共情是指由感知到他人情绪引起的、自身产生了一种相似的且指向他人的情绪状态，并且个体能够明确意识到他人是这种情绪的来源。疼痛能够稳定地引起共情反应，因而成为最常见的共情诱发场景。大量影像学研究表明，他人疼痛能够引起观察者前脑岛和前、中扣带回等加工自身疼痛的脑区的激活，它们被称为共享表征脑区。影响疼痛共情的因素包括观察者因素、疼痛者因素、二者关系因素和环境因素等， 它们对于共情的调节主要体现为促进和削弱共情的因素分别使共享表征脑区激活增加和减少。但是，当前疼痛共情研究主要集中在脑成像领域，缺乏对于疼痛共情现象学指标的研究。并且，对于各种因素调节共情的作用机制也不清楚。 情绪模仿是指人们在社会互动过程中复制他人的情绪，主要体现为感知到他人情绪后会产生一个相应的情绪表情。先前研究表明情绪模仿可能是共情的先兆，并且调节情绪模仿能够改变人们的情绪感知和情绪理解。因此，情绪模仿可能是疼痛共情的现象学指标。
    为了解决上述问题，本论文共设计了三个研究。研究一使用面部肌电（EMG）技术测量了被试在疼痛共情过程中的情绪模仿现象。研究发现，与观看非痛视频相比，观看疼痛视频会引起被试皱眉肌和颧肌的 EMG活动增加，表明他们在疼痛共情过程中会产生疼痛表情。并且，被试的皱眉活动与其共情特质成正相关。这一结果表明，疼痛共情中的情绪模仿可以作为共情的指标。研究二主要考察情绪模仿在疼痛共情中所起到的因果作用，包括两个实验，分别从实验室和临床考察了不同水平的情绪模仿对于疼痛共情的影响。实验一采用 5（被试间变量：空白对照组、阴性对照组、咬笔抑制组和弹力绷带抑制组以及主动模仿组）× 2（被试内变量：疼痛 vs.非痛）的混合设计，发现主动模仿能够显著提升被试对于他人疼痛的共情，但短时的实验室抑制手段（即，抑制模仿）对共情没有影响。实验二考察了半侧面肌痉挛（HFS）患者由于面部活动紊乱对于疼痛共情能力的影响。面部肌电活动表明，健康对照组能表现出典型的情绪模仿活动，而 HFS 病人大多数时间面部活动处于紊乱状态。与健康对照组被试相比，HFS 患者对于他人疼痛的强度评分都显著降低。表明由于病理状态破坏了情绪模仿能力从而削弱疼痛共情。该研究证明了操控面部模仿能够改变共情水平，说明情绪模仿是共情的作用机制之一。研究三使用高时间分辨率的事件相关电位（ERP）技术考察了人们对真痛和假痛区分的机制及其时间过程，以探究共情差异是如何发生的。该研究使用 2（疼痛 vs.非痛）×3（表情图片、胳膊图片和面孔图片）的被试内设计对比了被试在观看痛和非痛以及真痛（疼痛表情、针扎胳膊）和假痛（针扎在无表情的脸上）时的 ERP 活动。研究发现，观看疼痛图片相对于非痛图片会引起N2 和P3 成分的波幅正向漂移。 但只有当疼痛为真痛时，上述效应才会出现。
    这一结果表明 N2 和 P3 成分可能充当着一个共情过程中的筛查机制，它使得只有真实的、自我相关性高的或意义重大的疼痛线索才能进入进一步共情加工。 总之，上述研究表明疼痛共情过程中存在情绪模仿现象，并且情绪模仿是共情发生发展的作用机制之一。人类大脑对环境中疼痛线索具有筛查能力，这是共情差异产生的原因。讨论部分还从模拟论的理论视角探讨了这些现象，整合了先前研究发现与当前研究结果，指出模拟是共情产生的核心机制。

 Empathy refers to  one's  emotional state  that  is elicited by, and  isomorphic to, another person's emotional state, with a clear self-other distinction. Painful situations are  the most commonly used  triggers for eliciting empathy. Numerous neuroimaging studies have shown that exposure to another’s pain can activate brain regions  that overlap those  involved in the direct experience of pain,  including anterior insula and anterior/middle cingulate cortex. This shared neural activation  is called  a 'shared representation' account. Another line of research has  focused on various  factors that modulate  shared neural representation  during empathy, including observer  and sufferer characteristics, the relationship between the empathy dyad, and the contextual factors. Although  the neural underpinnings of pain empathy  have been extensively studied, the phenomenological observation has been poorly investigated.  It  remains unclear how the modulation of pain empathy elicited by above-mentioned factors occur .
