The importance of autonomic mimicry in healthy social cognition:
The theoretical framework incorporating evidence from
developmental psychology, social neuroscience, and robotics
Eliska Prochazkova (10629661)
ABC, Brain and Cognition, University of Amsterdam
Cognitive Science Center Amsterdam, University of Amsterdam, Nieuwe Prinsengracht 130, 1018 VZ Amsterdam, the Netherlands, soused@gmail.com, +31-687-379443
Abstract
The development of the social mind has fascinated people since time immemorial. Today, accumulated evidence from developmental psychology, cognitive neuroscience, and robotics has begun to shed light on these processes. In order for a complex system such as social cognition to develop, there is need for a so called ‘physical embodiment’, and, more importantly, ‘mimicry’. But what the exact role mimicry plays in healthy social development is not fully understood. The current review is the first to combine theories from robotics and the social sciences to support the notion that spontaneous mimicry is essential for the development of healthy social cognition. In this review, I will argue for three main points: A) I suggested that spontaneous mimicry provides a physical-cognitive link during an organism’s development that facilitates implicit emotional communication between species. To support this view, I will provide evidence from the social sciences about a strong physiological linkage that exist between people and describe how this physiological association translates into emotional attachment and emotional contagion. B) I will further propose that behavioral mimicry represents a physiological precursor that is necessary for the development of higher cognitive abilities. To provide a theory for this argument, I will adapt a theoretical model by Asada (2015) from computational science and robotics and apply it to evidence from the social and neural sciences. Here, I will describe how simple autonomous mimicry can give rise to complex social cognition and behavior. By doing so, the current review will be the first to combine theories from robotics and the social sciences in order to support the notion that spontaneous mimicry is necessary for healthy social development. Finally, I will propose C) that measures of mimicry may represents potential physiological marker of healthy social cognition.
1. Introduction
1.1. The Evolution of Social Cognition
In environments where many factors change quickly over ime, brains provide an advantage to survival by allowing organisms to extract patterns of information that aid predictions (Adolphs, 2001). Humans, as well as many other social animals, live in groups. On one hand, groups can offer better prospects for survival by cooperation and by communication but on the other hand, groups can generate competitions that pose a risk of exploitation by others. This is why compared to the physical environment the social environment is more abstract and highly unpredictable. Still, humans are often able to intuit other people’s feelings and predict others actions and interpret their meaning even before they take place(Kim, 2012). They do so seamlessly, without effort, and often without conscious awareness (Heider & Simmel, 1944). This remarkable social capacity has been claimed to be the key characteristic of many of humanity’s’ modern achievements (Wertsch, 1998; Kin et al., 2003; Cheney et.al, 2008). The ascertainment that other people have needs and desires has allowed humans to cooperate, communicate, learn, share and trade with each other, which separated us from any other species living on this planet. In the past decade the field of artificial intelligence and robotics has attempted to create a mechanism that would exhibit social intelligence and behavior similar to humans (Scassellati, 2007; Asada, et al., 2001, Asada, et al., 2012, Meltzoff, A.N., 2007; Miwa, et al. 2004). However, despite the considerable progress in computational knowledge and technological development, the attempt to create artificial social intelligence has so far failed. Nowadays, after many years of ontological and methodological reductionism1, the fields of artificial intelligence and social neuroscience have finally found common ground. It has been
11. Ontological reductionism: a belief that the whole of reality consists of a minimal number of parts.
agreed that in order for a complex system such as social cognition to develop, there is need for a so called ‘physical embodiment’, and more importantly social interaction. This is because from robotics, we know that in order for a simple design (predisposition) to become more complex, there is a need for information structuring through interaction with other social agents. From this standpoint it seems reasonable to say, that in order for social cognition to evolve there is a need for basic physical structure (the human body and the brain) to interact with other social agents (people) during social development. But where does the development of social mind begin? It has been agreed that in order for social cognition to develop, there is need for a so called ‘mimicry’. But what is the role of mimicry and how important it really is in healthy social development is not fully understood. Both robotics and social neuroscience have a lot to offer for the understanding of social mind. Although, both take rather different approaches to study these processes, both disciplines generally compliment and contribute each other’s views. Combining artificial intelligence theories with social neuroscience findings, here I will argue (A) that social cognition is a direct result of physical-cognitive interaction during an organism’s development that is facilitated by spontaneous mimicry. Specifically, (B) I will propose that mimicry provides perception– behavior link that forms implicit emotional communication between species and thus represents an evolutionary precursor of human´s higher cognitive abilities. By doing so, current review will be the first to provide comprehensive theoretical framework supporting the notion that spontaneous mimicry is essential for healthy social development and represents potential physiological marker of healthy social cognition.
1.1. 1. Emotional Contagion & Mimicry
Recent theories of social behavior in neuroscience and robotics propose that facets of social cognition expand from simple unconscious process (emotional contagion)to higher cognitive conscious processes (perspective thinking) (ibid). It has been further proposed that human cognition is similar to computer in the way its conscious levels exist on the physical level (body & brain) but not vice versa (the mind/software). According to Asada (2015) the development of
social cognition began with ‘emotional contagion’ which is both unconscious and requires body
contagion is an experiment where one mouse receives an electrical shock accompanied by a tone while being observed by another mouse. Eventually, the mouse which has been merely observing the scene also freezes in response to the tone, even though the mouse itself never experienced the sensation of an electrical shock. A similar example for humans would be observing a spokesman on a stage that is experiencing strong feelings of anxiety, and suddenly having a very uncomfortable feeling of anxietyas well. The mechanism of emotional contagion in both humans and animals is thought to result from mimicry or synchronization (Goldman, 2006; Stel & van Knippenberg, 2008; Anders, et al. 2011). People mimic all types of behaviors ranging from automatic mimicry of partner’s unconscious physiological responses to spontaneous synchronization of facial expressions, vocalizations, postures and movements controlled by voluntary muscles (ref review paper).
According to Meerloo (1957), mimicry may be defined as an archaic communication and rudimentary remnants of animal signals originally used as warning signs for fellow creatures to flee or to hide. It other words, automatic mimicry transmits a state of alarm and relays on ontogenetically older neural networks (i.e.; brain-stem and Limbic system), that are implemented in automatic and fast emotional responses. Hatfield and colleagues (1994) argued that people possess neural systems that pick up subtle social cues from the environment. By attending to streams of moment-to-moment expressions, ‘‘people and animals can feel themselves into the emotional landscapes inhabited by their partners’’. With regards to Asda’s (2015) model, during the development of social cognition ‘ontogeny recapitulates phylogeny’ where the increase in social behavior proficiency is observable on both the evolutionary as well as the developmental level (Fig. 1).The story of social cognition thus follows this progressive slope, where exchange of socially relevant information passes from the ‘physical-motor’ (unconscious) to ‘emotional-mental’ (conscious) level. The affective meaning of these social cues is first expressed within organisms’ bodies and brains before they can become consciously realized. Via neural feedback from own muscles to the brain, spontaneous mimicry of another’s bodily expressions allows individual to feel other’s emotions directly. For this reason spontaneous mimicry is believed to be a primitive form of empathy as it enables the direct transfer of emotional information from one organism to another (Preston & Waal, 2002).
