Varga Nicole L, Bauer Patricia J
Department of Psychology, Emory University, 36 Eagle Row, Atlanta, GA, 30322, USA.
Mem Cognit. 2017 Aug;45(6):1014-1027. doi: 10.3758/s13421-017-0711-6.
The present research investigated the retention of new factual knowledge derived through integration of information acquired across temporally distributed learning episodes. Young adults were exposed to novel facts as they read long lists of seemingly unrelated information, one sentence at a time. They then were presented open-ended questions, the answers to which could be self-derived through integration of pairs of facts from the list. Experiment 1 was the first test of self-derivation of new factual knowledge through integration in adults using open-ended testing (as opposed to forced-choice testing). Participants successfully self-derived integrated knowledge under these more challenging conditions. Experiment 2 was a test for long-term retention of newly self-derived information. Newly derived knowledge remained accessible after a 1-week delay. Striking individual differences were also observed, which were related to whether individuals spontaneously identified the relational structure of the learning task. Insight into the relation between explicit task knowledge and strategic processing was also revealed through examination of response speed at the time of test. Specifically, knowledge of the task structure was associated with response latencies on unsuccessful (but not successful) trials, such that participants who were aware of the opportunity to integrate spent longer when they were subsequently unsuccessful, presumably reflecting directed search strategies and heightened perseverance when those processes failed. Together, the present findings provide direct evidence for the role of memory integration in the long-term accumulation of a semantic knowledge base and have theoretical implications for our understanding of this fundamental form of learning.
本研究调查了通过整合在时间上分散的学习片段中获取的信息而衍生的新事实性知识的留存情况。年轻人在阅读一长串看似无关的信息时，每次阅读一个句子，从而接触到新的事实。然后向他们提出开放式问题，这些问题的答案可以通过整合列表中的成对事实自行推导得出。实验1是首次使用开放式测试（与强制选择测试相对）对成年人通过整合自行推导新事实性知识的测试。在这些更具挑战性的条件下，参与者成功地自行推导得出了整合后的知识。实验2是对新自行推导信息的长期留存测试。经过1周的延迟后，新推导得出的知识仍然可以获取。还观察到了显著的个体差异，这些差异与个体是否自发识别学习任务的关系结构有关。通过检查测试时的反应速度，还揭示了对明确任务知识与策略性加工之间关系的洞察。具体而言，任务结构的知识与未成功（但非成功）试验时的反应潜伏期相关，即那些意识到有整合机会的参与者在随后未成功时花费的时间更长，这大概反映了定向搜索策略以及当这些过程失败时更强的毅力。总之，本研究结果为记忆整合在语义知识库长期积累中的作用提供了直接证据，并对我们理解这种基本学习形式具有理论意义。
