相反的结果。根据这些结果,讨论了“积极脑”的理论解释。

关键词: 关系推理, 表面相似性, 呈现方式, 工作记忆, 积极脑