1.1.2. Emotional Empathy & Cognitive Empathy
Today, developmental theorists however make clear distinctions between the spontaneous mimicry which is the primitive form of empathy and the more cognitive, ‘‘sophisticated’’ processes such as cognitive empathy and theory of mind (ToM), (Hatfield, Cacioppo, & Rapson, 1994).Over the course of evolutionary history the ability to pick up subtle social signals by spontaneous mimicry has matured to the level that humans can consciously speculate about other peoples’ mental processes. This type of social cognition however requires perspective taking. In other words, even though both empathy and emotional contagion induce representations of another individual’s emotional states, unlike in emotional contagion, when people are empathizing they are generally aware of a separation between self and other. This in turn promotes metacognition-the ability to think about one’s own thinking-which is a mental capacity that is not known to exist in any other species (Asada, 2015). Metacognition is important for social integrity as it enables individuals to regulate their own emotions. This is because self/other separation allows for further distinctions between own basic/animalistic goals (implicit emotional urges) and higher order goals (rational/cultural rules) learned through past social interactions. Even though some basic forms of perspective taking have been reported in neonates (Geangu et al. 2010; Clark, 1987), robust evidence demonstrates that theory of mind does not fully develop in humans before the age of three (Happé, 1995; Frye, Zelazo, & Palfai, 1995).This supports the view that perspective thinking is not a purely innate capacity but is likely to result from the maturing brain interacting with its dynamic social environment.
While emotional contagion is fast, automatic and unconscious and is shared by most species, both emotional empathy (EE) and cognitive empathy (CE) reacquire representation of conscious self-awareness and may only occur between primates and perhaps other animals living in social groups such as dolphins, elephants and wolfs (ibid). EE stands between the basic emotional contagion and CE because it allows individuals to form mental representations of other’s feelings by embodied simulation, while still suiting self/other distinction. In CE the self/other separation
envy where an individual thinks about the self in the third person perspective (e.g. Why am I sad when the other is happy?). Following this developmental trend, it has been proposed that during a child’s development the steppingstones of CE begin with basic emotional contagion. In the first section of this review, I will thus focus on evidence from social sciences showing that a strong emotional linkage exists between humans. I will further illustrate how this emotional contagion translates to human physiology and psycho-emotional attachment.
1.2. Evidence for Emotional Contagion in Humans
Mothers and children share a deep physiological connection. This type of physiological linkage is shared by most mammals and represents the earliest form of emotional contagion that occurs between mother and child already before birth. In 2010, a team of doctors at Sydney hospital have witnessed nothing less but miraculous power of this strong physiological connection.
‘‘Kate Ogg has put her prematurely-born son on her chest, whispering words of comfort in soothing words. Doctors told her that he had died, and she was prepared to say her last good bye. Then something unexpected happened, little Jamie moved. She cried for help but doctors refused to come back, believing it was only a reflex; they didn’t want to keep Kate in denial. After two hours of skin-to-skin contact Jammie opened his eyes. Jammie is 5 years old today, he is healthy and lives with his family and twin sister in Sydney’’ (Crane, 2015, March 13).
Current literature agrees that what saved little Jammie’s life was a physiological synchrony between him and his mother. Accumulating evidence now shows that skin-to-skin contact between mother and infant can significantly reduce neonatal mortality (Lucas, Stoll & Bale, 2003). Researchers clarify that this is because when newborns are put into direct contact with the skin of their mothers, they synchronize their temperature, breath, and heart beat with that of their caregiver. Because infants breathe irregularly and have a faster heart rate than adults, by feeling mothers’ heart palpations and breath’s movements, they automatically mimic their mothers’ cardiovascular responses and reach homeostasis faster.
Interestingly, the physiological synchrony can also occur even without any direct physical contact. For instance, during close interactions both animals and humans tend to synchronize their heart rate, respiration rate, hormonal expression, temperature and pupil-sizes (Feldman et al., 2011; Luijk et al., 2010; Papp, Pendry, & Adam 2009; Kret, Fischer, & De Dreu, 2015, Cooper et al. 2014). In psychology, this phenomenon is also called a physiological linkagewhich refers to any association pattern between physiologies of interacting partners. For instance, research shows that mothers and infants mimic each others’ heart rhythms and breathing patterns during close interactions (Porter, 2003). Feldman et al. (2011) collected cardiac outputs from mothers and their 3-month-old infants who were interacting face to face. This study showed that mother and infant heart rhythms synchronized within lags of less than 1 s. In addition, the synchronization of heart rate significantly increased when the mother and child mimicked each others’ smiles and showed vocal synchrony. A growing body of evidence suggests that physiological mimicry is positively correlated to mother-infant attachment. For this reason it has
been proposed that physiological mimicry plays a role in the development of emotional ties between mother and child and promotes infant’s healthy social development (Feldman, 2012). Although previous research demonstrates that synchrony in autonomous systems between mother and child is generally a positive marker, physiological linkage can also have a negative impact on infant social behavior (Hill-Soderlund et al., 2008; Luijk et al., 2010). This is because attachment quality is also associated with dysregulation of the hypothalamic-piuitary-adrenal (HPA) axis or sympathetic nervous system (SNS) response. The ANS regulates unconscious (involuntary) bodily functions such as heart rate, respiration rate and pupillary response. Numerous studies have reported that maternal stress can affect the development of a child’s HPA axis (Laurent, Ablow & Measelle, 2011; Kivlighan, 2006). Dysregulation of an infant’s HPA axis is expressed by an increased vulnerability to stress and anxious personality trait. Interestingly, recent epigenetic findings suggest that maternal stress can be even transmitted to the next generation, which indicates that prenatal maternal stress can actually influence the offspring’s genetic information that is then passed on the new generations (Weaver et al., 2005; Oberlander, et al. 2008).
