长时重复启动范式下言语理解中的信息表征

Information Representation in Language Comprehension with the Long-Time Repetition Priming Paradigm

采用长时重复启动范式进一步探讨言语理解中信息表征的方式。实验1探讨听力理解中的信息表征,学习阶段采用句子理解任务,测验阶段进行图片判断任务,结果发现匹配和不匹配的图片的反应时和正确率没有显著差异,学过和未学过的图片的反应时和正确率也没有显著差异,说明听力理解中的信息表征是命题符号表征,而非知觉符号表征。实验2探讨阅读理解中的信息表征,结果同实验1,说明阅读理解中的信息表征也是命题符号表征,而非知觉符号表征。实验3在学习阶段和测验阶段都采用句子理解任务,结果发现学过的句子的反应时快于未学过的句子的反应时,学过的句子的正确率高于未学过的句子的正确率,说明前两个实验中没有出现学习效应是因为两个阶段任务类型差异较大造成的。综合3个实验的结果,可以初步得出结论,排除策略影响后,言语理解中没有产生知觉符号表征,先前句图匹配范式下的知觉符号表征很可能是一种策略性的表征。

The present study adopted the paradigm of repetition priming to explore the effect of participants＇ strategy on information representation in language comprehension. A study phase and a test phase were included in each experiment. In Experiment 1, the task in the study phase was sentence comprehension, and that in the test phase was picture judgment. The results showed that there was not perceptual symbol representation in listening comprehension when the paradigm of repetition priming was used, neither was the study effect. In Experiment 2, the reading comprehension task also showed thesame results as Experiment 1. In Experiment 3, the tasks in the study phase and test phase were the same, and the results showed that the reason why there was no repetition priming in Experiment 1 and 2 was due to the gaping difference of tasks between the two phases. Altogether, the results suggest that after eliminating the strategy used, there is no perceptual symbol representation representation in language but propositional sym comprehension.