蒙语词汇识别中语音自动激活的ERP研究

采用快速掩蔽启动范式、Go/No-go范式和事件相关电位（ERP）技术,探讨蒙语词汇识别过程中语音自动激活的时程及对语义通达的作用。实验中要求被试执行一个额外的语义分类任务,对关键刺激不做明显的行为反应。结果发现,SOA为67ms时,语音启动在头皮的额区诱发了明显的P150和N400成分;SOA为167ms时,语音启动在头皮的额区和中央区诱发了明显的P200成分,在整个大脑皮层诱发了明显的N400成分。结果表明,语音在蒙语词汇识别的早期自动激活并促进语义的通达。     

纳西象形文字识别中的形、音、义激活

对纳西象形文字识别中的形、音、义激活进行了考察。实验1采用色词干扰的实验范式。结果表明,纳西象形文字的字形比字音和语义更易被激活。命名颜色字的形似字的颜色比命名颜色字的音同或音似字、语义联想字的颜色快。实验2采用色词干扰实验范式的变式。结果表明,当SOA=100ms时,颜色字及其形似字对颜色块命名的启动效应显著。当SOA为200ms和400ms时,颜色字的语义联想字对颜色块命名的启动效应显著。整个研究表明,在纳西象形文字识别中,字形信息首先被激活,其次是语义,字音的激活不明显。纳西象形文字的识别符合直通假设。所以如此,与纳西象形文字的特点有关     

Hearing a melody in different ways: multistability of metrical interpretation, reflected in rate limits of sensorimotor synchronization

Music commonly induces the feeling of a regular beat (i.e., a metrical structure) in listeners. However, musicians can also intentionally impose a beat (i.e., a metrical interpretation) on a metrically ambiguous passage. The present study aimed to provide objective evidence for this little-studied mental ability. Participants were prompted with musical notation to adopt different metrical interpretations of a cyclically repeated isochronous 12-note melody while tapping in synchrony with specified target tones in the melody. The target tones either coincided with the imposed beat (on-beat tapping) or did not (off-beat tapping). An adaptive staircase method was employed to determine the fastest tempo at which each synchronization task could be performed. For each metrical interpretation, a significant advantage for on-beat over off-beat tapping was obtained - except in a condition in which participants, instead of synchronizing, were in control of the target tones. By showing that a self-imposed beat can affect sensorimotor synchronization, the present results provide objective evidence for endogenous perceptual organization of metrical sequences. It is hypothesized that metrical interpretation rests upon covert rhythmic action.     

Mirror Neurons and the Evolution of Embodied Language

ABSTRACT— Mirror neurons are a class of neurons first discovered in the monkey premotor cortex that activate both when the monkey executes an action and when it observes the same action made by another individual. These neurons enable individuals to understand actions performed by others. Two subcategories of mirror neurons in monkeys activate when they listen to action sounds and when they observe communicative gestures made by others, respectively. The properties of mirror neurons could constitute a substrate from which more sophisticated forms of communication evolved; this would make sense, given the anatomical and functional homology between part of the monkey premotor cortex and Broca's area (the "speech" area of the brain) in humans. We hypothesize that several components of human language, including some aspects of phonology and syntax, could be embedded in the organizational properties of the motor system and that a deeper knowledge of this system could shed light on how language evolved.     

汉字识别中的同音字效应:语音影响字形加工的证据

两个实验考察汉字识别中语音对字形加工产生的影响。实验一采用视觉呈现材料的词汇判断法,没有发现如同英语研究中的同音字效应,实验二采用视、听跨通道的实验范式,首先呈现汉字的语音,然后进行视觉的词汇判断,结果发现,当出现的汉字为低频字时,其同音字越多,词汇判断的时间越长。这说明汉字的识别过程中,语音对低频字字形的加工有一定影响。     

句子语境中汉语词汇形、音作用的眼动研究

采用眼动方法, 通过两个实验考查了句子语境中汉语词汇形、音的作用及其作用的时间进程。结果发现：(1)在高限制性句子语境中, 音同的凝视时间和总注视时间显著短于无关控制; (2)在低限制性句子语境中, 音同、形似的首视时间均显著短于无关控制; (3)词频效应出现在低限制性句子语境中。该结果表明, 句子语境影响词汇形、音的作用及作用的时间进程, 汉语词汇的意义可以由字形直接通达, 也可以由形和音两条路径得到通达。     

汉语言语产生中语音、字形启动的位置效应

以起始掩蔽启动效应(the Masked Onset Priming Effect, 简称MOP效应)为研究背景, 通过3个实验, 考察了语音和字形在MOP效应产生中的作用, 以及在汉语单字词产生中语音、字形启动的位置效应。实验1表明, 在汉语单字词的产生中, 语音启动的位置效应显著, 声母启动的效应量显著大于韵母启动的效应量, 与拼音文字研究得到的MOP效应类似, 说明在MOP效应中至少包含有语音效应。汉字单字词语音编码的最小单位是音位, 而且具有从左到右序列加工的特点, 支持言语产生的“音节临时形成论”。实验2表明, 与目标字左、右部件相同的启动刺激均能促进掩蔽启动范式中的命名反应, 被试对与启动字的右部件相同的目标字的反应显著快于对与启动字的左部件相同的目标字的反应, 与起始掩蔽启动效应的假设相反, 说明在MOP效应中不包含有字形的作用, 却发现了汉字单字词特有的命名的字形位置效应：右部件启动的效应量大于左部件启动的效应量。实验3采用图-词干扰范式对汉字左、右部件的启动效应做了进一步的探讨, 仍然发现汉字右部件启动的效应量显著大于汉字左部件启动的效应量。整个研究表明, MOP效应是单纯的语音效应, 并不包含有字形的作用; MOP效应产生于言语产生中的语音编码阶段, 支持言语计划理论的观点。     

语音在汉字语义通达中作用的研究述评

语音在汉字语义通达中的作用是二十多年来心理语言学研究的热点之一。直通假设和形音交互假设是学者们主张的理论观点,它们在语音和语义的相互关系上存在分歧,并分别有各自的实验证据。从比较语音语义激活时间进程、同音字效应、神经心理学与脑成像、语音作用的发展转换这四个方面对过去的研究进行回顾,指出了实验范式和实验材料中存在的问题,并提出结合汉字特点进行研究的思路。     

Sometimes spelling is easier than phonemic segmentation

Poor spellers from the Netherlands segmented and spelled the same words on different occasions. If they base their spellings on the segmentations that they produce in the segmentation task, the correlation between segmentation and spelling scores should be high, and segmentation should not be more difficult than spelling. The predicted correlation was found, but spelling proved to be easier than segmentation, particularly for postvocalic consonant-clusters. Correct spellings despite incorrect segmentation were also more frequent than would be expected on the basis of unreliability of segmentation. Ways in which writing may facilitate and improve segmentation are therefore discussed. These findings raise doubts about the validity of segmentation test and about the efficiency of teaching children to segment without using letters.     

视觉经验缺失对盲人听觉词汇识别的影响

听觉词汇识别包含复杂的认知加工过程。视觉通道受阻的盲人在听觉词汇加工中具有一定听觉补偿优势; 但由于视觉经验的缺失, 盲人对一些视觉相关词(比如颜色词)的语义加工和理解比明眼人弱。未来的研究应对词汇的视觉相关性进行分类讨论; 对音、形、义等多层面及其神经生理机制进行深入探究, 发展符合盲人感知特点的听觉词汇加工模型; 并拓展不同年龄段的发展性研究。最终, 揭示视觉经验缺失对盲人听觉词汇识别影响机制的全貌。