In support of emotion contagion theories, a recent human study found that physiological linkage between mother and child allows for the direct transfer of affective experience between the pair (Laurent, Ablow & Measelle, 2012). In this study, cortisol levels were collected from a large sample of 88 mother-infant pairs during the Strange Situation task. The Strange Situation task is a psychological method testing the attachment style between caregiver and child. Results showed that mothers who experienced strong negative emotions in this task also showed synchronous increases in cortisol in their child. On the other hand, caregivers of ‘disorganized’ children who displayed atypical behaviors, in the sense that they were not securely attached to their mothers compared to normally developing groups, showed anti-phase synchrony where infant’s cortisol levels decreased and mothers’ cortisol levels increased. Further studies suggest that the level and type of attachment style between the pair in turn influences the child’s personality traits and the way he/she interacts with others (Buss et al., 2015). This finding supports the view that the physiological state of a mother can directly affect the physiological profile of her child, which is
also translated in the psycho-emotional interaction between the pair. However, the physiological linkage is only beneficial if the mother is psychological health or has normal HPA activity. Saxbe’s and colleagues (2014) investigated whether emotional contagion also occurs later in life and ‘spreads’ within other members of families. In this study, cortisol levels were collected in a sample of adolescents after they engaged in family conflict with both parents. The results showed that cortisol levels were positively associated between family members, supporting the hypothesis that people synchronize their physiological stress levels during emotional social interactions. In addition, the analysis showed that fathers’ cortisol levels predicted adolescents’ cortisol levels, while adolescents’ cortisol predicted mothers’ cortisol, and mother’s cortisol predicted fathers’ cortisol level. This finding suggests that different individuals may exert various degrees of physiological influence on each other. In case of this particular study, father’s stress levels uniquely contributed to children’s stress levels, youths to their mothers’, and wives to husbands’.
1.2.1. Psychological Moderators of Physiological Mimicry
One open and interesting question is whether physiological dependence extends beyond families and holds across cultures. For instance it could be argued that the physiological dynamics may vary according to different types of emotional dynamics within each family unit, (i.e. the level of attachment between individual members of the family). In support of this view, Papp, Pendry, and Adam (2009) observed that the level of physiological alignment is moderated by the time family members spend together, where the cortisol synchrony was strengthened among mother-adolescent dyads who spent more time together, and in families who reported higher levels of shared activities and parental monitoring. Furthermore, apart from attachment level, research has shown that the physiological linkage can also be moderated by psychological factors. Laurent and Powers (2007) studied the synchrony of cortisol release within mother and father couples. In this study, cortisol was collected from couples 7 times per day over the course of 2 typical days. As expected, couples demonstrated physiological synchrony. However, this association was
moderated by psychological characteristics such as feelings of closeness/loneliness at the time of cortisol collection.
Above findings indicate that the strength of physiological linkage is associated to level of attachment between people and is moderated by psycho-emotional factors. However, apart from families, research showed that physiological alignment also occurs between adults who are not necessary directly related but are familiar to each other. In a study by Butler (2015),couples of familiar women watched a disturbing film and then were asked to discuss it. In each dyad, one woman was instructed to suppress her emotions, reappraise them positively, or converse naturally. This study reported that all pairs showed physiological synchrony (measured by heart rate and skin conductance). The women who reappraised the movie positively also influenced their partners to feel more positively and produced in-phase physiological synchrony with their partner; the suppression of emotions produced anti-phase synchrony, and did not have significant impact on partner’s reappraisal. In the control pairs who were not instructed to positively re-apprise the content of the movie or suppress their emotions, the synchrony alternated between in-phase and anti-in-phase. The change between in-in-phase and anti-in-phase synchrony is likely to be result of mixed emotions that are transmitted/experienced during emotional conversation. This study again suggests that emotional contagion is related to unconscious physiological linkage between partners that can lead to conscious reappraisal of their partner.
In summary, the here reviewed literature demonstrates that at the beginning of life people align their physiology with their caregivers; this in turn has an impact on their affective behavior, relationships and mental health later in life. Current findings further suggest that emotional contagion and social bonds operate both on the cognitive and physiological level. The magnitude of physiological linkage is strengthened by the amount of time people spend together and seems to be associated with the level of attachment as well as with other psychological factors. However, since most of these studies rely on correlations, it is impossible to determine whether the level of synchrony is a result or a consequence of these factors. This poses a major question for the theories of emotional contagion. On one hand, it could be argued that physiological linkage is a form of emotional contagion that allows sharing affective experiences between
species. On the other hand, it could be contra-argued that social cognition precedes physiological synchronization. In other words, people first psychologically align with each other before they show physiological synchrony. From this standpoint, synchrony in humans physiology could be seen as a result of shared cognitive experience rather then vice versa. Although, the truth is likely to lay in both of these statements, nevertheless, the crucial question is to what extend one contributes to another. According to de Waal (2008), because emotional contagion develops earlier than cognitive empathy, cognitive empathy is likely to be built on top of emotional contagion. Or even more precisely, emotional contagion is the identical processes with added functions.
There are two possible methods to tackle this empirical question. The first takes the retrospective approach, for example by the use of longitudinal studies we could measure to what extend the level of physiological linkage contributes to child´s later social abilities. The second option takes the prospective approach. For instance by the use of robotics we could test whether by developing robots that can detect subtle social signals and exhibit mimicry, the robots start to develop social cognition and behavior similar to humans. Nevertheless, the first step before we can implement either of these approaches is to understand the mechanisms of physiological linkage. This is because if we know how physiological linkage works, we can than identify its function and understand to what extend it effect human health and social behavior.
1.3. The Mechanism of Emotional Contagion: Motor Mimicry & Autonomous
Mimicry
Current literature agrees that the key mechanism of emotional contagion is spontaneous mimicry. What is important to note is that affective experience is expressed through three main compartments: cognitive, physical and behavioral (Kret, 2015). Thus, emotional contagion is likely to take place between all these three compartments (see Fig. 2). For instance, during social interactions both person A and person B are experiencing feelings and emotions (explicit, level 1), these emotions are expressed within their physiological arousal (implicit, level 2) and body
expressions which are frequently implicit but can be also voluntarily controlled (implicit/explicit, level 3).
People can influence each other’s feelings consciously, for instance verbally. An example could be when a verbal critique from person A affects person B’s feelings, which increases person’s B physiological arousal which results in a change in his facial expression and body posture (top-down route).Yet, during early development, emotional contagion is rather likely to take the bottom-uproute (Asada, 2015). This is because when infants are born their verbal and motoric abilities are still very limited and their communication relay mainly on subtle social cues from the environment. During this process, person A implicitly picks up on subtle social signals from person B’s face and body; this in turn impacts on person’s A own physiology and cognition. Embodiment theories propose that this type of emotional contagion is a result of overlapping neural networks that are activated during both motor action and the observation of motor action (see Zaki, 2014). As a result, perceptual activity spreads to behavioral representations, which in turn increases the probability of imitating that same behavior (Zaki, 2014). It is believed that during social development, by mimicking caregivers’ nonverbal expressions (through afferent feedback from own bodily expressions to the brain) a child learns to recognize different expressions of emotions and understand other people’s mental states. From this perspective, the development of social cognition is a direct result of physical-cognitive interaction during child’s development with its social surrounding. This supports current reviews argument that mimicry works as an evolutionary precursor that enables direct transfer of emotional information from one organism to another and thus precedes development of higher social cognition (i.e. Emotional Empathy/Cognitive Empathy). Spontaneous mimicry thus could be perceived as an innate characteristic that allows a child to resonate with or tune into other people’s internal states. If this is the case, measure of spontaneous mimicry could be potentially used as physiological markers of social disabilities in young children.
