情绪韵律调节情绪词识别的ERP研究

采用韵律/词汇干扰范式和延迟匹配任务, 通过两个ERP实验, 考察了汉语口语中情绪韵律能否、以及如何调节情绪词的识别。实验一中, 不同类型的情绪韵律分组呈现, ERP结果显示, 同与情绪韵律效价一致的情绪词相比, 与情绪韵律效价不一致的情绪词诱发了走向更负的P200、N300和N400成分; 实验二中, 不同类型的情绪韵律随机呈现, 上述效价一致性效应依然存在。实验结果表明, 情绪韵律能够调节情绪词识别, 主要表现在对情绪词的音韵编码和语义加工的双重易化上。     

情绪词加工过程及其情绪效应特点：ERP的证据

语言是传递情绪信息的重要媒介, 探讨情绪语言的加工过程一直是情绪研究的重点。文章聚焦于情绪词加工的ERP研究, 介绍了情绪词加工中的ERP成分、效价与唤醒度的作用关系, 以及情绪词情绪效应的特点。情绪词加工的ERP成分包括非常早期的P1-N1成分, 早期的EPN成分和晚期的LPP成分; 效价和唤醒度均在情绪词加工中具有重要作用, 但两者的关系还未得到一致结论。相比情绪图片, 情绪词加工所引发的情绪效应更弱, 并且存在积极效价偏向。今后的研究还需进一步探讨影响情绪词情绪效应产生的相关因素; 进一步区分情绪标注词和情绪负载词, 探讨两类情绪词加工的异同。此外, 结合眼动追踪技术探讨自然语境中的情绪词加工机制也是未来研究的一个重要方向。     

神经重用假说的汉语情绪词汇加工证据

神经重用假说认为对词的情绪效应可以不经语义形成而先行产生, 这更有助于人类适应环境。为了检验这一假说, 设计脑电实验和行为实验, 记录被试加工汉语厌恶情绪词和中性词的特征。结果发现, 170 ms左右厌恶词与中性词的ERPs发生分离, 且在枕部出现情绪刺激视觉注意关联的EPN; 厌恶词与中性词N400差异波溯源定位于脑岛附近, 其最大激活点时间窗口开始于380 ms。这表明在情绪词汇的语义分析之前就出现了情绪效应, 结果支持了神经重用假说。     

高频靶词情绪启动的反转效应

实验采用词-词情绪启动范式,在SOA 5个水平上以高频情绪词为靶词,考察情绪启动的反转效应。结果在SOA各个水平上均发现反转效应,即被试对与启动刺激效价一致的靶刺激的反应较慢,而对与启动刺激效价不一致的靶刺激的反应较快。实验结果支持情绪转换模型和快速-准确策略理论。     

负性情绪词错误再认的跨语言实验

根据词语情绪标准编制汉语双字词关联词表和英汉双语关联词表,考察DRM任务下的英汉双语者对负性情绪诱词产生的语言内、语言间关联性错误再认。结果表明：（1）汉语双字词语义关联下,情绪色彩的强度影响错误再认水平,负性诱词的错误再认率显著高于中性诱词;（2）英汉双语语义关联下,错误再认发生跨语言现象,汉语负性诱词的错误再认显著高于英语负性诱词与汉语中性诱词的错误再认;（3）汉语负性诱词的语言间关联性错误再认率显著增加, 语言类型、词表类型与情绪强度有显著交互作用。     

唤醒度对汉语情绪词情绪启动效应的影响

使用词汇情绪启动的实验范式,通过两个行为实验探讨了不同SOA(100ms、300ms和500ms)条件下情绪唤醒度在情绪启动效应中的作用,以及情绪启动效应是否受到任务形式的影响。结果发现,在情绪效价判断任务中,SOA为100ms时高唤醒和低唤醒启动词都出现了情绪启动效应,SOA为300ms时只有高唤醒启动词出现了情绪启动效应。在真假词判断任务中,无论高唤醒还是低唤醒启动词都没有出现情绪启动效应。唤醒度是影响词汇情绪启动效应的重要因素,但这种效应不是完全自动化的过程,会受到SOA和任务导向的影响。情绪启动效应可能来自于启动词与目标词在反应趋向上的一致性趋势。     

基于情绪产生离身观和具身观的时距知觉情绪效应

时距知觉情绪效应指个体时距知觉因情绪影响而相对扭曲的现象。这一效应基于研究者所持“情绪观”的差异而被从不同视角进行理论解释。情绪离身观主张情绪产生基于对情绪刺激本身维度的加工。离身观下的时距知觉情绪效应可以用单一唤醒机制、单一注意机制、单一工作记忆衰减机制、唤醒和注意共同受唤醒度调节机制以及唤醒、注意及工作记忆受目标时间长度动态调节机制等进行解释,情绪具身观主张情绪产生是身体和情绪信息共同参与的结果。具身观下时距知觉情绪效应可以用具身化调节内部时钟速率、内感受性强度以及神经能量高低等机制进行解释。未来研究应注意：其一,基于离身情绪观需要首先明确“唤醒”、“注意”在具体研究的含义,操纵并观测“唤醒”和“注意”等变量主效应及其他变量（机体或环境变量）的调节效应。同时,厘清时距知觉情绪效应的判定标准,并为情绪刺激时距知觉分段综合假说提供佐证;其二,基于具身情绪观需要首先澄清内感受性强度、神经能量在具体研究中的内涵,然后,厘清时距知觉情绪效应产生机制的因果链,并为情绪刺激具身化调节时距知觉假说提供佐证。其三,需要在解释时距知觉情绪效应之前首先确立应该持何种情绪观的前提条件,然后再探索何种因素在...     

