汉语普通话声调感知的老年化效应：来自ERP的证据

研究考察汉语普通话老年人前注意阶段声调感知状况, 探究是否存在领域特殊的老年化。运用事件相关电位技术, 采用被动oddball范式诱发MMN回应, 控制领域一般性因素的影响。结果显示涉及范畴变化的声调和非语音音调诱发MMN强度衰减, 不涉及范畴变化的声调诱发MMN强度未衰退。研究结果表明在前注意阶段, 在特定汉语普通话声调范畴知识加工能力上存在领域特殊的衰退, 而不涉及母语音位知识的声调的感知存在领域特殊的一定程度的保留, 这一保留与时间维度上补偿机制的调用有关。受补偿机制调节, 语言加工呈现出衰退或保留等不同的老年化进程。     

不同语境中声调早期加工的作用及时间进程

作为一种声调语言,汉语词汇在语音构成上存在独特的特点。本研究利用汉语不同于非声调语言的特点,探讨已有研究中尚未澄清的问题,即汉语声调在早期自动加工和早期控制加工中对语义激活的作用,以及在不同认知阶段中汉语声调加工的神经机制。本研究拟采用ERP方法,结合LORETA源定位技术,通过Oddball范式考察以下内容在早期阶段,汉语声调加工的大脑激活模式,拓展以往基于非声调语言研究建立的口语加工理论的适用范围,为完善言语认知加工的理论模型提供新的实验证据。     

汉语声调加工过程的脑神经机制述评

汉语是一种声调语言。文章从三方面总结了汉语声调加工的脑神经机制研究:1)声调加工是语言性质,还是语音性质;2)声调加工和音段音位加工的异同;3)声调加工与语调加工的异同。首先,汉语声调加工是语言性质,虽然有右脑的参与,但主要表现为左脑偏侧化。其次,汉语声调是一个独立的语音单位,声调加工与音段音位加工的脑神经机制存在一定程度的分离。最后,声调加工与语调加工在脑神经机制上有一定相似性也有很大不同,虽然汉语语调加工主要是语言性质,但存在明显的语音性质。     

汉语声调认知加工的脑机制及语言经验的影响

汉语母语者对汉语声调的认知加工可分为亚词汇和词汇两种加工水平。亚词汇水平涉及对汉语声调的声学和语音加工,当无需主动注意参与时,现有研究在左右半球优势上尚未取得明确结论;当需要主动注意参与时,优势主要表现为左侧化,涉及左侧的布洛卡区、前运动区、前部脑岛、顶叶、顶枕联合区以及双侧听皮层附近脑区,这与语言加工双通路模型(HickokPoeppel,2007)中的背侧通路有很大重合。词汇水平涉及对汉语声调的识别,由双侧半球参与,顶叶、颞叶等部分脑区左侧优势更加明显,同时涉及颞中回等语言加工腹侧通路的核心区域。此外,个体对汉语声调认知加工的脑机制会受到语言经验的影响。     

汉语口语韵律的作用及其神经机制

汉语是一种声调语言, 在口语韵律表达中存在独特的特点。本项目利用汉语不同于非声调语言的特点, 探讨已有研究中尚未澄清的问题, 即汉语声调在早期自动加工中对语义激活的作用, 以及在不同认知阶段中, 汉语声调和语调加工的神经机制。这些问题是当前的研究热点, 但仍存在着各种分歧和激烈的争论。本项目拟采用ERP方法, 结合LORETA源定位技术, 通过不同的实验范式考察以下内容：(1)在早期自动加工过程中, 汉语声调对词汇语义激活的作用; (2)在早期阶段, 汉语声调和语调加工的大脑激活模式; (3)在晚期阶段, 汉语声调和语调加工的大脑激活模式。对这些问题的考察将有助于澄清当前的激烈争论, 拓展以往基于非声调语言研究建立的口语加工理论的适用范围, 为完善言语认知加工的理论模型提供新的实验证据。     

