Nelson Douglas L, Goodmon Leilani B
Department of Psychology, University of South Florida, Tampa, Florida 33620-8200, USA.
Mem Cognit. 2003 Jan;31(1):65-76. doi: 10.3758/bf03196083.
Sometimes in conversation, something is said that causes us to want to comment, but before our impending but implicit thoughts can be expressed, the conversation is disrupted. Later, we cannot recall what we wanted to say, but still later, we can. We used the extralist cuing task to model this phenomenon, and across experiments we varied the strength, direction, and directness of the relationship between the retrieval cues and the targeted information. Disruption was varied by switching attention to a different task before testing and by changing the testing context. Such disruptions reduced recall for the target and its implicitly activated memories. Following a disruption, stronger cues that were related to the target or to both the target and its implicit memories were more effective than those that were related to implicit memories. The findings were consistent with a model of long-term working memory that attributes forgetting to a loss of access to what has been activated, which loss is relative to the strength of the retrieval cue. Decay alone does not explain the results, indicating that many models of working memory need to be revised to take the nature of retrieval cues into account.
有时在交谈中,会有人说出一些话让我们想要发表评论,但在我们即将表达那些潜在的想法之前,交谈就被打断了。之后,我们想不起来自己想说什么,但再后来,我们又能想起来了。我们使用额外列表线索任务来模拟这一现象,在各个实验中,我们改变了检索线索与目标信息之间关系的强度、方向和直接程度。通过在测试前将注意力转移到不同任务以及改变测试环境来改变干扰情况。这种干扰降低了对目标及其隐含激活记忆的回忆。干扰之后,与目标相关或与目标及其隐含记忆都相关的更强线索比那些仅与隐含记忆相关的线索更有效。这些发现与长期工作记忆模型一致,该模型将遗忘归因于无法获取已激活的信息,这种损失与检索线索的强度有关。仅衰退并不能解释这些结果,这表明许多工作记忆模型需要进行修订以考虑检索线索的性质。