海洛因戒断者的阈下情绪启动研究

研究表明海洛因戒断者的情绪加工异常,但其阈下情绪加工是否异常尚无定论。本研究采用阈下情绪启动范式,考察其阈下情绪加工能力。结果：戒断组的反应时显著高于控制组;两组中,相比不一致、控制状态,一致状态反应时均显著最短;在戒断组中,不一致状态的反应时显著大于控制状态,负性图片的反应时显著大于正性图片,而在控制组中差异不显著。结论：戒断者存在阈下情绪启动效应,阈下情绪加工存在正性偏向,且能力弱于正常人。     

中性抽象词和具体词情绪信息获得的差异

采用词汇迫选学习范式考察不同类型词汇获得情绪信息的认知机制。结果发现：（1）在迫选过程中与积极图片匹配的具体词汇喜好度显著高于与消极图片匹配的喜好度,而抽象词学习并未表现出这样的效应。（2）积极图片下被选择词汇的喜好度显著高于未被选择词汇的喜好度,而消极图片下被选择词汇与未被选择词汇的喜好度差异不显著。该结果表明具体词相较于抽象词更易获得相应情绪信息,积极情绪比消极情绪信息更易被习得。     

时间间隔和情绪效价对双字词重复启动效应的影响

已有研究大多以中性刺激为材料,尚缺乏同时关注预存知识表征和情绪效价对重复启动效应影响及时间间隔对情绪材料相应效应调节机制的探究。本研究以国际情绪词库翻译而成的汉语双字词为材料并控制其效价,旨在探讨时间间隔对不同情绪效价词的重复启动效应的影响,并与转置词的结果相比较。结果显示,转置词较标准词的重复启动效应更强;立即重复较长间隔条件的重复启动效应更强;情绪效价与时间间隔交互影响词的重复启动效应。上述结果表明,情绪效价对重复启动效应的影响受时间间隔调节且支持注意资源缩减理论;该作用为材料是否具有预存知识表征所调节,支持自下而上加工的观点。     

Affective picture processing is modulated by emotion word type in masked priming paradigm: an event-related potential study

ABSTRACT

Although a large body of research demonstrates the role of language in emotion processing (e.g. emotional facial expressions), how emotion-laden words (e.g. poison, reward) and emotion-label words (e.g. fear, satisfaction) differently impact affective picture processing is not clear. Emotion-label words label affective states straightforwardly, whereas emotion-laden words engender emotion via reflection. The current study adopted the masked priming paradigm to examine how Chinese emotion-laden words and emotion-label words distinctively influence affective picture processing. Twenty Chinese speakers decided the valence of the pictures with their cortical responses recorded. Emotion-label words facilitated affective picture evaluation behaviourally. Moreover, pictures that were preceded by emotion-laden words generated larger electrophysiological activation than those preceded by emotion-label words. Combined behavioural and ERP evidence revealed that emotion word type modulated affective picture processing, suggesting different roles of emotion-laden and emotion-label words in how emotion is shaped by language.     

The face wins: Stronger automatic processing of affect in facial expressions than words in a modified Stroop task

Affective words and faces each seem to be evaluated automatically, but it is unclear if they differ from one another in perceptual salience as measured by automaticity. The current study examined a possible hierarchy among affective stimuli using a modified photo–word Stroop task. Positive and negative words were superimposed across faces expressing positive (happy) and negative (angry, sad) emotions. Participants categorised the valence of faces and words. Across two experiments, interference effects were seen for both stimulus types. However, larger interference effects occurred during judgements of words than expressions, suggesting affective faces are processed more automatically than affective words. When the strength of positive and negative expressions is compared, angry expressions resulted in a larger interference effect than sad, and happy expressions produce interference similar to that of angry faces. The latter result contrasts with research suggesting potential threat stimuli are processed more automatically than positive stimuli. Implications are discussed.     