方言经验对早期沪语儿童汉语语音意识的影响

已有研究对汉语儿童的英语语音意识进行研究,且取得了很多有价值的研究成果,但对方言儿童的汉语语音意识的研究很少。汉语是一种具有众多方言的独特语言,许多方言和标准语的语言体系有着显著差异。具有方言和普通话早期经验的双语儿童在汉语语音意识上是否存在优势,有两种不同的看法：一是从心理语言学层面上,母语学习中获得的语言能力或普遍语法能促进第二语言的学习;二是从语言学层面上来讲,母语对第二语言的学习存在干扰作用。那么双语儿童在汉语语音意识上比单一普通话儿童是否存在优势呢？本研究将进行调查。实验选取72名4岁、5岁和6岁学前儿童为被试,每个年龄组各24人,其中沪语儿童和单一普通话儿童各12名。采用听觉呈现方法对所选被试进行汉语语音意识水平的测试,包括声调意识、声母意识和韵母意识三部分,并收集儿童的口语资料。结果表明: 虽然沪语儿童声母意识和韵母意识主效应不显著,但沪语儿童语音意识总成绩和声调意识成绩均比单一普通话儿童高;沪语儿童汉语语音意识优势不受年龄的影响。本研究启示是,上海地区学前儿童学习标准语（普通话）之外应鼓励使用沪语。     

辅音特征和声调识别中的耳优势

本文用双耳分听方法,用识别率和知党空间结构指数等指标,探讨左右耳在辅音特征和声调辨别上的优势问题。得到的主要结果是:(1)右耳辅音识别率高于左耳,声级对侧化效应有显著影响;(2)右耳辅音特征识别率高于左耳;(3)两耳的辅音知觉空间维度和维度的主次关系一致,辅音在知觉空间中的分布也无显著差异;(4)两耳声调识别率无差异,且声调知觉空间也极相似。     

汉语普通话语音审美特质在新闻播音中的体现

汉语语音在世界语言中具有独特美感。新闻播音是汉语普通话发音的标杆和榜样,能集中体现汉语普通话语音的审美特质。其主要表现在吐字的字正腔圆、声调的抑扬顿挫、语流的舒畅自然、表意的清晰准确、情声气的和谐统一等方面。     

2~4岁普通话儿童前注意阶段的声调感知机制

以MMN、p-MMR作为汉语词汇声调感知的神经关联物,探究2~4岁汉语普通话儿童前注意阶段对声调刺激的失匹配响应,关注范畴信息和偏差大小两个因素对儿童感知水平的影响。结果显示:在范畴间大偏差条件下(T1/T3),诱发明显的MMN;在范畴内或小偏差条件下(T3a/T3、T3b/T3、T2/T3),都未诱发显著的MMR。表明:2~4岁普通话儿童正处于声调感知能力的发展过程中,音位信息和声学信息共同影响其声调感知。     

声调在汉语音节感知中的作用

探讨超音段(如声调)与音段信息的共同作用机制, 对口语词汇识别研究具有重要的理论意义。有研究探讨了声调在口语词汇语义通达阶段的作用, 但在相对早期的音节感知阶段, 声调与声母、韵母的共同作用机制还缺乏系统的认识。本研究采用oddball实验范式, 通过两个行为实验考察了声调在汉语音节感知中的作用。实验1发现检测声调和声母变化的时间没有差异, 均比检测韵母变化的时间长, 表明在汉语音节感知中对声调的敏感性不及韵母。实验2发现声母和韵母的组合并没有明显优于对韵母的觉察, 但声调与声母或韵母的同时变化都促进了被试对偏差刺激的觉察, 表明声调通过与声母、韵母的结合来共同影响汉语音节的感知加工。研究结果在认知行为层面为声调在音节感知中的作用机制提供了直接的实验证据, 为进一步探讨超音段与音段信息共同作用的认知神经机制提供了基础。     

Dichotic listening performance to CVs between versus within groups

The present study is concerned with comparing group mean data between groups with individual performance within groups in dichotic listening (DL). It is argued that only comparing mean ear performance in dichotic listening between groups may obscure important individual differences within groups in DL performance. In the present study we compared between group vs. within group performance in four groups of subjects (male and female, right- and left-handers). The stimuli were 66 trials of consonant-vowel combinations with the six stop-consonants and the vowels a and u. The results showed a group mean right ear advantage for consonants in the two right-handed groups, and a left-ear advantage in the two let-handed groups. The analysis of the individual distributions within each group, however, showed clear differences between the male and female groups. Generally, though, only the male right-handed group revealed enough homogeneity within the group to warrant the conclusion that the group mean REA reflects a left hemisphere specialization for speech perception.     