In the next section, I will discuss various types of spontaneous mimicry providing evidence about a link between mimicry and social cognition development. Due to limited scope of this thesis I will focus only on mimicry of facial displays. Distinction will be made between motor
mimicry(Fig. 2. Level 3)- controlled by facial muscles which is partly implicit but as well can be
consciously controlled, and autonomous mimicry (Fig. 2. Level 2)- phylogeneticaly older signaling system that is fully unconscious and autonomic as it relays on changes in the ANS.
Fig. 2: Schematic representation of emotion processing during social interactions, adapted from Fig.1, Kret (2015).
Level 1: Explicit
Level 2: Implicit
1.3.1 Mimicry of Voluntary Expressions: Motor mimicry
One physical characteristic that distinguishes humans from any other species is the high level of expressiveness of human faces. Robert Sapolsky (2002), the evolutionary biologist and neuroendocronologists from Stanford University, once said that in a way ‘‘compared to other animals people are basically see-through’’. Emotions people experience are often spontaneously displayed in facial expressions without conscious awareness or voluntary intention. The stronger the emotion is, more difficult it generally becomes to control and sometimes the affect can be so strong that it actually brings humans to tears. Yet, from the evolutionary perspective, considering that we are living in the world of ‘selfish genes’ where recourses are scares and competition high, and where expression of weakness could attract exploitation by predators, this characteristic does not seem to be very adaptive. On the contrary, it could be argued that the reduced emotional expressiveness that masks organism’s intentions provides survival advantage. Some researchers believe that at the price of private thoughts humans have evolved communicative faces between which information can flow freely. It is this quality that is by many believed to be the stepping-stone of human verbal communication and development of complex social cognition (Meerloo, 1957; Hatfield, Cacioppo, & Rapson, 1994).
1.3.2. Facial Mimicry
When infants are born they have underdeveloped sensory organs and limited motor abilities. For instance, neonates’ vision is blurry (visual acuity is sharpest at the edges- rather than at the center of the visual field), (Dobson & Teller, 1978; Ayres & Robbins, 2005). Still, people generate, attend to and mimic facial expressions early after the birth. Apart from humans, chimpanzees also show spontaneous mimicry (Ross, Menzler & Zimmermann, 2008) which indicates that spontaneous mimicry is an evolutionary adaptive mechanism in non-human primates too .Still, even though humans and chimpanzees have strikingly similar underlying mimetic musculature in their faces, humans use greater variety of facial expressions and also
detect facial movements with much more speed and precision (Burrows, et al, 2006, Vick, et al, 2007).
The mimicry of people’s facial expressions is very fast and delicate. Psychophysiological research has found that facial mimicry is at times almost instantaneous as people seem to be able to track the most subtle moment-to-moment changes in their partner’s faces. Additionally, their own facial muscles tend to reflect at least some of partner’s muscle changes (Hatfield, Cacioppo, & Rapson, 1994). Research shows that people can synchronize their movements within 21 milliseconds, which is almost twice beyond the threshold of people’s voluntary movement abilities (Condon & Ogston, 1966). Furthermore, these microexpressions are so subtle that they sometimes cannot be detected by the human eye and can only be measured by Electromyograph (EMG) systems sensitive to micromovements of facial muscles. In addition, people display facial mimicry even when facial stimuli are presented to participants subliminally. Dimberg and colleagues (2000) investigated activity of facial muscles, which move during smile and during a frown. The authors argued that if distinct emotions can be automatically elicited by subliminal cues, then unconscious exposure to happy or sad faces should differentially activate these muscles. In line with this hypothesis, the results showed that participant’s muscle responses were implicitly elicited and corresponded to muscle movements that are generated during happy and sad facial expressions and that even though participants’ reported not being aware of the stimuli presentation nor their own muscle movements. This supports the theory that facial mimicry is an automatic and innate reflex-like mechanism that is activated in response to people’s emotional states.
A landmark study by Meltzoff and Moore (1983) showed that very young infants ranging between 1 hour and 3 days of life tend to already imitate the behavior of strangers. The authors proposed that imitation of a caregiver’s facial displays is important for the communication of affective states. In support of this theory, Oberman et al. (2007) revealed that blocking participant’s facial muscles impairs the ability to recognize the expressed emotions in target’s facial expressions. Evidence exists that people recognize emotions from other people’s faces by experiencing changes in their own physiological state (Ekman, Levenson & Friesen, 1983). This
affective shift is proclaimed to be the result of spontaneous mimicry. In support of this theory, Ekmanand colleagues (1983) showed that the mimicry of facial expressions can generate ANS arousal that is comparable to experienced emotion itself. In their study, participants were asked to produce six basic emotions (disgust, surprise, anger, fear, sadness, and happiness). They were requested to either relive times when they had experienced such emotions or arrange their facial muscles according to these emotions. The authors found that both the act of reliving emotional experiences or production of facial expressions produced the same ANS response. This suggests that facial expressions can generate autonomic nervous system activity which serves as physiological feedback to inform individuals about their own/another’s emotional experience. Current findings demonstrate that despite the fact that people are generally a) not consciously aware of subtle changes in partner’s facial characteristics and b) do not voluntary react to them, in accordance to emotion contagion theory, people are still able to detect these subtle signals -as demonstrated by mimicry -and by doing so enhance their understanding of partner’s emotional expressions.
1.3.3. Eye gaze
Apart from facial expressions, one of the earliest and most salient types of automatic mimicry is eye-gaze. During close interactions, both infants and adults focus on their interactive partner’s eyes, grasp emotion signals from the eye whites and pupils, and follow gaze (Baron-Cohen, 1997; Driver et al. 1999, Hood et al., 2002). Research shows that the eye region captures more attention than other areas of the face in adults as well as infants (Adolps et al., 2005; Heith, Bergman, & Moore, 1997). Neonates prefer to look into the eyes of their caregiver compared to other people, and also prefer to look at people who are looking at them rather than to the eyes of others with averted gaze. The explanation of why humans are implicitly drawn towards the eye region is not fully understood. Yet what is known is that direct eye-contact increases amygdala signaling and is associated with elevated arousal levels (Spezio et al., 2007, Kawashima et al., 1999). Furthermore, recent evidence shows that mother-infant physiological linkage increases when the pair looks into each other’s eyes (Feldman et al., 2011). The same effect was also
found in romantic couples, where physiological synchrony was strongest when lovers looked directly into each other’s eyes(Helm, Sbarra, & Ferrer, 2012). Another study by Heyes et al. (2005) demonstrated that eye gaze increases motor mimicry in humans. In this study, direct gaze increased mimicry of hand actions by 13 ms compared to averted gaze. These results suggest that the speed of motor mimicry can be affected by social engagement cues. It has been argued that direct eye-gaze is an innate biological system that automatically stimulates arousal levels in the observer; this in turn leads to faster processing of the social situation and social learning. According to the mind-eye hypothesis, what people are looking at is usually the same as what they are thinking about. By following gaze, people can track the path of partner’s intentions, get insights into his/her emotions, and share experiences (Just & Carpenter, 1976; Duchowski, 2007). Complementary research shows that high frequency and long duration of eye contact is positively correlated with trust, sexual attraction, and openness. On the other hand, people who do not look into other people’s eyes are considered to be cold, distant and indifferent (Kleinke,1986; Farroni, et al., 2002). For this reason, direct eye contact has been claimed to be an innate adaptive behavioral trait that is essential for the development of ToM and attachment behavior. In support of this theory, direct eye-contact avoidance is related to social pathologies. 12For example, people with autistic spectrum disorders (ASD) as well as people with social anxieties show reduced direct eye contact (Farroni, et al., 2002). Accordingly, people with ASD also show problems with social bonding, ToM and empathic abilities. Based on this evidence, eye tracking has been suggested to be a potential neurocognitive tool that could be used to detect early onset of ASD (Boraston & Blakemore, 2007).
