University of California, Berkeley, Berkeley, CA 94720, USA.
Boston University, Boston, MA 02215, USA.
J Exp Child Psychol. 2020 Jan;189:104701. doi: 10.1016/j.jecp.2019.104701. Epub 2019 Oct 8.
Children use speakers' past accuracy to make inferences about novel word meanings those individuals provide in the future. An open question is whether children can retrospectively reevaluate information after learning that the source was inaccurate. We addressed this question in two experiments where a speaker first introduced labels for novel objects and then revealed that she is either accurate or inaccurate in naming familiar objects. Experiment 1 showed that 3.5- to 6.5-year-olds displayed enhanced performance on a word knowledge test when they had learned novel words from a speaker who then showed herself to be an accurate labeler as opposed to an inaccurate labeler. Experiment 2 replicated this procedure but had a different speaker provide inaccurate label information. This manipulation did not affect learning, suggesting that children discount speakers and are not simply influenced by the demands of processing inaccurate information. Together, these results indicate that 3.5- to 6.5-year-olds continue to monitor the speakers' accuracy after learning new words from them, update their beliefs as accuracy data become available, and selectively retain words learned from speakers who they deem to be epistemically competent.
儿童会利用说话者过去的准确性来推断他们未来提供的新单词的含义。一个悬而未决的问题是，儿童在得知来源不准确后，是否可以回顾性地重新评估信息。我们在两个实验中解决了这个问题，在第一个实验中，说话者首先介绍了新物体的标签，然后揭示她在命名熟悉的物体时是准确的还是不准确的。实验 1 表明，3.5 至 6.5 岁的儿童在从一个准确的标签提供者而不是不准确的标签提供者那里学习新单词后，在词汇知识测试中的表现得到了提高。实验 2 复制了这个程序，但使用了不同的说话者提供不准确的标签信息。这种操作并没有影响学习，这表明儿童会对说话者进行评估，而不是简单地受到处理不准确信息的要求的影响。总的来说，这些结果表明，3.5 至 6.5 岁的儿童在从他们那里学习新单词后，会继续监测说话者的准确性，根据准确性数据更新他们的信念，并选择性地保留他们认为具有认知能力的说话者教授的单词。
