PSYCH OpenIR  > 认知与发展心理学研究室
汉字组块加工式问题解决的脑机制探讨:来自EEG和FMRI的证据
庞娇艳
学位类型博士
导师罗劲
2010-05
学位授予单位中国科学院研究生院
学位授予地点北京
学位专业心理学
关键词汉字 组块加工 问题解决 功能性磁共振成像 事件相关频谱分析
摘要组块是由于对一类物体或事件的熟悉而产生的对其特征或成分稳定组合的认知模式,它作为一种独特的信息加工机制在人类的感知觉、记忆、学习和问题解决中都发挥着重要的作用。有研究表明专家的问题解决过程伴随着许多与问题相关的组块的使用,但由于知觉组块的应用是一个自动化的过程,在某些问题的解决过程中,知觉组块的自动激活却会使人形成对问题不适当的表征从而阻碍问题解决,这时就需要组块破解过程的参与。组块破解的概念最早是由OhlSSon于1990年提出的,指的是将不适当的知觉组块破解为其各个组成部分,从而使这些成分有可能以更合理有效的方式重新组合。Knoblich等人以火柴棍算式问题为实验材料系统的研究了组块破解理论,结果表明组块破解在改变问题表征方面起着重要的作用;随后它与约束解除(constraint relaxation)一起被认为是实现问题表征变换从而达到顿悟问题解决的两个主要认知操作。但是由于实验材料方面的限制,组块破解的神经机制一直没有得到研究。直到2006年,Luo等人发明了一种汉字组块破解任务并对由提示诱发的汉字组块破解过程的大脑机制进行了开创性的研究。但是汉字组块加工式问题解决的大脑机制尚没有得到充分的探讨,LuO的研究以特殊的研究范式为基础,探明了由提示诱发的顿悟式紧密组块破解的脑活动,但是代表一般性组块加工式问题解决的松散组块破解的脑活动没有被提及,顿悟与非顿悟式汉字组块破解在脑机制方面的差异以及这类问题解决在时间方面的特性也有待进一步探讨。本研究首先使用脑成像技术探讨自发的汉字组块破解过程,重点分析自发产生的松散组块破解和紧密组块破解各自的脑活动以及二者的差异,以验证Luo实验巾"诱发出来的组块破解过程与自发产生的加工过程类似"的假设,同时对松散组块的加工过程以及两类任务的差异进行初步的探索;随后,为了深入探讨顿悟与非顿悟类汉字组块加工任务的脑活动差异,我们将实验材料进行了细化,采用参数设计的方式来考察任务间脑活动在质和量上的差异;最后,使用脑电技术对汉字组块加工类问题解决的时间特性进行研究。对自发组块破解的脑成像研究验证了Luo的实验假设,自发组块破解引起了与诱发组块破解相类似的大脑激活,自发的紧密与松散组块破解因为任务性质而共享了大部分的脑区激活,差异仅存在于负责汉字细节和视觉心理想象加工的双侧锲叶;参数设计的脑成像研究结果发现随着组块紧密程度增加,大脑双侧枕叶和左侧前额叶区域的激活逐渐增强,而左侧顶叶和右侧前额叶区域的活动则因是否需要问题表征变换而存在质的差异;对汉字组块加工任务的脑电时频分析发现松散类问题解决在脑内是一个由枕叶负责知觉表征和特征整合的区域激活开始,经由中央前回对当前信息进行记忆提取和匹配最后到达前额叶对整个任务完成状态进行监控的自下而上的加工过程。本研究具有一定的理论意义,通过系列实验探讨组块加工的脑机制,丰富了组块及组块破解理论;并且探讨了-以表征变换为基础的顿悟问题解决的脑机制。
其他摘要Familiarity with a class of objects or events leads to the creation of patterns that capture recurring constellations of features or components, which is generally referred to as chunking processes. As an unique information processing mechanism, chunking plays an important role in human perception, memory, learning and problem solving. Expert problem solving performance is associated with a large repertoire of problem-relevant chunks, and the application of perceptual chunks is always an automatic process. However, the automatically inspiration of a chunk in some kind of problem solving may result in inapposite representations and thus hinder the processes, in which the chunk decomposition is needed. The chunk decomposition was firstly studied by Ohlsson in his research of geometry; it means to decompose the inappropriate chunks into their component features, so paving the way for an alternative parse of the problem situation. Knoblich et al., systematically studied the chunk decomposition theory in their matchstick problem solving experiments and confirmed its role in restructuring problems’ representation. They further put forward that chunk decomposition and constraint relaxation were the two main cognitive processes in changing problem's representation, which would finally lead to insight.Owing to the lack of proper experimental materials, no further investigation of brain mechanism underlying chunk decomposition was explored.
Until recently, Luo et al., developed a novel chunk-decomposition task to explored its brain mechanism. But there were not enough researches for the brain mechanism of Chinese chunking problem solving. Luo’s experiment was under a special paradigm and although it found the brain activities of the hint triggered insightful tight chunk decomposition, the loose chunk decomposition task which represents the general problem solving was not mentioned, and the difference of Chinese chunking problem solving with or without insight and its time characteristics needs further researches.
In this research, we firstly use the Fmri technique to explore the brain activities of the inner generated Chinese chunk decomposition processes, including the loose problems and tight problems and their differences, the propose is to confirm Luo’s assumption that solution hints trigger similar processes as internally generated solution attempts; then , in order to explore the unique brain mechanism of insight problem solving, we subdivided the materials into three groups and using a parametric design to find the quality and quantity difference in the brain activities of the tasks; finally, we use the time frequency analysis of EEG to study the time course of the problem solving.
Our first experiment confirmed Luo’s assumption, the internally generated solution attempts triggered similar processes as the solution hints; the loose and tight chunk decomposition share similar brain activations, the only difference was the stronger activation in bilateral cuneus in the tight tasks which related to the details of Chinese character and visual mental imagery; the result of the parametric fMRI experiment showed gradually increased activations in the bilateral occipital areas and the left prefrontal cortex with the strength of tightness, and the tight chunking tasks selectively activated the right prefrontal cortex and the left precuneus; the time frequency analysis of the EEG data revealed a bottom up information processing which includes an initially perceptual representation and feature integration in the occipital cortex, and then the prefrontal cortex retrieves related information with the answer and match them, finally the accomplishment of the task was monitored by the prefrontal cortex.
This research has important theoretical significance. Serial experiments were done to explore the brain mechanism of chunk processing; the results enriched the chunking theory, and also inspected the brain activation of insight problem solving which based on perceptual representation.
学科领域认知神经科学
语种中文
文献类型学位论文
条目标识符http://ir.psych.ac.cn/handle/311026/20322
专题认知与发展心理学研究室
作者单位中国科学院心理研究所
推荐引用方式
GB/T 7714
庞娇艳. 汉字组块加工式问题解决的脑机制探讨:来自EEG和FMRI的证据[D]. 北京. 中国科学院研究生院,2010.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
汉字组块加工式问题解决的脑机制探讨:来自(2459KB)学位论文 限制开放CC BY-NC-SA浏览 请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[庞娇艳]的文章
百度学术
百度学术中相似的文章
[庞娇艳]的文章
必应学术
必应学术中相似的文章
[庞娇艳]的文章
相关权益政策
暂无数据
收藏/分享
文件名: 汉字组块加工式问题解决的脑机制探讨:来自EEG和FMRI的证据.pdf
格式: Adobe PDF
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。