Cerebral specialization and verbal-motor integration in adults with and without Down syndrome

Persons with Down syndrome (DS) tend to exhibit an atypical left ear-right hemisphere advantage (LEA) for the perception of speech sounds. In the present study, a recent adaptation of the dichotic listening procedure was employed to examine interhemispheric integration during the performance of a lateralized verbal-motor task. Although adults with DS (n = 13) demonstrated a right ear-left hemisphere advantage in the dichotic-motor task similar to their peers with (n = 14) and without undifferentiated developmental disabilities (n = 14), they showed an LEA in a free recall dichotic listening task. Based on a comparison of the laterality indices obtained from both dichotic listening procedures, it appears that the manifestation of lateral ear advantages in persons DS may dependent on the response requirements of the task.     

Cerebral specialization for speech production in persons with Down syndrome.

The study of cerebral specialization in persons with Down syndrome (DS) has revealed an anomalous pattern of organization. Specifically, dichotic listening studies (e.g., Elliott & Weeks, 1993) have suggested a left ear/right hemisphere dominance for speech perception for persons with DS. In the current investigation, the cerebral dominance for speech production was examined using the mouth asymmetry technique. In right-handed, nonhandicapped subjects, mouth asymmetry methodology has shown that during speech, the right side of the mouth opens sooner and to a larger degree then the left side (Graves, Goodglass, & Landis, 1982). The phenomenon of right mouth asymmetry (RMA) is believed to reflect the direct access that the musculature on the right side of the face has to the left hemisphere's speech production systems. This direct access may facilitate the transfer of innervatory patterns to the muscles on the right side of the face. In the present study, the lateralization for speech production was investigated in 10 right-handed participants with DS and 10 nonhandicapped subjects. A RMA at the initiation and end of speech production occurred for subjects in both groups. Surprisingly, the degree of asymmetry between groups did not differ, suggesting that the lateralization of speech production is similar for persons with and persons without DS. These results support the biological dissociation model (Elliott, Weeks, & Elliott, 1987), which holds that persons with DS display a unique dissociation between speech perception (right hemisphere) and speech production (left hemisphere).     

Dichotic Perception of Mandarin Tones by Chinese and American Listeners

The dichotic perception of Mandarin tones by native and nonnative listeners was examined in order to investigate the lateralization of lexical tone. Twenty American listeners with no tone language background and 20 Chinese listeners were asked to identify dichotically presented tone pairs by identifying which tone they heard in each ear. For the Chinese listeners, 57% of the total errors occurred via the left ear, indicating a significant right ear advantage. However, the American listeners revealed no significant ear preference, with 48% of the errors attributable to the left ear. These results indicated that Mandarin tones are predominantly processed in the left hemisphere by native Mandarin speakers, whereas they are bilaterally processed by American English speakers with no prior tone experience. The results also suggest that the left hemisphere superiority for native Mandarin tone processing is similar to native processing of other tone languages.     

Writing induces a right hemisphere linguistic advantage in dysphonetic dyslexic children: implications for attention and capacity models of laterality

An experiment demonstrated a complete hemispheric processing reversal in 10 male, dysphonetic dyslexic children that occurred during a dichotic listening test of their verbal working memory. Requiring a written response to dichotic digits produced a right hemisphere/left ear superiority in the dysphonetic dyslexics whereas normal subjects and other dyslexics maintained a left hemisphere/right ear advantage. This reversal was unaffected by changes in task difficulty. A second experiment assessed the influence on producing the reversal of concurrent manual interference with left hemisphere verbal processing (responding orally vs. manually) and selective right hemisphere priming (Forward Writing vs. Backward Writing). The dysphonetic children reverted to a strong left hemisphere superiority when recalling the dichotic digits orally. Backward writing produced no ear advantage in either direction. The findings suggest that dysphonetic dyslexia may be related to (1) left hemisphere processing demands that exceed capacity, (2) easily activated right hemisphere processing strategies and (3) failure to coordinate linguistic processing interhemispherically. The results supported a novel hybrid conceptualization of dyslexia consisting of a synthesis of selective activation, and dual processor-limited capacity, theories.     

