Perspective-taking, intersubjectivity and non-verbal communication

The ability to take the perspective of others has been posited as central in the transition from pre-reflective to reflective forms of agency (Sugarman 8i Sokol, 2012), along with emerging symbolic and linguistic capacities (Sokol et aL, 2015). According to Martin (2006), it is through taking the perspective of others that we come to exist as self-interpreting beings. Perspective taking occurs when one attunes to the world: it is 'an active sensing of the surroundings that involves incipient responses to what is sensed’ (Stone, Carpendale, Sugarman & Martin, 2012, p. 169).

Contemporary theorists like Sokol, Martin and Sugarman acknowledge their debt to George Herbert Mead, referenced in Chapter 3, whose theory focused on the dynamic interaction between the actor and the world. For Mead, who wrote in the first decades of the twentieth century, mind and self arise primarily from social interactions. Mead termed the early responses to the world as attitudes, or a readiness to engage. An array of attitudes can arise from a moment of attuning, as individuals do not simply sense what is taking place, but also can sense what is going to or what can take place (Mead, 1934). The array of attitudes constitutes a perspective, and people occupy different perspectives as they attune from moment to moment to their surroundings and their own attitudes (Stone et al., 2012).

In contrast to the ‘like me’ and simulation theories referred to in the previous section, the ability to work with and through perspectives is not an instantiation of the workings of complex internal processes and structures. Rather, it is an embodied, situated and dynamic way of being and interacting in a world with objects and with others (Martin, Sokol 8i Elfers, 2008). According to Martin (2008), the simulation and ‘like me’ theories are flawed because they fail to explain how it is possible for the infant to draw analogies between his experience and that of another, in the absence of any understanding of others and their perspectives. He states: ‘what possible criteria might the analogically reasoning infant use to establish identity across his own and another’s experience if such an identification is based only on his own experience?’ (p. 101). Similarly, Muller and Carpendale (2005) argue that it is doubtful that young infants are capable of the counterfac-tual thinking required by the ‘like me’ and simulation theories, because according to those perspectives infants must model someone else’s feelings that they, at that moment, do not have. Thus, these authors are very much rejecting the cognitive perspectives proposed by theorists like Tomasello and Meltzoff. Instead, they turn to the social and the relational.

Within what will be termed the perspectival view, the emergence of perspective-taking abilities is grounded in the early interactions between infants and their caregivers, as well as the objects and artefacts in their environment. After birth, infants engage in sequences of interaction with others, most notably their caregivers. These interactions have been likened to a dance as infants and caregivers vocalise, gaze, smile and touch in turn-taking sequences (Stern, 1985). These interactions have been termed 'protoconversations' (Bateson, 1979; Trevarthen, 1979). Through repetition of social exchanges, infants come to delimit and anticipate the significance of social gestures, and distinguish between persons and objects, and their different properties (for example, that physical objects cannot predict the intention of the infant, but social partners can). They begin to understand the perspective of others through coordinated social acts (such as being fed, dressed or picked up), initially reflected in their ability to adjust from moment to moment in line with the anticipated conduct of the interactive partner.

From about 9 months of age, infants begin to engage with other persons in interactions that involve a third object or event - a triadic interaction -and, into the second year, shared understanding and shared engagement emerge in coordinated acts (Moore, 1999; Stone et al., 2012). Common examples of these triadic interactions include joint visual attention, gaze following, pointing and social referencing (where the infant looks to the other person to gauge their emotional reaction to a situation). By about 12 months, infants demonstrate their motivation for joint attentional engagement in numerous ways: they follow others’ focus of attention (they follow gazes and pointing); they actively direct others to follow their focus of attention (they point and show), and they will voluntarily turn away from interesting sights and look to others’ faces to invite them to share attention with them (Carpenter et al., 1998; Carpenter, 2011).

Several studies illustrate infants’ adeptness at using pointing to direct the attention of others in order to share attention and interest (Liszkowski, Carpenter, Henning, Striano 8i Tomasello, 2004; Liszkowski, Carpenter 8i Tomasello, 2007). In one study, an experimenter reacted in one of four ways to 12-month-old infants’ pointing at interesting events: the experimenter attended to the infant only, not the referent, and displayed positive emotion to the infant (face condition); the experimenter reacted only to the event, and not the infant, and did not comment or display emotion about it (event condition); the experimenter alternated attention to both the infant and the referent and commented positively (joint attention condition); and finally, the experimenter ignored the infants’ pointing and did not attend to either the infant or the event (ignore condition). Infants displayed dissatisfaction in all conditions except the joint attention condition, leading to the conclusion that infants wanted the adult to attend to both themselves and the object, so that attention and interest in the object could be shared. In a later experiment, the researchers demonstrated that what is important to infants is that the adult attends to the specific referent to which they are pointing, and not some random object in a similar direction. In this study, infants repeated their pointing to the original referent if the adult 'misunderstood’ the referent and attended to a different object. However, if the adult reacted as though uninterested, then the infant stopped pointing altogether. The authors concluded that, by this age, declarative pointing is a communicative act aimed at sharing with others attention to a referent, and a specific attitude about that referent.

One perspective on these early social interactions is that of intersubjectivity (Trevarthen, 1979), which assumes that infants are born with an innate sense of the other’s mental states, such as intentions and feelings - 'readiness to be aware of a human presence and to follow and communicate with human states’ (Trevarthen, 1993, p. 121).

