Department of Psychology, University of Illinois, Chicago.
J Exp Psychol Appl. 2024 Jun;30(2):318-330. doi: 10.1037/xap0000490. Epub 2023 Sep 4.
Teachers and students often report using examples to support learning. Research has shown benefits of viewing provided examples and generating examples during declarative concept learning; however, there is less work showing clear benefits when learners generate their own examples on comprehension measures while students are attempting to learn from expository science texts. The present study tested whether generating examples would be useful for improving comprehension and comprehension monitoring in the context of an undergraduate science course. In a pre-post design, students completed an initial reading activity, followed by taking practice tests on each topic. Some students were assigned to complete an additional example generation activity after taking the practice tests. Some students also evaluated the quality of generated examples and received explanatory feedback. While there was an overall improvement in comprehension for all students, those who generated examples without the opportunity to evaluate the quality had the smallest overall learning gains. Students who evaluated the quality of examples showed the greatest learning gains on application-based test questions. And, although overall overconfidence decreased, there were no differences between conditions. These results suggest that example generation may not always be as helpful for improving learning as many students and teachers assume. (PsycInfo Database Record (c) 2024 APA, all rights reserved).
教师和学生经常报告说使用示例来支持学习。研究表明，在陈述性概念学习中，观看提供的示例和生成示例都有好处；然而，当学习者在试图从说明文科学文本中学习时，生成自己的示例来衡量理解时，效果并不明显。本研究测试了在本科科学课程的背景下，生成示例是否有助于提高理解和理解监测。在前后设计中，学生完成初始阅读活动，然后对每个主题进行练习测试。一些学生在完成练习测试后被分配完成额外的示例生成活动。一些学生还评估了生成示例的质量并收到了解释性反馈。虽然所有学生的理解能力都有所提高，但那些没有机会评估示例质量的学生的整体学习收益最小。评估示例质量的学生在基于应用的测试问题上表现出最大的学习收益。并且，尽管整体过度自信有所下降，但各条件之间没有差异。这些结果表明，示例生成并不总是像许多学生和教师认为的那样有助于提高学习。（PsycInfo 数据库记录（c）2024 APA，保留所有权利）。