However, the problem with eye gaze and facial mimicry research is that they are facilitated by facial muscles. Although previous literature suggests that spontaneous facial mimicry and eye gaze are largely implicit and unconscious processes (i.e. are related to changes in ANS and amygdala signaling), both eye gaze and facial expressions can still be voluntarily controlled (see Fig 2, Level 3). For instance, Hess and Fischer (2013) showed that facial mimicry can also be used as a social regulator, suggesting it is partly conscious. Additionally, understanding facial expressions is bound to specific social contexts, and the level of eye-contact and facial mimicry during social interaction is modulated by cultural differences and the level of social engagement
during childhood development (Farroni, et al., 2002). These factors may however produce major scientific confounds when considering direct eye gaze and facial mimicry as an innate characteristic which could be potentially used as physiological markers of social disabilities in young children (Farroni, et al., 2002).Recently, researchers have started to argue for a broader exploration of emotional signals from other sources beyond the face muscles. Specifically, synchrony of pupil-diameter, blushing, and contagious crying has been suggested to provide alternative autonomous pathways of emotional contagion (Kret, 2015). According to Kret (2015),these signals are directly related to changes in autonomic nervous system and much harder to control than facial muscles, yet because they are still visible to the observer, they might add to the perceived intensity of facial expression or even overrule the emotional signals the facial muscles try to reveal (ibid).
1. 3. 4. Mimicry of Autonomous signals: Autonomous mimicry
1.3.5. Pupil synchronization
The recent research of Kret, Tomonaga, & Matsuzawa, (2014)revealed that apart from direction of eye gaze primates also synchronize their pupil sizes. Pupil change is directly related to activity of ANS. While pupil dilation is a physiological marker of the sympathetic ‘flight or fight response’, the pupil constriction is part of the parasympathetic ‘rest and digest response’. Research shows that human pupils dilate when people are excited, aroused, when they experience high cognitive load or when they are trying to deceive others (Gilzenrat et al., 2010; Alnæs et al., 2014). What makes pupils especially interesting is that in contrast to other physiological expressions of autonomic arousal such as GSR (Galvanic Skin Response), cardiovascular changes (heart rate and heart rate variability), and electrical activity measured by Electroencephalography (EEG), pupil-size changes are visible to others. Hess (1965) was the first to argue that pupils fulfill a social function as they constitute an implicit form of communication between people. In one of his first experiments, he presented groups of men with pictures of young woman; the pictures were completely identical except for one small difference. In one of these pictures the woman had larger pupils while in the other her pupils were made to
look relatively small. Participants, unaware of this manipulation, perceived the woman with large pupils as friendlier, softer and more trustworthy. This evidence was the first to show that pupillary changes are implicitly picked up by observers and can actually influence their judgments. Later research has replicated these findings showing that people judge partners with large pupils as more positive, and partners with small pupils as cold and distant (Amemiya & Ohtomo, 2012; Demos, et al. 2008). Kret, et al. (2014) suggest that this positive association is formed through pupil-synchronization. In their study, which included both humans and chimpanzees, they found that apart from a social signaling function, pupil sizes synchronize between partners of the same species during social interactions. In a second human study, a link with behaviour was observed. Here Kret et al. (in press) found that if participants synchronized their pupil-size with the dilating pupils of their virtual partner they established greater trust in their partner. This remarkable observation is supportive of the view that pupil-mimicry may constitute neurocognitive precursors from which social judgments and behaviours are formed. Yet, whether pupil-synchronization always increases trust or actually allows individuals to understand other people’s emotional states and emotions is not yet known.
1.3.6. Contagious Blushing
Another form of autonomous mimicry is blushing. When people experience strong affect, their skin gets perfused with oxygenated blood. Such a change is directly observable by increased redness of the face. People associate redness in face with health, anger or aggression. However, blushing may also signal shyness or embarrassment (Kret, 2015). It has been argued that mimicry of face redness may have evolved as a passive behavioral defense. Meerloo (1957) proposed that redness could a twofold defensive strategy that serves as a warning or a disguise in response to an external event. The idea is that when in danger, the victim overcomes the anxiety through physiologically playing the role of the aggressor. In other words blushing is the involuntary imitation of the feared person. This in turn influences the behavior of aggressor. In support of this theory, redness of face has been shown to affect the observers’ social judgments. Higher level of redness has been associated with physical dominance (Leary, et al. 1992, Dijk, Ketelaar & de Jong, 2011). Again, as in previous examples of pupil size, because blushing is an
autonomous response that is difficult to control, and is highly contagious, it may serve a role of social signaling providing an implicit form of communication between people.
1.3.7. Contagious Crying
Most people who have visited a newborn ward have noticed that crying is contagious. Martin and Clark (1987) played audio recordings to newborns. They showed that one-day-old babies were more likely to mimic crying when they heard a recording of another newborn crying. However, when they heard their own cries, or heard much older infant crying, mimicry did not occur. This study supported the view that crying mimicry is not merely the result of elevated noise but is a contagion mechanism. Additionally, neonates seem to make a self/other distinction and mimic the behavior of similar others. Geangu et al. (2010)tested infants at 1, 3, 6, and 9 months of age in response to different types of cries. Their emotional reactions were recorded in terms of vocal (presence of vocal distress, latency and intensity) and facial expressions (anger and sadness). Results showed that all age categories mimicked crying, whereas the distress was highest in response to cries of pain. The ability to distinguish between different types of crying that is accompanied by similar response of distress has been claimed to be one of the first signs of empathy in humans.
1.4. The neural mechanisms of mimicry
In the previous section, I reviewed literature suggesting that both children and adults automatically mimic facial expressions as well as autonomous signals expressed in their partner’s face. It is believed that spontaneous mimicry during face-to-face communication relies on a biologically old signaling system that occurs automatically and largely outside awareness. The study of these phylogenetically old signaling systems is of crucial importance because human nonverbal communication often makes use of these elementary biological signaling devices.
