Abstract: One crucial field of the research on retrieval-induced forgetting (RIF) is to explore the mechanism which underlying this phenomenon from the perspective of individual-differences. However, there remain some inconsistent findings so far, because of failing to consider the correlated costs and benefits problem. Such ignoring make it difficult to separate costs created by an inhibition process during retrieval practice from the benefits produced by an inhibition process in the final test. The goal of this study is to examine the effect of the correlated costs and benefits problem on RIF for undergraduate students, using two variables of item competitive intensity and response inhibition ability, and furtherly clarify the issue of the mechanism that underlies RIF. In the present experiment, the stop-signal task was first administered for 137 undergraduate students, and intended to distinguish the group with high levels of response inhibition ability from groups with low levels of response inhibition ability, consisting of 32 participants respectively. And then each participant from both groups was instructed to perform retrieval practice paradigm, which consisted of three stages including study, retrieval practice, and test, in an individual cubicles. Specifically, participants were asked to study 60 categories-exemplar pairs presented successively at 5-s rate on a computer screen, each exemplar together with its category name. The order of presentation was random with the constraint that no two pairs from the same category were presented consecutively. Immediately following the study phase, participants were cued to retrieve half of exemplars from half of the categories using a category-cued word-stem completion test (30% of ending strokes were removed from the first character of each exemplar). Each pairs were practice three times. Finally, after a limited 2-min distractor task, category-cued word-stem completion test that 50% of ending strokes were removed from the first character of each exemplar were administered for all studied exemplars. A 2 (response inhibition ability: high, low) × 2 (item competitive intensity: strong, weak) × 2 (item type: Rp+, Rp-, Nrp) mixed design were conducted. The findings found a significant difference in performance of stop-signal reaction time (SSRT) between two groups of participants assessed by stop-signal task, with high levels of response inhibition individuals, as compared with low levels of response inhibition individuals, presenting less SSRT. More importantly, in retrieval practice task following the stop-signal task, the results showed that participants with high levels of response inhibition ability presented significant RIF in the condition that unpracticed items interfered the recalling of practiced items strongly, but not in the condition in which the retrieval of practiced items were disrupted slightly. In contrast, participants with low levels of response inhibition ability didn’t show significant RIF for unpracticed items regardless of item competitive intensity. Taken together, the current results reveal that, for participants with a normal level of response inhibition ability, RIF can be attributed to inhibitory control that is generally considered as an intentional, resources-demanding processes. In particular, the finding supports that item competitive intensity is an effective operation that can separate the effect of the correlated costs and benefits problem from RIF.

Key words: retrieval-induced forgetting, item competitive intensity, response inhibition, the correlated costs and benefits problem

摘要： 采用提取练习范式，通过操作反应抑制能力和项目竞争强度两个变量，考察相关代价与效益问题（CCB）对大学生提取诱发遗忘（RIF）的影响。实验为2（反应抑制能力：高、低）×2（项目竞争强度：高、低）×3（项目类型：Rp+、Rp-、Nrp）三因素混合设计。结果发现，高反应抑制能力者在高项目竞争强度下产生了RIF，在低项目竞争强度下则未产生；低反应抑制能力者在高、低项目竞争强度下均未产生RIF。这些结果表明，对反应抑制能力正常的大学生而言，需要认知资源的抑制控制加工是RIF的产生机制，且这种抑制性RIF不受CCB的影响。

关键词: 提取诱发遗忘, 项目竞争强度, 反应抑制, 相关代价与效益问题