内、外部情绪标注对情绪作用的ERPs研究

以往情绪标注的研究发现描述情绪面孔的情绪（外部情绪标注）可以降低被试的情绪。但是描述自己的情绪（内部情绪标注）结果如何尚存争议。研究采用了ERPs技术,以愉快和悲伤情绪面孔为实验材料,选择18名大学本科生为被试,比较了内部情绪标注、外部情绪标注及性别标注之间的差异。结果发现：外部情绪标注和性别标注相比,其LPP波幅均小于性别标注下LPP波幅,并且这种差异在Pz、CPz、Cz、FCz、Fz五个电极点均类似;而内部情绪标注和性别标注相比,其LPP波幅均小于性别标注下LPP波幅,但仅表现在Pz、CPz、Cz三个点上。这说明：无论内部情绪标注还是外部情绪标注,在命名后均能抑制情绪。研究结果支持阻断理论。     

Processing emotional category congruency between emotional facial expressions and emotional words

Facial expressions are critical for effective social communication, and as such may be processed by the visual system even when it might be advantageous to ignore them. Previous research has shown that categorising emotional words was impaired when faces of a conflicting valence were simultaneously presented. In the present study, we examined whether emotional word categorisation would also be impaired when faces of the same (negative) valence but different emotional category (either angry, sad or fearful) were simultaneously presented. Behavioural results provided evidence for involuntary processing of basic emotional facial expression category, with slower word categorisation when the face and word categories were incongruent (e.g., angry word and sad face) than congruent (e.g., angry word and angry face). Event-related potentials (ERPs) time-locked to the presentation of the word–face pairs also revealed that emotional category congruency effects were evident from approximately 170 ms after stimulus onset.     

Processing emotional category congruency between emotional facial expressions and emotional words

Facial expressions are critical for effective social communication, and as such may be processed by the visual system even when it might be advantageous to ignore them. Previous research has shown that categorising emotional words was impaired when faces of a conflicting valence were simultaneously presented. In the present study, we examined whether emotional word categorisation would also be impaired when faces of the same (negative) valence but different emotional category (either angry, sad or fearful) were simultaneously presented. Behavioural results provided evidence for involuntary processing of basic emotional facial expression category, with slower word categorisation when the face and word categories were incongruent (e.g., angry word and sad face) than congruent (e.g., angry word and angry face). Event-related potentials (ERPs) time-locked to the presentation of the word-face pairs also revealed that emotional category congruency effects were evident from approximately 170 ms after stimulus onset.     

Proactive and reactive control depends on emotional valence: a Stroop study with emotional expressions and words

We examined proactive and reactive control effects in the context of task-relevant happy, sad, and angry facial expressions on a face-word Stroop task. Participants identified the emotion expressed by a face that contained a congruent or incongruent emotional word (happy/sad/angry). Proactive control effects were measured in terms of the reduction in Stroop interference (difference between incongruent and congruent trials) as a function of previous trial emotion and previous trial congruence. Reactive control effects were measured in terms of the reduction in Stroop interference as a function of current trial emotion and previous trial congruence. Previous trial negative emotions exert greater influence on proactive control than the positive emotion. Sad faces in the previous trial resulted in greater reduction in the Stroop interference for happy faces in the current trial. However, current trial angry faces showed stronger adaptation effects compared to happy faces. Thus, both proactive and reactive control mechanisms are dependent on emotional valence of task-relevant stimuli.     

具身模拟程度和具体性对词语效价表征的影响

采用词性判断和启动加工任务考察具身模拟程度和具体性对词语效价具身表征的影响。结果发现,在具身模拟程度低的空间表征中,只有敏感的负性词出现了效价和远近距离的相容效应;在模拟程度高的趋避表征中,正性词和负性词都与趋避动作产生了相容效应,并且具体词的效价和趋避表征之间的联结强于抽象词。     

The effects of subliminal anger and sadness primes on agency appraisals

Two studies examined whether appraisals can be differentially affected by subliminal anger and sadness primes. Participants from Singapore (Experiment 1) and China (Experiment 2) were exposed to either subliminal angry faces or subliminal sad faces. Supporting appraisal theories of emotions, participants exposed to subliminal angry faces were more likely to appraise negative events as caused by other people and those exposed to subliminal sad faces were more likely to appraise the same events as caused by situational factors. The results provide the first evidence for subliminal emotion-specific cognitive effects. They show that cognitive functions such as appraisals can be affected by subliminal emotional stimuli of the same valence.     

The role of valence and frequency in the emotional Stroop task

People are generally slower to name the color of emotion-laden words than they are to name that of emotionally neutral words. However, an analysis of this emotional Stroop effect (Larsen, Mercer, & Balota, 2006) indicates that the emotion-laden words used are sometimes longer, have lower frequencies, and have smaller orthographic neighborhoods than the emotionally neutral words. This difference in word characteristics raises the possibility that the emotional Stroop effect is partly caused by lexical rather than by emotional aspects of the stimuli-a conclusion supported by the finding that reaction times to name the color of low-frequency words are longer than those for high-frequency words (Burt, 2002). To examine the relative contributions of valence and frequency in color naming, we had 64 participants complete an experiment in which each of these variables was manipulated in a 3 x 2 factorial design; length, orthographic neighborhood density, and arousal were balanced. The data indicate that valence and word frequency interact in contributing to the emotional Stroop effect.