Engagement: Looking beyond the mirror to understand action understanding

1 British Journal of Developmental Psychology (2015) © 2015 The British Psychological Society www.wileyonlinelibrary.com Special issue paper Engagement: Looking beyond the mirror to understand action understanding Vasudevi Reddy1 and Sebo Uithol2,3* 1 Department of Psychology, University of Portsmouth, UK Bernstein Centre for Advanced Neuroimaging, Charit
e Universit€atsmedizin, Berlin, Germany 3 Department of Neuroscience, University of Parma, Italy 2 In this paper, we argue that the current focus on mirroring as the route to explaining the development of action understanding is misleading and problematic. It facilitates a fundamentally spectatorial stance, ignoring engagement and dialogue; it focuses on similarity between self and other and neglects difference; and it succumbs to the static terminology of mechanism rather than a dynamic language of process. Contrary to this view, dialogic exchanges are evident from the start of life, revealing infants’ ability to engage with and respond appropriately to actions that are outside their own motor repertoire. We suggest that engagement rather than mirroring better accounts for many current findings in action understanding. The neurological evidence to date shows that action perception involves a process of continuous synchronization and change, suggesting that it might be more fruitful for research and theory to look beyond mirroring and instead adopt dynamic processual explanations of action understanding within interaction. The argument that infants’ understanding of others’ actions originates in infants’ own action capabilities (Hunnius & Bekkering, 2014; Woodward & Gerson, 2014) is both cogent and consistent with the majority of current views in developmental neuroscience. In essence, the argument is that the neural systems that underpin infants’ own actions are strongly connected to, and indeed drive, infants’ ‘analysis’ of others’ actions. This claim appears to be both strongly supported by current evidence and logically unimpeachable. If action is prospectively controlled (and there is good reason to believe that it is), then there must be ‘abstract’ intentional (and prospective) relations between goal and own movement which, when present, must also be available to infants when they perceive others’ movements in relation to potential goals. The claim is heavily grounded in neuroscience with discoveries of mirror mechanisms allowing the perception of intentional relations in others’ actions on the basis of own action production. In all, despite some recent criticisms,1 the claim seems compelling. *Correspondence should be addressed to Sebo Uithol, Charite Universitätsmedizin Berlin, Bernstein Centre for Advanced Neuroimaging, Philipstrasse 13, Haus 6, 10115 Berlin, Germany (email: sebouithol@gmail.com). 1 Recent criticisms of the mirroring theory of action understanding have argued that mirroring is in fact a by-product of associative learning (Heyes, 2010) and not necessary for action understanding (Hamilton, 2009; Hickok, 2009). DOI:10.1111/bjdp.12106 2 Vasudevi Reddy and Sebo Uithol The claim becomes problematic, however, when one realizes that there are at least three ways in which infants are exposed to others’ actions. The infant could merely observe the frequent occurrence of an action by the other without involvement. The infant could observe the similarity of an action by the other to that within the infant’s repertoire. The infant could feel and produce a response to an action by the other which is directed to the infant – even in the absence of the capacity to produce the original action. The received interpretation of mirror mechanisms in action understanding – the directmatching hypothesis (Rizzolatti, Fogassi, & Gallese, 2001) – proposes that by mapping the observation of another’s action onto one’s own motor repertoire, a first-person experience of the observed action is created (Rizzolatti & Sinigaglia, 2010). This means that although action mirroring can be embedded in interaction, it need not be. And while producing a similar action (i.e., imitation) may be based on action mirroring, producing a different response falls outside the scope of such a mechanism. In this paper, we take issue with the claim that infants’ capacity to understand an action is dependent on their capacity to produce the same action for three reasons. First, the claim is, despite the best intentions of its proponents, fundamentally spectatorial. The crucial feature in such a first-person or third-person account2 is the infant’s perception of the similarity of the other’s action to an action in her own repertoire. The infant is still an observer and analyser of the action, not necessarily a participant with it. Second, dialogue between self and other needs difference – not identity – to be meaningful. Although neither similarity nor difference are absolute or categorical entities, mirroring explanations focus on similarity between self and other, while action understanding within dialogue would require us to focus on infant actions as responses to difference. Third, mirroring explanations incline towards a focus on mechanism and static representation (i.e., a dedicated mechanism that produces a mental or motor representation similar to the one present in the observed actor). We argue that understanding action is a more fluid and dynamic process, without a pre-defined and clear endpoint. Cognitive neuroscience and developmental science are better off