题目(Title):The transformative nature of human episodic memory 演讲人(Speaker): 薛贵,教授,北京师范大学 主持人(Chair): 蒋田仔,研究院,中国科学院自动化研究所 时间(Time): 2021.07.02, 09:30 - 10:30 地点(Venue): 智能化大厦三层第一会议室 |
简介(Abstract):
Memory has long been conceived as a dynamic process. As Bartlett (1932) noted, "Remembering is not the re-excitation of innumerable fixed, lifeless and fragmentary traces, but rather an imaginative reconstruction or construction...". Nevertheless, existing studies have largely emphasized the reactivation of encoded neural representations during memory retention and retrieval, whereas the dynamic nature of memory are still poorly understood. To address this question, we employ functional magnetic imaging (fMRI), intracranial electroencephalogram (iEEG) and magnetoencephalography (MEG) imaging technology to examine the neural representations with high spatial and temporal resolutions. In addition, by linking neural representations with deep neural network model, we could infer the representational formats during different memory stages and in different brain regions. Our results suggest that the reactivated representations during long-term memory retrieval have been systematically transformed from the encoded representations, in terms of both the representational formats and the brain regions containing the representations. Furthermore, this representational transformation occurs much faster than previously thought: it starts right after stimulus offset and continue to evolve from short-term memory retention to long-term memory retrieval. Finally, it is suggested that greater representational transformation during encoding is predictive of better longterm memory, and effective learning strategies, such as spaced learning and retrieval practice, often involve greater representational transformation. These results not only help us understand how memory is constructed and reconstructed, but also hold the potential to understand the role of episodic memory in supporting other cognitive functions, such as problem-solving and creative thinking.
摘要(Biography):
薛贵,北京师范大学认知神经科学与学习国家重点实验室教授,麦戈文脑研究院PI,北脑学者,长江学者特聘教授,博士生导师。主要从事人类学习和记忆的认知及神经机制研究,包括记忆编码、提取和巩固和优化,执行功能、决策与适应性学习。近年来主持国家自然科学基金重点项目、973课题,中德、中以国际合作等项目。以第一作者或者通讯作者身份在Science, PNAS, TiCS, Current Biology, eLife, Journal of Neuroscience, Cerebral Cortex等学术刊物上发表100多篇具有影响力的学术论文。目前担任Cortex杂志副主编,Journal of Cognitive Neuroscience 咨询编委,《心理学报》编委。