  Emotional mimicry is defined as an automatic tendency to match another one's emotional state in social interaction. By imitating a facial expression of others, we have direct access to the emotion that trigger the facial expression. Some investigators argue that emotional mimicry may precede empathy. There is also solid evidence that inhibiting or facilitating facial mimicry have an impact on the perception and understanding of  other's  emotion. Hence, emotional mimicry may be the phenomenological signature of empathy.
  The present doctorial project involved 3 studies, aiming to  address  these issues. Study 1  examined observers' emotional mimicry  when they watched video clips depicting painful scenes using  facial electromyography (EMG)  recording  method. Enhanced EMG activity of the corrugator supercilii (CS) and zygomaticus major (ZM) muscles was found when observers viewed others in pain, supporting a unique pain expression that is distinct from the expression of basic emotions. CS activity was positively correlated with the empathic concern score  as well as  ZM activity, suggesting  a  facial mimicry  that  mediated empathy for pain. The results demostrated that emotional mimicry can serve as a phenomenological signature of empathy.  
  To  clarify  the causal contribution of emotional mimicry in pain empathy,  two experiments were designed in study 2 in laboratory and clinical settings respectively. Experiment 1 adopted a 5 (group:  blank  control,  negative control, pen-biting inhibition, plastic bandage inhibition, and voluntary imitation) × 2 (condition: pain vs. No-pain) mixed model design. The results showed that, compared to control groups, imitation group showed more sympathy towards others in pain, whereas inhibition manipulations had no influence on empathy reactions. Experiment 2  investigated  the impact of facial mimicry disturbance on empathy for pain in hemifacial spasm (HFS) patients. Facial EMG  indicated that healthy controls (HCs) showed a normal,  typical emotional mimicry activity when facing others in pain, whereas HFS patients showed involuntary facial twitching during most of the experimental trials. Between-group comparisons showed that, relative to HCs, HFS showed a  lower pain-intensity rating to another's pain. Experiment 2 demonstrated that HFS had reduced empathy towards others' pain as a result of facial mimicry disorder. Taken together, study 2  revealed that different levels of mimicry led to different levels of empathy, suggesting  that emotional mimicry is one of the mechanisms underlying pain empathy.  
  To investigate how empathy modulators affect empathy, in study 3, we compared event-related potentials (ERPs) in response to empathy-eliciting visual stimuli with a 2 (condition: pain vs. no-pain) × 3 (stimulus category: facial expressions, face pictures, and arm pictures) within-subject design. Pictures of painful expressions and needle-penetrated arms were treated as true-pain stimuli, while pictures of needle-penetrated faces with neutral expressions were considered false-pain as these situations were not commonly seen  in daily life. The results revealed that, observers allocated more attentional resources to painful events compared to neutral ones, manifested as a positive shift in N2 and P3 components, when true-pain stimuli but not false-pain stimuli  were presented.  Therefore, the complex of N2 and P3 components may serve as a screening mechanism  in brain  through which observers bias their attentions towards more important or more relevant events in surroundings.
   Taken together, the present study verified that pain empathy was accompanied by emotional mimicry, which had an causal role in empathic responses. People holds an attentional screening ability to automatically distinguish pain cues in the environment, which may lead to empathy  difference towards different targets. We also discussed these phenomenon in an integrated theoretical perspective, i.e., 'simulation theory'.