With regard to Asada’s (2015) work describing the development of artificial empathy, emotional contagion from one individual to another is an automatic (unconscious) bottom-up process where variety of implicit environmental cues are automatically picked up by an observer’s senses. These cues are then expressed in the observer’s bodily sensation before they become consciously realized. To elucidate the underlying mechanisms, I will review the neural correlates of social cognition. Here the main distinction will be made between the basic emotional contagion (facilitated by motor mimicry and autonomous mimicry) and emotional empathy (EE)/cognitive empathy (CE) implemented by higher cognitive networks.
1. 4.1. Basic Emotional contagion
1.4.2 Autonomous mimicry
According to Asada (2015), basic emotional contagion may be defined as an archaic communication signaling system originally used for warning signals. In other words, automatic mimicry transmits a state of alarm and relies on ontogenetically ancient neural networks (i.e.; brain-stem and limbic system) that allow fast automatic emotional responses (LeDoux, 1994).Here I provide possible explanation of how autonomic mimicry transfers arousal between human bodies and brains during basic emotional contagion.
To illustrate an example, we can imagine two people having a face-to-face interaction. The sender is experiencing a strong feeling of anxiety, while the receiver is observing his partner’s facial characteristics. From previous literature we know that anxiety is related to activation of sympathetic nervous system (SNS).
The sender:
The SNS activation in the sender (Fig.2, A) mediates the hormonal sympathetic stress response and also regulates the body's unconscious actions such as heart rate, digestion, and breathing. Sympathetic nerves are located near the brain stem and the stress response is initiated by
hypothalamus-pituitary-adrenal (HPA) axis activation. When the HPA axis is activated, the adrenal medulla secretes acetylcholine, increasing adrenaline (epinephrine) and noradrenaline (norepinephrine) release. This fuels the cardiovascular system and as a result heart rate and respiration rate increase and digestion slows down. Importantly, apart from inner organs, the SNS is directly connected to sensory channels on the surface of the body where information is transmitted in a bidirectional flow. The outer parts of the body of this system are pupils, skin, and muscles. This is why people’s pupils dilate, cheeks blush, hands sweat, and why people make facial/body expressions when they experience high levels of arousal (Barboi,2013).
The receiver:
According to De Waal (2008) behavioral mimicry is an automatic, matched motor response, based on a perception– behavior link. This hypothesis assumes that merely perceiving a specific nonverbal display automatically entertains the same expression in the perceiver’s brain. During close interactions, sensory inputs converge in the amygdala either directly from the sensory thalamus or from the various sensory cortices. The amygdala is a brain region located in the deep layers of limbic part of the brain and is involved in detection of biologically relevant cues. The amygdala is involved in the recognition of emotion in people’s faces and its’ main function is to link perceptual representations to cognition and behavior on the basis of the social value of the stimuliAdolphs, 2001; Amemiya & Ohtomo, 2012). The HPA axis activity is directly associated with increased amygdala signaling as it is the key node used in fear conditioning (Phillips & LeDoux,1992). Thus, when receivers perceive signals of partner’s increased arousal, the amygdala picks up on these social cues and projects this information to the brain stem initiating a stress response in the receiver. As a consequence, the receiver experiences a reflection of the sender’s arousal in his or her own body, and both partners’ synchronize their physiological responses (sweating, blushing, pupil dilation) - they converge emotionally.
The synchrony of autonomous signals (heart rate, skin conductance, hormonal expression) is not sufficient for emotional contagion to occur. This is because emotions have two fundamental dimensions, the arousal level (intensity) and the valence of emotion (interpretation) (Lane, Chua, & Dolan,1999).Research demonstrates that increases in arousal can be interpreted as both positive and negative depending on the environmental context (Dolcos, LaBar & Cabeza, 2004). This however requires the incorporation of several visual and cognitive processes at the same time. Apart from synchrony in autonomous arousal, another mechanism that has been proposed to play an essential role in emotional contagion is ‘embodied simulation’ (Gallese, 2013). Research shows that when people observe another person's facial expression, gesture or movement, this automatically activates similar neural networks that are used for the generation of these motor expressions (Anders et al., 2011). Among the most well known systems used for ‘embodied stimulation’ is the mirror neuron system (MNS). The regions of MNS are assumed to contain ‘mirror’ neurons similar to those studies in equivalent regions in macaque monkeys. The MNS system is a neural network connecting several brain areas including inferior parietal lobe (IPL), inferior frontal gyrus (IFG), and superior temporal sulcus (STS), (Hamilton, 2008).The MNS system is part of motor system which is robustly activated when goal directed hand movements are performed (Grafton, et al., 1992). However, apart from own hand movements MNS is also activated when people observe others enact hand gestures (Buccino et al., 2001), imitate movement (Aziz-Zadeh, et al., 2006) or only imagine that someone is performing motor action (Grafton, 1996). Over the last decade the MNS has attracted scientific attention because it was suggested that, in addition to motor functions, it may support social function.
In support of this proposition, recent research demonstrated that the mimicry of facial expressions and the production of facial expressions both rely upon the same neural networks. For instance, recently Anders et al. (2011) used information-based functional magnetic resonance imaging (fMRI) to investigate the flow of affective information between two interactive brains of romantic partners while both participants were engaged in ongoing face-to-face communication in the scanner. The sender was instructed to pose emotional expressions to share her emotional feelings with her partner while the partner was trying to understand the affective experience of the sender. The skin conductance measures showed that in both senders
and perceivers SCR increased during emotion periods, which supports the idea that affective communication leads to an increase of autonomic arousal. In addition, the fMRI results showed that the level of neural activity within a motor network of the sender’s brain predicted the neural activity in the same network in perceiver's brain and that with a temporal resolution corresponding to the phase of partners’ affective interactions. These findings are fundamental in the way they show that apart from synchrony in autonomous system, during face-to-face interactions the movements in partner’s face are directly projected to observer’s motor areas. Embodied simulation theories suggest that since the same neural networks are involved in the motor production as well as perception, the stimulation of these areas automatically generates coherent motor movements in the observer which results in motor mimicry. Through afferent feedback (from the muscle movements to the brain), mimicry helps observers to feel what others are feeling and to understand their mental states (Goldman, 2006; Stel & van Knippenberg, 2008).Here, I propose that the generation of facial expression by the sender is fundamental for the receiver to interpret the perceived emotion. While arousal during social interactions gives the intensity of communicative signals, a facial expression provides visual input that helps observers to label the physiological arousal with appropriate emotional valence. In other words, by pairing physiological synchrony with motor synchrony emotional meaning can be transferred from one individual to another. This evidence supports the view that spontaneous mimicry represents a basic form of emotional empathy (EE). This is because in order for motor synchrony to arise, there is a need for both partners’ neural networks to be simultaneously activated (De Waal, 2008). Thus, motor mimicry mirrors neuro-psychological alignment between communicating brains. The stronger and more coherent the neural activity between two individuals, the stronger the mimicry and the more likely the partners are to share the same mental state. This supports Asadas’ (2015) proposition that emotional contagion is a bottom-up process where the fast and automatic thalamo-motor and thalamo-amygdalar pathways constitute the basic form of emotional empathy. During dynamic social interactions, physiological feedback allows the receiver to make fast and intuitive decisions about their partner’s intentions without the need for conscious scrutiny of their partner’s characteristics and thus to act quickly and adaptively in a rapidly changing social environment (Adolphs, 2001).
