ISSN 0439-755X
CN 11-1911/B

心理学报 ›› 2019, Vol. 51 ›› Issue (4): 462-470.doi: 10.3724/SP.J.1041.2019.00462

• 研究报告 • 上一篇    下一篇


张丹丹1,2, 陈钰1, 敖翔1, 孙国玉3, 刘黎黎3, 侯新琳3, 陈玉明1()   

  1. 1 深圳大学心理与社会学院
    2 深圳市情绪与社会认知科学重点实验室(深圳大学), 深圳 518060
    3 北京大学第一医院儿科, 北京 100034
  • 收稿日期:2018-07-31 发布日期:2019-02-22 出版日期:2019-04-25
  • 通讯作者: 陈玉明
  • 基金资助:
    * 国家自然科学基金(31571120);深圳市基础研究自由探索项目(JCYJ20170302143246158);北京市科委基金(Z161100002616011)

Early preference for positive over negative prosody in neonates: Evidence based on event-related potentials

ZHANG Dandan1,2, CHEN Yu1, AO Xiang1, SUN Guoyu3, LIU Lili3, HOU Xinlin3, CHEN Yuming1()   

  1. 1 College of Psychology and Sociology, Shenzhen University, Shenzhen 518060, China
    2 Shenzhen Key Laboratory of Affective and Social Cognitive Science, Shenzhen University, Shenzhen 518060, China
    3 Department of Pediatrics, Peking University First Hospital, Beijing 100034, China
  • Received:2018-07-31 Online:2019-02-22 Published:2019-04-25
  • Contact: CHEN Yuming


准确解码语音中的情绪信息能让个体更好地适应社会环境, 此能力对新生儿和婴儿尤其重要, 因为人类刚出生时听觉系统远比视觉系统发育得完善。虽然已有研究表明5~7月龄的婴儿能分辨不同情绪种类的语音, 但目前对新生儿的研究还非常少。人类是否在出生时即具有分辨不同种类情绪性语音的能力?新生儿对情绪的加工是否存在正性或负性偏向?本文选用odd-ball范式考察高兴、恐惧、愤怒三种韵律性语音在1~6天龄新生儿大脑中诱发的事件相关电位。实验1直接对比三种情绪性条件, 发现新生儿大脑的额区(F3和F4电极点)可以区分情绪性语音的正负性, 正性(高兴)语音诱发的“失匹配反应”幅度明显大于负性(愤怒和恐惧)语音。实验2采用偏差和标准刺激反转的odd-ball范式, 证实了实验1的结果并非源于三种情绪语音物理属性的差异。本文的结果提示, 新生儿大脑可自动辨别正性与负性情绪语音, 但尚不能将愤怒和恐惧两种负性语音区分开来。更重要的是, 高兴语音比两种负性语音诱发了更大的失匹配反应, 这一结果首次从神经学层面(电生理指标)为新生儿情绪性语音加工的正性偏向提供了证据。

关键词: 新生儿, 正性偏向, 高兴语音, 恐惧语音, 愤怒语音


Our ability to process emotional prosody, that is the emotional tone of a speaker, is fundamental to human communication and adaptive behaviours. Very early in development, vocal emotional cues are more critical than facial expressions in guiding infants' behavior. However, the processing of emotional prosody in the very early days of life is still far from clearly understood. It is unclear whether the discrimination between prosodies with different emotional categories is present at birth. Furthermore, it is unknown whether there is a preferential orientation (negativity bias versus positivity preference) in neonates.

Here, we used event-related potentials (ERPs) to examine the ability of neonates (from 1 to 6 days old) to discriminate different types of emotions conveyed by speech prosody. The experiment was conducted in the neonatal ward of Peking University First Hospital, Beijing, China. Electroencephalogram recording was carried out when the infants were in a state of active sleep. Using an oddball paradigm, the current study investigated the neural correlates underlying automatic processing of emotional voices of happiness, fear and anger in 18 (Experiment 1) and 29 (Experiment 2) sleeping neonates. In Experiment 1, each category of emotional prosody (20%) was separately mixed into emotionally neutral prosody (80%), forming three blocks with different emotions. In Experiment 2, we not only repeated the procedure of Experiment 1, but also reversed the standard and deviation stimuli in the odd-ball task.

Event-related potential data showed that the frontal scalp distribution (F3 and F4) of the neonatal brain could discriminate happy voices from both angry and fearful voices; the mismatch response (MMR) was larger in response to the deviant stimuli of happiness, compared with the deviant stimuli of anger and fear. However, the latter two stimuli, i.e., angry and fearful voices could not be differentiated. The MMR amplitudes at the other four electrodes, i.e., C3, C4, P3, and P4 did not show significant differences across emotional conditions. Note: the MMR is a prototype of the mismatch negativity, i.e. a preattentive component of the auditory ERP that shows a positive (MMR) or negative (MMN) displacement in response to deviant sounds compared to standard sounds in the oddball paradigm.

The neural responses recorded here indicate very early preference for positive over negative stimuli, which is contrary to the ‘negativity bias’ phenomenon established in the affective prosody literature of adult and infant studies. It is suggest that the range-frequency hypothesis could help to interpret the transformation from the ‘positivity preference’ during the first half year of life to the ‘negativity bias’ later in development. The present finding provides the first neuroelectrophysiological evidence for the hypothesis of positivity preference in neonatal participants. In addition, this special discrimination between positive and negative prosody in early life may provide a foundation for later emotion and social cognition development.

Key words: neonate, positivity preference, happy prosody, fearful prosody, angry prosody