focusing on the role motor resonance plays in enabling the organism to respond to relevant features (Hutto, 2013), and to engage in dialogue with it, than on the static recognition of isolated actions. Before we discuss evidence that action understanding is not best interpreted solely in terms of mirroring, we briefly discuss the (largely neuroscientific) background of the current interpretation. From neurons to mechanism As is widely known, mirror neurons were first discovered in area F5 of the ventral premotor cortex of macaque monkeys. Monkeys were seated with their heads and bodies fixed while watching objects being picked up and grasping these objects themselves. Single neurons appeared to be selectively activated not only when the monkey grasped 2 Discussions about knowing other minds note the spectatorial stance implied by both first-person and third-person accounts (Costall, 2013; Leudar & Costall, 2009). In both, the knower is fundamentally an observer of the other mind. In first-person accounts, the knowing is accomplished through either explicit analogy with the self or implicit simulation of the other within the self. In third-person accounts, the knowing is accomplished through deduction and theory applied to the observed behaviour. Both adopt a Cartesian rather than a Hegelian view of knowing (Markova, 1982), in which the knower is independent of that which is known. Engagement beyond the mirror 3 the object, but also when the experimenter grasped it. Di Pellegrino, Fadiga, Fogassi, Gallese, and Rizzolatti (1992) hypothesized that ‘the actions performed by other monkeys must be a very important factor in determining action selection’, explicitly relating their findings to response generation and social interaction. Fairly soon, however, the role attributed to these neurons shifted from action selection to action understanding (Gallese, Fadiga, Fogassi, & Rizzolatti, 1996; Rizzolatti, Fadiga, Gallese, & Fogassi, 1996) and ‘mind reading’ (Gallese & Goldman, 1998), and mirror neurons were hypothesized to be the neural instantiation of a simulation of the observed action in a motor format (Gallese & Sinigaglia, 2011; Goldman, 2006). This move from action selection to action understanding strongly facilitates an interpretation in terms of passive observation rather than active involvement. As it is hard to establish mirroring at the single cell level in humans (but see Mukamel, Ekstrom, Kaplan, Iacoboni, & Fried, 2010), the word ‘mirror neuron’ was generally omitted when discussing human motor activation during action observation. Instead, the notion of ‘mirror system’ or ‘mirror mechanism’ (Rizzolatti et al., 2001) provides a more neutral term to account for the activation of motor areas during action observation (including the ventral pre-motor cortex, and the rostral part of the intraparietal lobule, with the posterior superior temporal sulcus being its main visual input (Iacoboni & Dapretto, 2006). By referring to these areas as the ‘mirror system’, or implementing a ‘mirror mechanism’, claims about the exact implementation at the single cell level are avoided. Although this agnosticism is praiseworthy, the particular notions of ‘system’ and ‘mechanism’ implied here do involve a crucial conceptual step. They suggest a confined entity that has the purpose of accomplishing a specific task: a mechanism for understanding others. Thus was born the idea of a dedicated mechanism for understanding observed actions by mapping them onto the observer’s own motor repertoire. As premotor areas are part of this action observation system, activation in these areas is easily interpreted as instantiating simulation of the observed actions. Indeed, TMS (Fadiga, Fogassi, Pavesi, & Rizzolatti, 1995), fMRI (Buccino et al., 2001), MEG (Hari et al., 1998), and mu suppression-based EEG paradigms (Cochin, Barthelemy, & Roux, 1999) showed activation in pre-motor and motor areas during action observation and were interpreted within the framework of action understanding. The idea of a mirror mechanism rapidly found its way into developmental science. As the motor repertoire develops during infancy, the resulting framework generates testable hypotheses for the development of action understanding. And indeed, previous motor experience alters perceptions of others’ actions. Sommerville, Woodward, and Needham (2005) showed that 3-month-olds who are given artificial motor experience of reaching and grasping – in contrast to 3-month-olds not given such experience – can detect the goal directedness of the reaches of a disembodied adult arm. Woodward and Guajardo (2002) were able to directly link age-related motor production abilities with perception of motor acts. This link was further established using eye tracking and EEG studies (Ambrosini et al., 2013; Cannon & Woodward, 2012; Daum, Prinz, & Aschersleben, 2011; Kanakogi & Itakura, 2011). In all, the evidence seems strongly in favour of the hypothesis that understanding patterns and meanings in others’ actions comes from having such patterns and meanings in one’s own actions, with the idea of a dedicated action mirroring system that develops with the observer’s motor skills leaving little room for a different interpretation. Yet, there are reasons for rethinking the developmental evidence and looking beyond the mirror for explanations of the development of action understanding. 