信息提取的基本层加工优势及其转移神经机制研究

来源 :北京工业大学 | 被引量 : 0次 | 上传用户:jueduizhi
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
Information retrieval plays an important role in high level cognitive activities such as learning,problem solving.It is commonly held that objects are first identified at the basic level.However,there are researches suggest that basic level advantage do not necessarily take place.It will switch to subordinate level or superordinate level.A possible explanation for different level precedence may relate to external differences in the processing demands of task used in the experiment.The functional magnetic resonance imaging(Fmri),with the high spatial resolution,can probes the brain activies.It is revealed that some regions are task-specific.Therefore,the present study will take effort to reveal the neural mechanism underlying different cognitive tasks.Two tasks are designed to measure information retrieval performance,word-picture matching task and picture-word matching task respectively.They are exactly the same excepting the reverse order of the word and picture stimuli.Behavioural results have shown that superordinate concepts were performed most quickly and differentiated from intermediate and subordinate concepts in word-picture matching task,however,in word-picture matching task,intermediate concepts took shortest to perform,Fmri results are consistent with previous studies in areas involved with memory retrieval and semantic processing,some areas are commonly actived in both tasks,mainly including inferior frontal gyus,anterior cingulated gyus,inferior parietal lobe,fusiform and occipital lobe.Contrast analysis show that areas of inferior parietal lobe(BA40)and fusiform gyri(BA37)were activated more strongly in word-picture matching task compared with picture-word matching task.This may reflect much visual requirements in this task.In picture-word matching task,however,activations in IFG(BA45A7)may relate to semantic process.Besides,between-level differences in each task suggeste that inferior parietal lobe(BA40)are mainly activated at the precedence level.Further functional connectivity analysis demonstrates that connections between regions are simplest at the superordinate level in word-picture matching task while at mediate level in picture-word matching task.On the contrary,More cognitive resources are needed when participants identify items at subordinate level.It is also found that regions connect closely with inferior parietal lobe and fusiform gyri in word-picture matching task while in picture-word matching task they are highly correlated with inferior frontal gyus.All of these results strengthen the hypothesis that Word-picture matching task may load up more visual-based processes since comparison between mental imagery and visual stimuli.Howerver,in the case of picture-word matching task,semantic memory retrieval is much involved.In conclusion,our research provides evidence for the viewpoint that external environment such as the processing demands of tasks has its effect over information retrieval strategy.
其他文献
近些年,随着互联网的不断发展,数据交换也变得频繁,由于XML的可移植性以及自述性等特点,使得它逐渐成为商业、工业以及生活中重要的一种数据交换标准。XML不断地被应用,从而
烟雾和云的模拟广泛应用于电脑游戏、影视特技、科学计算可视化以及军事模拟和训练等领域,是计算机图形学的一个热点问题。本文采用基于物理的方法对烟雾和云进行模拟,在基于
计算机网络的普及伴随着业务复杂程度的提高,使得新发布的基于Web的应用产品使用难度大幅提高。而且基于Web的应用普遍缺乏对普通用户的实时技术支持,使得新的技术或产品在较长
互联网的普及和信息技术的快速发展,网络数据呈指数级增长,尤其近年社交网站和电子商务网站的兴起与迅猛发展,出现了“信息资源过载”和“数据爆炸”的现象。这些现象已经成
对等网络(Peer-to-Peer Network,简称P2P)是一个完全非中心化、自组织的资源使用系统。它打破了传统的“客户/服务器”模式,具有高效、高扩展性、良好的容错性等优点,是当今
对等计算(Peer-to-Peer Computing,P2P)是在互联网上实施网络计算的一种新的计算模型。P2P打破了传统的Client/Server(C/S)模式,网络中的每个节点是逻辑对等的,拥有对等的功
描述逻辑是人工智能的逻辑基础,能够对客观世界的概念和角色进行表示和推理。但是由于概念的扩展有严格的限制,传统的描述逻辑系统不能够处理粗糙的(或模糊的)概念。本文将基
形态学联想记忆(Morphological Associative Memories,简称MAM)是一种新型的联想记忆模型。在无噪声情形下,形态学自联想记忆(Morphologicalauto-Associative Memories,简称a
大量的端到端(P2P)业务尤其是共享较大视频类文件的软件以及下载软件已占据了越来越多的互联网业务总量。由此带来的带宽的巨大消耗所引起的网络拥塞以及网络性能的降低使P2P
近年来,集中供热已成为国家大力推行的节能与环保的有效措施,在我国的北部供热区域已经被广泛采用。为了提高供热效率,供热计量方式也逐渐由原先按建筑面积和供热单价进行僵