PSYCH OpenIR  > 认知与发展心理学研究室
时间结构信息加工促进动态视觉意识的认知神经机制
Alternative TitleTemporal Structure Processing Fosters the Emergence of Dynamic Visual Conscious Experience: Cognitive and Neural Mechanisms
胡瑞晨
Contributor蒋毅 ; 王莹
2021-06
Abstract外界信息包含了丰富的时间结构,提取这些时间结构信息对于人类适应动态变化的环境具有重要的意义。已有研究表明,人类可以捕获并利用视觉信息时间结构中蕴含的规律,并借此促进注意和知觉等认知过程。但是,我们尚不清楚规律时间结构信息加工能否促进视觉意识内容的产生,以及如果该效应存在,那么其背后机制又如何从认知和神经层面得以实现。为此,本论文采用行为学和脑电相结合的方法,进行了两项研究,尝试对以上问题做出解答。 研究一采用双眼竞争范式,考察了基于语义和视觉特征的时间结构信息在意识内容生成中的作用。结果表明,相比于随机序列,人的意识体验更加自发地偏向于物理属性与随机序列一致,但包含规律的结构序列。值得注意的是,这种结构优势并不仅仅局限于基于视觉特征变化的知觉水平时间结构,还可以拓展到基于成语规则的语义水平时间结构。然而,两种水平的时间结构优势效应存在着明显的分离,集中体现在各自对信息整合窗口时空规律性的依赖以及对意识水平的要求上。研究一的发现揭示了意识和视觉信息时间整合之间的复杂关联,并强调了基于多水平规律的时间信息整合是促进连续意识经验产生的基本机制。 研究二采用EEG 技术,进一步探究不同水平时间结构信息的神经编码及其在意识内容生成过程中的加工机制。结果发现,知觉和语义水平的时间结构信息都可以同步化内生神经振荡,在刺激和结构频率上引发对应的响应,但只有结构频率上的响应强度可以预测被试在规律探测任务中的行为表现。不仅如此,后续分析还发现不同水平的时间结构信息共用同一个额顶-枕顶神经网络以完成对高层级时间结构的追踪,但是在相关的神经表征又存在一定的特异性。另一方面,采用双眼竞争范式,同时呈现两种频率不同的结构或随机序列,结果发现,知觉和语义水平的时间结构信息同样可以同步化内生神经振荡,然而对不同水平结构的响应存在意识阶段的分离:语义结构信息可以在意识上阶段增强大脑对当前主导序列的表征;与之相反,知觉结构信息可以在意识下阶段减弱大脑对当前主导序列的表征。研究二的发现揭示了多水平的时间结构信息如何通过一个共享的神经网络及刺激特异的神经表征在有意识和无意识阶段促进意识内容的连续产生。 综上所述,本论文的结果表明,在动态视觉竞争中,意识的内容偏向于具有规律时间结构的信息。这种时间结构优势可以推广至语义和知觉水平,但二者的优势在信息整合方面存在分离。上述行为表现可能是由部分共享但在刺激间分离的神经机制所介导,表明时间结构信息在视觉意识内容生成过程中的神经编码存在着一定的共性和特性。这些发现强调了规律时间结构在生成动态意识体验中的功能性作用,并初步阐明了多水平的时间结构信息在视觉意识中享有优势效应的认知及神经机制。
Other AbstractExtracting the temporal structure of sensory information is of great significance to our adaptive functioning in the dynamic environment. The existing research shows that humans are endowed with the capability to capture and exploit regularities embedded in the temporal structure of the visual inputs to assist in cognitive processes (e.g., attention and perception). Nevertheless, it remains unclear whether the processing of regular temporal structure confers a direct benefit to the emergence of conscious content in the continuum of time; and if so, how this effect is complemented at the cognitive and neural levels. To address these questions, here we conducted two studies using a combination of behavioral and electrophysiological methods. In Study 1, using the binocular rivalry paradigm, we demonstrated that awareness is spontaneously biased towards rhythmically structured visual information streams relative to their temporally randomized but otherwise matched counterparts. Notably, this effect was not restricted to perceptually defined temporal structures based on visual-feature changes but extended to idiomatically defined structures organized by idiomatic rules. However, there was a significant dissociation between the advantages of these perceptual- and semantic-level temporal structures, especially regarding their resistance to spatiotemporal perturbations and demands for conscious processing over the visual integration process. These findings reveal an intricate link between conscious awareness and temporal integration and highlight temporal integration by multi-level regularities as a fundamental mechanism to foster the emergence of continuous conscious experience. In Study 2, using the EEG techniques, we found that oscillatory cortical activity concurrently entrained to the stimulus rhythm and the structure rhythm for both semantic- and perceptual-level temporal structures. However, only the strength of the structure frequency correlated with participants’ behavioral performance related to the perception of the rhythmic temporal structure. More interestingly, a frontal occipitoparietal neural network was associated with the cortical tracking of temporal structures regardless of stimulus types, suggesting that the processing of semantic- and perceptual-level temporal structures may share a common mechanism. However, further analyses revealed that the neural representation of multi-level temporal structures could be well distinguished in terms of topographical activation patterns related to structure processing, suggesting the existence of stimulus-specific representations for the perceptually and semantically structured information. Furthermore, we recorded EEG during a BR experiment in which two structured/random streams competed with each other. We observed enhanced neural responses at the structure-related frequencies for semantic and perceptual-level temporal structures in the structured relative to the random condition. Further analysis revealed that relative to the random condition, the neural representation of a structured idiom stream was enhanced only when it became dominant in awareness, whereas the neural representation of a structured shape stream could suppress the representation of the rivalrous stream when it remained at the non-conscious stage. Taken together, these findings reveal how multi-level temporal structure facilitates the generation of conscious content both at the conscious and the unconscious stages, probably through a shared neural network but distinct neural representations. In summary, the results discussed in this dissertation demonstrated that during the dynamic visual competition, the content of visual awareness is biased towards information with regular temporal structures at both the perceptual- and semanticlevels. Despite the apparent resemblance, the observed privileges differentiate across information levels regarding the information integration process, with the cross-level effects mediated by partially overlapped but distinct neural mechanisms. Altogether, these novel findings stress the functional role of temporal structure in generating dynamic conscious experience and shed light on the cognitive and neural bases for the prioritization of multi-level temporal structures in conscious visual perception.
Keyword时间结构 视觉意识 信息整合 双眼竞争 脑电
Subtype博士
Language中文
Degree Name理学博士
Degree Discipline认知神经科学
Degree Grantor中国科学院心理研究所
Place of Conferral中国科学院心理研究所
Document Type学位论文
Identifierhttp://ir.psych.ac.cn/handle/311026/39609
Collection认知与发展心理学研究室
Recommended Citation
GB/T 7714
胡瑞晨. 时间结构信息加工促进动态视觉意识的认知神经机制[D]. 中国科学院心理研究所. 中国科学院心理研究所,2021.
Files in This Item:
File Name/Size DocType Version Access License
胡瑞晨-博士学位论文.pdf(8156KB)学位论文 限制开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[胡瑞晨]'s Articles
Baidu academic
Similar articles in Baidu academic
[胡瑞晨]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[胡瑞晨]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.