1.4. 4. Emotional Empathy and Cognitive Empathy
Developmental theorists propose that social cognition is complex system that is influenced by social context. As was already mentioned in the introduction, there is a distinction between basic emotional contagion and more sophisticated types of social cognition such as emotional empathy and cognitive empathy (ToM). This is because if social cognition would be based purely on implicit bottom-up signaling systems, then human social behavior would be entirely driven by social cues and emotional contagion could not be controlled or inhibited. However, the mimicry literature indicates that this is not the case. Research shows that spontaneous mimicry similar to physiological linkage is influenced by social context as well as by psychological processes. For instance, a large number of studies have reported that spontaneous mimicry is stronger among relatives and familiar others (Gueguen, Jacob, & Martin, 2009; Hess & Fischer, 2013; Chartrand and Bargh, 1999). Mimicry also occurs more readily within groups rather than between groups, and is stronger between members of the same species (humans to human /chimpanzees to chimpanzee interactions) but not across different species (humans-chimpanzees) (Kret et al, 2014). In addition, previous findings indicate that spontaneous mimicry is associated to higher cognitive functions such as trust (Kret et al, in press), feelings of closeness (van Baaren et al., 2004) , liking (Lakin & Chartrand, 2003) and mimicry also increases when people are trying to understand other people’s mental states (Galinsky, et al., 2008).For this reason, it has been argued that physiological mimicry is not purely a bottom-up process, (driven by social cues in the environment), but is likely to be modulated by higher top-down processes that can prevent emotional contagion to occur.
In support of this theory, neuroimaging studies have shown that spontaneous mimicry is associated with activation of robust networks that span from lower and phylogenetically older limbic parts of the brain all the way to the neocortex (Buccino et al., 2004; Craig, 2003; Asada, 2015). This indicates that spontaneous mimicry is not restricted only to lower level (reflex-like) emotions (or emotion primitives) but is likely to involve several cortical regions and cognitive functions at the same time (Adolphs, 2001). According to de Waal (2008), because emotional contagion develops earlier than cognitive empathy, cognitive empathy is likely to be built on top
of emotional contagion. Or even more precisely, emotional contagion is the identical processes with added functions. Specifically, the main distinction is the sense of self and the ability to
control one’s own emotions which are present in cognitive empathy but absent from basic
emotional contagion.
The seed node related to self-awareness is located in the anterior cingulated cortex (ACC) and anterior insula (Craig, 2003). The ACC is known to be involved in the regulation of cognitive and emotional processes. Importantly, there are two major subdivisions of the ACC with distinct functions. The dorsal (cognitive) division and ventral (affective) division, thus the control of emotional and affective functions is separated within these two parts of the ACC. Because several studies have reported that emotional contagion and cognitive empathy are located in different neural regions, the authors frequently suggested that they are independent processes, having different roles and functions (Craig, 2003). For instance, the areas that are known to be part of MNS (Inferior Frontal Gyrus (IFG), Brodeman’s area 44) have been associated with more basic emotional empathy Rizzolatti, (2005),while ACC, prefrontal cortex, and TPJ have been proposed to play roles in perspective taking, self-awareness, and in more cognitive type of empathy such as ToM (Asada, 2011). For instance, Buccino and colleagues (2004) neurological findings suggests that MNS plays a role in understanding the meaning or goal of observed actions and is involved in predicting the future actions of other persons. In their study they asked subjects to inhibit their natural tendency to mimic an observed action, this excreted a neural response in the medial prefrontal cortex, precuneus and bilateral TPJ. This network is commonly activated in tasks requiring inferences about other people’s mental states, i.e. theory of mind task (Frith & Frith, 2003). The authors concluded that the ToM network may be responsible for controlling mimicry in typical adults.
1.4.5. Neural Network Studies
It is important to note that all these listed neural regions have been previously reported to be involved in many different functions. For instance, TPJ is known to be important for goal-related behavior (Corbetta, & Shulman, 2002), the medial prefrontal cortex, similar to the amygdala has
been suggested to play role in extinction of emotional learning (Morgan, Romanski, & LeDoux, 1993), whereas ACC is known to be important in error detection, and the online monitoring of performance (Carter, et al. 1998). The long list of functions subscribed to each region suggests that the localization of such an abstract social capacity such as ToM and empathy is a rather simplistic approach.Instead, contemporary computational models propose that with regards to the emotional and cognitive aspects of social cognition, the emergence of cognitive empathy is the result of an interaction between the bottom-up and top-down neural modulations (Asada, 2015; Decety & Lamm, 2006). This approach uses functional connectivity studies and neural network models to explain cognition as an interactive process incorporating several basic functions at the same time.
According to Decety & Lamm (2006) bottom-up functions, including physiological linkage of ANS and motor synchrony, are automatically activated by perceptual input. These bottom-up pathways are essential for direct matching between perception and action and are thus responsible for direct emotion sharing. On the other hand, higher level cortical regions such as the prefrontal and anterior cingulate cortex serve to distinguish self from other and regulate both emotional contagion and social cognition through top-down regulation based on contextual appraisal. These top-down modulatory processes however are not independent systems, as they are continually updated by bottom-up information. In these higher cognitive regions, information becomes more schematic and abstract as information clusters around learned goals and heuristic rules. In return, higher levels control the lower levels by providing top-down feedback inhibiting irrelevant information by selective attention. Top-down regulation thus modulates lower levels and adds flexibility, making the individual less dependent on external cues. This metacognitive feedback loop also plays a crucial role in perspective taking as it accounts for one’s own mental goals and retract them apart from the affective states of others (Decety & Lamm, 2006).
