Abstract: This study explored the developmental tendency of young children’s (4～6 year old) metamemory monitoring (EOL/JOL/JOC), including the age differences on the level, the developing speed and the metamemory efficiency. There were 91 children randomly selected to participate. Items were 20 pairs of pictures, in which half were antonymous picture pairs, and the remaining half were unrelated pictures pairs. Two of 10 pairs were used to practice. The remaining 8 pairs were randomly displayed by computer for 3 s each, with an interval of 2 s. Before studying, participants were asked to make an aggregate EOL by responding to a question. Following the EOL, participants studied 8 pairs only once and made their JOL for 8 pairs by responding to a question followed by a cued-recall test. The experimenter instructed them to make an aggregate judgment of confidence. A MANOVA, Age(3)×Item difficulty (2:easy vs. hard), was carried out. The results yielded overall main effect was significant, Wilks’λ= .91, p <.05, for age, Wilks’λ= .72, p <.05, for item difficulty and Wilks’λ= .93, p =.58, for the interaction. The interaction of age and item difficulty was significant on JOC and we conducted a simple effect, the main effect of age was significant when task was difficult, M= 1.44, SD =1.46, for 4-year-old, M= .65, SD =.71, for 5-year-old, and M= .46, SD =.61, for 6-year-old. Secondly, the slope of development curve was computed to examine the differences in developmental rate (EOL/JOL/JOC). When the task was difficult the magnitude of slope as follows: k JOC = .79, k EOL = .06, k JOL =.40, for the age of 4-5, and k EOL =.71, k JOL =.54, k JOC =.19, for the age of 5-6. Finally, a Paired t–test was used to compare the differences between predicted recall and actual recall. Under the easy task condition, the difference were significant, tEOL(30)=-5.38,p<.001; tJOL(30)=-4.01,p<.001;tJOC(30)=-3.08,p<.01,for 4-year- old, tEOL(29)=-6.51,p<.001, only for EOL of 5-year-old, and the difference was not significant for 6-year-old . All of children overestimated their recall performance when the task was difficult. The conclusions as follows: children’s monitoring accuracy increases from 4 to 6 years old. Accuracy is better on an easy task than a difficult task. The developing speed of 3 monitoring judgments is different, under the difficult task condition, JOC is the first one at the age of 4～5, EOL and JOL develop quickly at the age of 5～6. All of children exhibit to overestimate their recall performance when the task was difficult. Whereas, under the easy task condition, 4-year-old children underestimate their actual recall; 5-year-old children exhibit accurate monitoring on JOL and JOC, but they underestimate their actual recall on EOL; 6-year-old children exhibit accurate monitoring on EOL, JOL and JOC.

Key words: Metamemory Monitoring, EOL, JOL, JOC

摘要： 采用不同难度图对为记忆任务，考察了4～6岁儿童学习难易判断(EOL), 学习判断（JOL）和提取自信心判断(JOC)的发展特点。结果表明：三种元记忆监测水平随年龄增长均不断提高，任务难度不同三种元记忆监测发展水平不同，对低难度任务的元记忆监测显著好于高难度任务；三种元记忆监测在低难度任务下没有发展速度差异，高难度任务下4～5岁时回溯性监测(JOC)发展较快，5～6岁前瞻性监测(EOL/JOL)发展加速；有效元记忆监测存在任务难度差异和年龄差异。

关键词: 元记忆监测, 任务难易判断, 学习程度判断, 提取自信心判断