基于熟悉性的项目间语义关系再认

在项目间语义关系的熟悉性研究中, 一个关键问题是熟悉性效应依赖于哪种信息的加工, 关系概念还是抽象结构?研究通过两个实验, 借助无线索回忆再认范式(the recognition without cued recall paradigm, 简称RWCR范式)和四词类比材料, 采用混合设计, 直接分离与比较关系概念和抽象结构的加工对RWCR效应的影响。结果发现, 关系概念作为测验线索能够引发显著RWCR效应, 但仅仅学习关系概念却不能;只有学习关系概念的结构化材料时, 才能引发显著RWCR效应。该结果表明, 项目间语义关系的熟悉性过程与结构信息有关。

Familiarity-based Recognition of Inter-item Semantic Relations

Among cues that fail to elicit successful recall, participants can still discriminate based on familiarity between cues that do and do not resemble studied inter-item semantic relations. In the laboratory, this ability is referred to as recognition without cued recall (RWCR). One theoretical question is whether inter-item relational RWCR is driven by relational concept or by abstract structure processing. We hypothesized that the RWCR could be supported by abstract structure information between items.240 word pairs and relational concepts were used in RWCR paradigm, in which participants studied some items and were shown new items at test, half of which related to studied items and half of which did not. For each test item, participants first attempted to recall a related item from the study list. Then, regardless of whether successful recall occurred, participants were prompted to rate the familiarity of the test pair, which was said to indicate the likelihood that an item was similar to the test item studied. To examine how abstract structure vs. relational concept affected RWCR effect, Experiment 1 controlled test condition and used 2 study status (within-subjects: studied vs. unstudied)×2 test condition (between-subjects: word-pair vs. relation-concept), in which participants studied word pairs (e.g., lion-gazelle) and were shown two types of test cues (e.g., snake-mouse vs. prey on). Experiment 2 controlled study condition in which participants studied two types of materials (e.g., prey on vs. A prey on B) and were shown word pairs (e.g., snake-mouse).The results showed that participants demonstrated an ability to recognize inter-item semantic relations without cued recall through familiarity. RWCR effect was found only when relational concepts act as test cues rather than study materials. However, the processing of structured materials of relational concepts in study can cause RWCR effect.To conclude, the results support that RWCR effect of inter-item semantic relations depends on abstract structure processing at encoding.