Micro-analyses of protoconversations between mothers and their 2-month-old infants reveal the repertoire of expressions infants share with an adult, and that both are engaged in anticipating the response of their partner, such that if the normal flow of interaction is interrupted, either partner can become upset and the interaction breaks down. In one study, 6-to 12-week-old infants were observed while interacting with their mothers via a live image through a television monitor, and, after a period of time, the live image of the mother was switched with a non-contingent image of the mother that had been recorded earlier. Even the very young infants detected the change in the contingency structure of the interaction and reacted with distress to this loss of contingency. These findings were later replicated by Nadel and colleagues (1999) who concluded that at 2 months infants are highly sensitive to the timing and emotion of a mother’s expression in conversation. In a similar vein, studies adopting the Still-Face paradigm - whereby the mother interacted normally with her infant for three minutes and then, after a short break, sat in front of her seated infant, remained unresponsive and maintained an expressionless face - highlighted the importance of interactional reciprocity for infants. In the original study using the paradigm, Tronick and colleagues (1978) observed a sequence of steps through which infants (aged 1 to 4 months) proceeded in order to accommodate the ‘Still Face’ state of their interactive partner: an initial greeting, realisation of the perturbation, wariness, checking, repeated attempts to mobilise the mother and eventual withdrawal. Although language is not yet part of the interaction, a lexicon of expressions and actions conveys to each partner their inner emotional state.

Several points define the intersubjectivist position: infants are born with innate mechanisms to identify and attribute mental states (such as feelings and intentions) to others’ minds during early social interactions; from the beginning of life the infant has introspective access to a relatively rich set of mental states of the self, such as emotions, motives, goals; and mental states in the self can be recognised as being similar to or corresponding to mental states of the other, and as such are experienced as being shared with the other (Trevarthen & Aitken, 2001; Gergely, 2002; Fonagy, Gergely & Target, 2007). This account of perspective taking is richly mentalistic and, like the simulation models previously reviewed, has been bolstered by evidence which highlights the multi-component neurophysiological underpinning of perspective-taking capacities. Notably, the mirror neuron system, a class of visuomotor neurons, has been implicated as part of the neuro-biological grounding for the expression of intersubjectivity (Rizzolatti 8i Craighero, 2004). Based on research with monkeys, it was discovered that a particular set of neurons, activated during the execution of purposeful, goal-directed hand actions (such as grasping), also activated when the monkey observed similar hand actions being executed by another individual. It was hypothesised that these mirror neurons, by matching observation of the action with action execution, constitute one mechanism by which the actions of others can be understood (Ferrari & Gallese, 2007). Although there is ongoing debate about the role of mirror neurons in humans, they may represent one means for experiencing correspondences between experiences of others and similar experiences of self (Mascolo & Fisher, 2015).

However, the intersubjectivist position has been challenged on the grounds that evidence called upon to support this position can be explained without assuming an innate ability to identify and attribute mental states to the other’s mind. For example, Gergely (2002) notes that ‘innate attentiveness and reactivity to human facial/vocal displays together with an innate sensitivity to contingency are sufficient to explain the structured bi-directional nature of early affective interactions’ (p. 30). He goes on to suggest that these early affective and imitative interactions serve important evolutionary functions, such as to promote proximity to an attachment figure and to establish an emotional-intentional scaffolding environment. In brief, Gergely’s perspective, termed the no starting-state intersubjectivism position, proposes that the ability to detect contingency is fundamental to the infants’ developing self- and then other-awareness. Over the early months of development, infants are highly sensitive to the contingent relations between their motor responses and the ensuing stimulus events, and until around 3 months they display an attentional bias towards such perfect contingencies (Bahrick & Watson, 1985). From around 3 months, infants begin to orient more towards non-contingent stimuli, reflecting a new preference for less-than-perfect contingencies (Rochat & Morgan, 1995). According to Watson (1985), this switch in the contingency-detection device around 3 months of age orients the infant away from self-exploration where contingencies are perfect (for example the act of putting their hand to their mouth results in them being able to sense their hand at their lips) to exploration of the social world where contingencies are high but less than perfect (for example when they shake a rattle, a sound emerges, but the sound might differ depending upon how hard they shake the rattle, or which rattle they shake). This contingency-detection mechanism is also proposed to be central to the development of the infant’s emotional self-awareness (for example the feeling of how satisfactory something is) and control.

During early caregiver-infant interactions, caregivers repeatedly display attuned but partially imitative reflections of their infants’ affective or intentional behaviours. The theory proposes that infants register this affect-mirroring process and note through contingency detection that their behavioural-state expressions exert high contingent control over their caregivers’ mirroring expression. Through this process, the infant learns about the internal cues that are active when they display behaviour which controls the adults’ affect-mirroring expressions. Thus, the infant is sensitised to the internal cues that covary with their expression of emotions and actions. This leads to the establishment of representations which provide the cognitive means for accessing and attributing emotions to the self, and thus the development of self-control and emotional awareness. However, according to Gergely, rather than being innate, as suggested by Trevarthen, intersubjective understanding does not emerge until the second year, after the infant develops accessible representations of affective and intentional self-states (Gergely & Watson, 1999; Gergely, 2002). These representations are essential for an emerging sense of oneself as an agent in the world.