Unilateral forced nostril breathing affects dichotic listening for emotional tones

Unilateral forced nostril breathing (UFNB) through the left nostril is associated with enhanced spatial abilities, whereas UFNB through the right nostril is associated with enhanced verbal abilities. However, the effects of UFNB on standard tasks of laterality (e.g., dichotic listening) are unknown. This study employed dichotic listening for word targets, which typically exhibits a right ear advantage (REA), and dichotic listening for emotional targets, which typically exhibits a left ear advantage (LEA). Participants were asked to breathe either through their dominant nostril (congruent UFNB) or to breathe through their non-dominant nostril (incongruent UFNB) for the entire testing session. There was a significant three-way interaction between the type of dichotic listening task, nostril dominance, and nostril assignment, with the expected REA for word targets and the expected LEA for emotional targets-with one exception. Right nostril dominant participants assigned to the congruent condition exhibited an LEA for emotional targets (p < .05). The other three groups exhibited the expected LEA/right hemisphere advantage for the detection of emotional targets and all groups exhibited the expected REA/left hemisphere advantage for detection of word targets. As such, possible bias induced by tactile stimulation of the contralateral face cannot account for these results. Thus, our data are consistent with a selective enhancement of right hemispheric lateralized functions by right nostril UFNB.     

Is there parallel and independent hemispheric processing of intonational and phonetic components of dichotic speech stimuli?

The possibility of hemisphere interaction in the processing of spoken language was studied in two dichotic listening experiments. The stimulus material consisted of six CV syllable triplets each spoken with each one of six intonation contours. In Experiment I, 15 aphasic patients, 8 patients with unilateral right hemisphere lesions, and 10 normal controls were asked to identify the four components of a dichotic item from a multiple-choice (MC) set comprising all possible CV triplets and intonation contours. In Experiment II, 30 normal subjects were required to identify either the right or left ear stimulus alone from an MC set comprising the right and left ear stimulus together with the two wrong combinations of right ear CV triplet with left ear intonation and vice versa. It is concluded from the results that the left hemisphere is capable of processing both phonetic and intonational information and that there is neither the necessity nor the tendency for right hemisphere participation in the perception of spoken language.     

On the relation between auditory spatial attention and auditory perceptual asymmetries.

Laterality investigators have typically interpreted any perceptual asymmetry as a direct expression of the functional organization of the brain. However, many other confounding factors, including the asymmetric distribution of attention, may also contribute to either the magnitude or the direction of any of these advantages. In two experiments, attention was manipulated in a dichotic listening paradigm by presenting a preexposural tone cue to the ear from which the subject was required to report. The time available to orient attention was manipulated by varying the time period between the onset of the cue and the onset of the trial (stimulus onset asynchrony, or SOA). Results indicated that a right ear advantage for the identification of verbal material obtained at a 150-msec SOA was almost completely eliminated at an SOA of 450 msec. In addition, the direction of the ear advantage for emotion identification was found to depend on task difficulty. A left ear advantage, apparent when task difficulty was minimal, was reversed to a right ear advantage when difficulty was increased. These data are taken as evidence that, when subjects are faced with a difficult dichotic task, there is a general tendency for right-handed subjects to bias their attention toward the right ear. Such a tendency is shown not only to have likely seriously compromised the results of past investigations of functional perceptual asymmetries but also to be inconsistent with previously proposed theories of dichotic listening performance.     

On the relation between auditory spatial attention and auditory perceptual asymmetries

Laterality investigators have typically interpreted any perceptual asymmetry as a direct expression of the functional organization of the brain. However, many other confounding factors, including the asymmetric distribution of attention, may also contribute to either the magnitude or the direction of any of these advantages. In two experiments, attention was manipulated in a dichotic listening paradigm by presenting a preexposural tone cue to the ear from which the subject was required to report. The time available to orient attention was manipulated by varying the time period between the onset of the cue and the onset of the trial (stimulus onset asynchrony, or SOA).Results indicated that a right ear advantage for the identification of verbal material obtained at a 150-msec SOA was almost completely eliminated at an SOA of 450 msec. In addition, the direction of the ear advantage for emotion identification was found to depend on task difficulty. A left ear advantage, apparent when task difficulty was minimal, was reversed to a right ear advantage when difficulty was increased. These data are taken as evidence that, when subjects are faced with a difficult dichotic task, there is a general tendency for right-handed subjects to bias their attention toward the right ear. Such a tendency is shown not only to have likely seriously compromised the results of past investigations of functional perceptual asymmetries but also to be inconsistent with previously proposed theories of dichotic listening performance.     

Ear asymmetry in perception of emotional non-verbal stimuli

Hemispheric asymmetry for perception of non-verbal emotional human voices was investigated in normal subjects by dichotic listening utilizing non-verbal responses. As in former experiments, a slight but significant left ear superiority was found. This finding suggests that mediation of these stimuli is done by the right hemisphere. The dominant role of the right hemisphere in this perceptual task is discussed in terms of an earlier development of this hemisphere.