4 Vasudevi Reddy and Sebo Uithol The importance of engagement: Self and other as participants not spectators Our predominant social experience in the first year of life comes from situations where we are directly involved and often the direct recipients of others’ actions and utterances. At any point in our lives, actions directed to us put us immediately in a difference pragmatic frame where a response of some sort – even if only to ignore the action – is demanded of us. They also involve us in an immediately emotional relation – arousing interest, dislike, attraction, joy, pain, indifference, and so on. Although in older children and adults such emotional involvement can occur in spectators too – as in watching movies – they are far more typical of direct interactions and, we argue, developmentally crucial in early infancy. Various neuroscientific studies have shown that being a recipient rather than an observer makes a difference to neural responses (Kampe, Frith, & Frith, 2003; Schilbach, Eickhoff, Mojzisch, & Vogeley, 2008; Schilbach et al., 2006). Performing an action – even one as simple as looking at a stimulus on a screen together with someone – affects brain activation (Hoehl, Michel, Reid, Parise, & Striano, 2014). Joint looking evokes activity in a brain area argued to be involved in thinking about mental states, while solo looking does not (Schilbach et al., 2013). Interpersonal engagement reaches into the future as well: Not only do the actions of a partner in a game cause greater motor activation than the actions of a non-partner (Kourtis, Sebanz, & Knoblich, 2010), but observation of the actions of people with whom you have previously interacted produces greater sensorimotor activation than observation of the actions of people you have not interacted with (Kourtis, Knoblich, & Sebanz, 2013). At the simplest level, it seems that being engaged with another person in some way – whether by being called, being looked at, being smiled at, or by doing (or having done) simple actions with them – has a fundamental impact on brain responses. Engagement also leads to different neural activation in infants. Direct gaze to 4- and 5-month-olds evokes early and increased gamma band activity compared to averted gaze (Grossmann, Johnson, Farroni, & Csibra, 2007) and enhanced pre-frontal cortex activity (Grossmann, Parise, & Friederici, 2010). Using ERP studies, Farroni, Massaccesi, Pividori, and Johnson (2004) found that in 4-month-olds cortical processing of direct gaze is enhanced even when the face is averted; and there is better recognition of (upright) faces with direct gaze than with averted gaze (Farroni, Massaccesi, Menon, & Johnson, 2007). Mutual gaze also enhances the possibility of learning about the world and the processing of objects (Parise, Reid, Stets, & Striano, 2008; Reid, Striano, Kaufman, & Johnson, 2004) and the neural processing of facial expressions (Striano, Kopp, Grossmann, & Reid, 2006). Prior mutual gaze enhances gaze following in 6-month-olds (Senju & Csibra, 2008) and faster saccades to the peripheral targets of others’ gaze even in neonates (Farroni et al., 2004). Being addressed in one way or another is also directly ‘preferred’ by adults (Schnell, Bluschke, Konradt, & Walter, 2011) and by infants: 2- to 5-day-old infants already look preferentially at frontal photographs of faces in which the eyes are oriented forwards (looking at the infant) than those in which the eyes are averted to the side (Farroni, Csibra, Simion, & Johnson, 2002), an effect also present when using schematic faces (Farroni et al., 2004). Together these findings suggest that that the phenomenon of being addressed is developmentally crucial and has broader implications than simply the detection of ostensive cues as argued by some (Csibra, 2007) – it draws the infant further into emotional involvement with the other and the other’s interests, rather than only into analysing and interpreting the other’s behaviour as cues. The findings converge on the Engagement beyond the mirror 5 idea that social cognition is not just normal cognition with more complexity added. Rather, they show that our brains are deeply social and that a social setting alters even the most basic processes. Most of these findings focus on being addressed by gaze; being addressed by actions is likely to be much harder to avoid noticing and responding to and also shows increased brain activity in the relevant regions (Nagels, Kircher, Steines, & Straube, 2015). The world of newborn and very young infants is particularly full of actions performed by adults upon or towards their faces, limbs, mouths, and bottoms. Dialogue needs difference not similarity The infant’s engagement with adults from birth has a dialogic character (Bateson, 1979; Trevarthen, 1977), revealing reciprocal interactions with mutual adaptation to each other’s actions and orientations. From at least 2 months of age infants are capable of perceiving the temporal and affective contingencies between their own actions and orientations and others’ responses (Bigelow, 1998; Markova & Legerstee, 2006; McQuaid, Bibok, & Carpendale, 2009; Nadel, Carchon, Kervella, Marcelli, & R
