Suppr超能文献

阅读过程中汉字抽象正字法模式的快速内隐提取。

Rapid implicit extraction of abstract orthographic patterns of Chinese characters during reading.

机构信息

Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China.

Center for Psychology and Cognitive Science, Tsinghua University, Beijing, China.

出版信息

PLoS One. 2020 Feb 21;15(2):e0229590. doi: 10.1371/journal.pone.0229590. eCollection 2020.

Abstract

Orthographic processing is crucial in reading. For the Chinese language, sub-lexical processing has already taken place at radical level. Previous literature reported early position-specific radical representations and later position-general radical representations, implying a possible separating process of abstract position information irrespective of radicals per se from radical representations during orthographic processing. However, it remains largely unclear whether the abstract pattern of spatial arrangement of radicals can be rapidly extracted, and if so, whether this extraction takes place at the visual cortex, the very first processing center. As the visual cortex is documented to actively participate in orthographic processing, it may also play a role in the possible extraction of abstract orthographic patterns of Chinese characters. Hence, we hypothesize that abstract orthographic patterns of Chinese characters are covertly extracted at the visual cortex during reading. In this study, we investigated whether the visual cortex could rapidly extract abstract structural patterns of Chinese characters, using the event-related potential (ERP) technique. We adopted an active oddball paradigm with two types of deviant stimuli different only in one feature, structural or tonal, from standard stimuli; in each of the two sessions, subjects focused conscious attention on one feature and neglected the other. We observed that the ERPs recorded at occipital electrodes responded differentially to standard and structural deviant stimuli in both sessions, especially within the time range of the occipital P200 component. Then, we extracted three source waves arising from different levels of the visual cortex. Early response differences (from 88 to 456 ms after stimulus onset) were observed between the source waves, probably arising from left primary/secondary and bilateral associative visual cortices, in response to standard and deviant stimuli that violated abstract structural patterns, whether subjects focused their attention on the character structure or not. This suggests rapid extraction of abstract structural patterns of Chinese characters in the visual cortex, no matter the abstract structural pattern was explicit or implicit to subjects. Note that the source waves arising from right primary/secondary visual cortices in response to standard and structural deviant stimuli did not differ at all, indicating that this extraction of the abstract structural pattern of Chinese characters was left-lateralized. Besides, no difference was observed between source waves originating from any level of the visual cortex to standard and deviant stimuli that violated abstract tonal patterns, until 768 ms when a late effect related to conscious detection of targets occurred at higher levels of the visual cortex. Note that at late stages (later than 698 ms after stimulus onset), responses arising from bilateral associative visual cortices to standard and target stimuli differed for both sessions, no matter the structural or tonal feature was attended to. Our findings support the primitive intelligence of visual cortex to rapidly extract abstract orthographic patterns of Chinese characters that might be engaged in further lexical processing. Our findings also suggest that this rapid extraction can take place implicitly during reading.

摘要

正字法加工在阅读中至关重要。对于中文来说,亚词汇加工已经在部首水平上进行了。先前的文献报道了早期位置特异性部首表示和后来的位置一般性部首表示,这意味着在正字法加工过程中,可能存在一种将部首本身之外的抽象位置信息与部首表示分离的过程。然而,目前尚不清楚是否可以快速提取部首的抽象空间排列模式,以及如果可以,那么这种提取是否发生在视觉皮层,这是最初的处理中心。由于视觉皮层被证明积极参与正字法加工,因此它也可能在中文字符的抽象正字法模式的可能提取中发挥作用。因此,我们假设在阅读过程中,视觉皮层会在隐蔽处提取中文字符的抽象正字法模式。在这项研究中,我们使用事件相关电位 (ERP) 技术研究了视觉皮层是否可以快速提取汉字的抽象结构模式。我们采用了一种主动的奇偶校验范式,其中两种偏差刺激在结构或音调上与标准刺激仅存在一个特征差异;在两个会话中,受试者将注意力集中在一个特征上,而忽略了另一个特征。我们观察到,在两个会话中,在枕叶电极记录的 ERP 对标准和结构偏差刺激的反应不同,尤其是在枕叶 P200 成分的时间范围内。然后,我们从不同层次的视觉皮层提取了三个源波。在源波中观察到早期反应差异(在刺激开始后 88 到 456 毫秒之间),这可能是由于左初级/次要和双侧联合视觉皮层对违反抽象结构模式的标准和偏差刺激的反应,而与受试者是否关注字符结构无关。这表明在视觉皮层中快速提取汉字的抽象结构模式,无论抽象结构模式对受试者是明确的还是隐含的。请注意,源自右初级/次要视觉皮层的源波对标准和结构偏差刺激的反应完全没有差异,这表明对汉字的抽象结构模式的这种提取是左偏的。此外,在视觉皮层的任何水平上,对违反抽象音调模式的标准和偏差刺激的源波之间都没有差异,直到 768 毫秒时,更高水平的视觉皮层才出现与目标意识检测相关的晚期效应。请注意,在晚期阶段(刺激开始后 698 毫秒之后),来自双侧联合视觉皮层的对标准和目标刺激的反应在两个会话中都有所不同,无论关注的是结构还是音调特征。我们的发现支持视觉皮层具有快速提取汉字抽象正字法模式的原始智能,这可能涉及进一步的词汇处理。我们的发现还表明,这种快速提取可以在阅读过程中隐含地进行。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b597/7034908/73b21bbc8b68/pone.0229590.g001.jpg

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验