1.5. Mimicry in Health and Pathologies
In contrast to previous literature where ToM and empathy have been generally localized within specific neural regions and have been claimed to play distinct function, current findings imply
that empathic abilities are represented by a global synchronous change in neural activity of interacting brains. In modern neuroscience, this observation has transformed the way we think about neural architecture as it suggests that affective perception and expression of the same affect is implemented by the same neural representations and thus are not separate cognitive entities. If we accept the fact that mimicry is a physiological marker of synchronized brain activity between partners, it makes sense to argue that higher spontaneous mimicry will result in better understanding of a sender’s mental states, as the brain activity of both partners exhibit very similar patterns of neural representation. The level of mimicry should then be directly related to people’s ability to understand other people’s emotions and to empathize. A growing body of literature supports this notion. For instance, a study by Sonnby-Borgström, Jönsson & Svensson
(2003) showed that the spontaneous mimicry of facial expression is positively related to emotional empathy. In this study, high and low emotional empathy groups were tested on their facial mimicry reactions (measured by EMG) while exposed to pictures of angry or happy faces. Results showed that individuals who scored high on empathic abilities displayed a significant spontaneous mimicry in response to affective faces but the low empathy group did not display mimicry at all. These finding support the view that spontaneous mimicry is directly related to people’s empathic abilities.
1.5.1. Different Functions of Mimicry over the Lifespan
It has been proposed that mimicry plays different functions depending on the age of the organism (Asada, 2011). This is because during early childhood, the brain is still very malleable and heavily relies on external inputs. Social schemas and verbal skills are still to be developed and the communication between the infant and its caregiver is largly symbolic. Building on basic reflex-like emotional contagion, a child continuously learns new associations. Through interactions with other people, the social mind develops further. When the social schemas become more complex and stabilized, these mental representations start to serve as predictions for other people’s actions (Adolphs, 2001). During later stages in the development of the brain top-down modulations allow for the use of heuristic models which, by means of selective attention, reduce the need to process all information at once (Decety & Lamm, 2006). This saves
processing energy and the need for effortful bottom-up learning. Research suggests that a major shift between the bottom-up and top-down neural modulation occurs around 3 years of age, which is also the time when the ‘self-concept’ and the ToM is rapidly developing (Zelazo & Lyons, 2012; Ochsner et al. 209; Ochsner & Gross, 2005, Happé, 1995; Frye, Zelazo, & Palfai, 1995). This process is accompanied by the maturation of prefrontal regions and increased neural density in the ACC (Hamilton, 2008). For this reason it has been proposed that while in adulthood mimicry may play more of a social function where it represents so called ‘social glue’ promoting affiliation and trust among similar others and from the social cognitive point of view, mimicry in adulthood becomes more redundant (Lakin, et al., 2003). Yet during early childhood spontaneous mimicry is absolutely essential for the development of healthy social cognition as it allows for automatic and implicit understanding of other people’s affective experiences (Asada, 2011; Asada, 2015).
1.5.1. Mimicry in Autism Spectrum Disorder (ASD)
Indeed, from clinical research we know that reduced spontaneous mimicry is associated with social disorders such as autism spectrum disorder (ASD). For instance, several studies reveled that that people with high-functioning ASD do not spontaneously mimic other people’s facial expressions (Yoshimura, 2015). What is interesting is that people with high functioning ASD can imitate other people’s goal-directed movements (Hamilton, 2008). They can also recognize emotions from different facial expressions. However, they fail to infer other people’s intention and feelings automatically (Adolphs, 2001). Clinical tests demonstrate that ASD individuals display deficits in empathy and ToM. It has been proposed that this could be because during social interactions people with ASD show reduced eye contact, where they prefer to look into people’s mouth rather than into their eyes (Neumann, Spezio, Piven, & Adolphs, 2006). The reason for this behavior remains controversial. Direct eye contact is associated with increased autonomic arousal and increased pupil size (Harrison, 2007; Anders et al., 2011). Sepeta et al. (2012) investigated pupillary diameter as a measure of autonomic response in children with
ASD. In this task participants were presented to faces displaying happy, fearful, angry or neutral emotions with the gaze either directed or averted. Results showed that children with autism displayed reduce pupillary diameter compared to normally developing children when presented with pictures of happy faces with direct gaze. The authors concluded that a reduced pupillary response to positive social stimuli indicates that people with ASD do not find direct eye contact as stimulating/arousing as normally developing children. In general, research agrees with this conclusion as a large body of evidence exists showing that compared to normally developing children, children with autism show decreased interest in social cues (Adolphs, Sears & Piven, 2001, Pelphrey, Adolphs & Morris, 2004). It has been proposed that abnormal processing of social stimuli, commonly observed in autism, could be the cause and not the consequence of dysfunction in cognitive empathy and ToM in these disorders (Hamilton, 2008).
6.
While an abundance of literature has focused on the abnormalities in eye gaze in autism, and eye-tracking have been proposed to be a potential tool for the diagnosis of autism in its early development (Boraston & Blakemore, 2007; Riby & Hancock, 2009). Much less attention has been given to spontaneous mimicry as potential measurement of social deficits. However, eye contact represents only one of the possible emotional contagion channels, social signals are transmitted also via other means than eye gaze (see Fig. 2, Level 2 and 3). If we accept that emotions are transferred from one individual to another through different sensory channels, then the way to understand other peoples’ emotions would be to ‘resonate with’ or ‘tune in’ to one or
more of these channels (Hatfield, Cacioppo, & Rapson, 1994). Thus, here I propose that the
measurement of several types of mimicries at the same time could provide a more holistic physiological profile of the level to which the infant understands/processes other people’s social signal. Especially useful might be measures of autonomous signals that are not likely to be influenced by learning, social interactions, or conscious control (Kret, 2015). Here I suggest that measure of spontaneous autonomous mimicry along with motor mimicry early after birth, could be used as potential markers of social deficits. If true, pediatrics could intervene early substantially reducing the adverse symptoms of these disorders.
2. Discussion
Because mimicry is an evolutionarily ancient characteristic that is observable in most species, it is generally agreed that it must play some important social/cognitive function. However what is the role of mimicry and how important it is in healthy social development is not fully understood. Current review provides overreaching overview of studies ranging from developmental psychology, to social sociology, evolutionary biology and neuroscience, supporting the notion that spontaneous mimicry is essential for the development of healthy social
cognition. In this review, I argued for three main points: A) I suggested that spontaneous
mimicry provides physical-cognitive link during an organism’s development that facilitates implicit emotional communication between species. B) I further proposed that behavioral mimicry represents physiological precursor that is necessary for development of higher cognitive abilities. Finally here I argued that C) measures of mimicry may represents potential physiological marker of healthy social cognition. Current review was the first to combine theories from robotics and social sciences to support the notion that spontaneous mimicry is necessary for healthy social development.
To provide theoretical framework for my argument, I adapted the Asadas´ et al. (2015) theory of artificial development of empathy where I described how basic emotional contagion gives rice to abstract mental faculties such as theory of mind and empathy. In support of embodiment theories, in the first section of this work, I reviewed studies showing that strong physiological linkage exists between mothers and neonates. Here I showed that this physiological association translates to psycho-emotional interaction between the pair. I further implicated that while physiological linkage is generally a positive marker promoting attachment, research shows if the mother is in distress, physiological synchrony can actually have negative consequences on child’s social development. This evidence supported this report´s first argument that physiological alignment allows for direct transfer of affective information from one individual to