eserbat Plantey, 1999; Reddy, Chisholm, Forrester, Conforti, & Maniatopoulou, 2007). Such contingencies must also be perceivable in the reverse direction – that is, the relatedness of their own responses to others’ actions must also be perceptually manifest. Dialogue needs a responsive voice – a different ‘other’ – to develop, not similarity. As Buber put it (Friedman, 1955) ‘Though the Thou is not an It, it is also not “another I”. He who treats a person as “another I” does not really see that person but only a projected image of himself’. Even at 3 months, infants prefer contingent responses to their own initiatives to be different but appropriate, rather than direct imitations of their acts (Markova & Legerstee, 2006). In other words, to engage with another dialogically, you do need a different other. The complex embeddedness of infant actions in participatory dialogues from the early months is evident in face-to-face as well as caretaking and other playful interactions (Nadel et al., 1999; Raczaszek-Leonardi, 2013; Rossmanith et al., 2014), with infants not only responding to the adult acts they receive, but receiving further adult actions in response to their own acts. From the first moment of life, the majority of infant experiences of others’ actions are of adults performing actions on them which the infant cannot possibly accomplish: holding the legs to lift the bottom, wiping the eyes, cleaning the nose, stroking the face, putting the arm in a sleeve, holding the chest to lift the baby up, and so on. Although these actions are not part of the infant’s action repertoire, infants do respond to them, often emotionally and appropriately. From a mirroring perspective, such adult actions to the infant would be either irrelevant or incomprehensible to the infant. However, the few studies that exist which test actions directed to the infant suggest that there is a lot more understanding going on in the early months than can be explained by mirroring. Habituation studies of action awareness – considered to be less cognitively demanding than studies of ‘online’ anticipatory responses (Cannon & Woodward, 2012) – show the earliest infant awareness by 5 months of age (Woodward, 1998) or 3 months with action training (Sommerville et al., 2005). Studies using measures of anticipation show their earliest effects at 6 months of age (Ambrosini et al., 2013; Kanakogi & Itakura, 2011). In all these studies, however, the infant was a mere spectator of adult actions directed to a distal object. Studies of actions where the infant is the recipient of the action yield earlier evidence of action awareness. By at least 2 months of age, infants adjust their bodies by increasing its rigidity and by opening their arms when adults extend their arms to pick 6 Vasudevi Reddy and Sebo Uithol them up (Reddy, Markova, & Wallot, 2013). Infants at 2 and 3 months cannot yet reach out with both arms in anything like the pickup gesture, yet have little difficulty anticipating and adjusting to its specific meaning. In preschool children with autism, on the other hand, where there is little difficulty with reaching out, anticipatory adjustments to being picked up have been reported to be absent or problematic (Kanner, 1943), suggesting that such anticipations may be connected to the perception of the intentionality of actions. Similarly, unlike the anticipatory mouth opening of typically developing infants being fed with a spoon or indeed from a bottle or breast, children with autism do not open their mouths in anticipation of the approaching spoon (Brisson, Warreyn, Serres, Foussier, & Adrien, 2012).3 Further, at 2 and 3 months, anticipation of being picked up occurs with constant gaze to the mother’s face rather than to her hands or away (Reddy et al., 2013), suggesting that this is not merely the anticipation of an impersonal event but of the action of a person (see Bruner, 1974, for a similar interpretation of the distinction between ‘giving’ with gaze to the hand and ‘giving’ with gaze to the other’s face). Our argument is that the infant’s responses both reveal and further build the infant’s understanding of the adult’s pick up act; the response itself illuminates and co-creates the meaning of the act. The mirroring framework has nothing to say about such clear instances of infant action understanding, as the observed action cannot be mapped onto the motor repertoire of the infant. Neonatal imitation studies offer the richest (as well as the most controversial) source of experimental data about infant responses to actions towards themselves. Debates continue about whether the infants’ ‘imitative’ responses to the actions modelled by adults extend beyond tongue protrusion, whether they are really ‘imitative’, and how one can explain the processes involved in it (see, for instance, Jones, 2009). Nonetheless, one key point of interest is the nature of the engagement involved in these interactions. Three sets of studies from three different laboratories attest to the conversational nature of the exchanges. Infants attend intently to facial actions directed towards them by adults, but not all attentive infants perform ‘imitations’, and those that do, do so with a range of patterns – from single actions to a series of acts either with increasing or decreasing proficiency (Kugiumutzakis, 1998). These imitative acts may also be anticipations of or invitations to the model to perform, also occurring (with infant gaze to the model) when the model is still and is no longer modelling the act (Meltzoff & Keith Moore, 1994; Nagy & Molnar, 2004). As with adults (see Dumas, Martinerie, Soussignan, & Nadel, 2012), even with very young infants the role of model and imitator can be hard to tell apart. It has been argued that even imitative interactions are not merely imitative and that the actions themselves are never identical to those observed (Csibra, 2007). They often involve varying degrees of approximation and varying degrees of hesitation and change: responses to, rather than mimicking of, another person; unless done robotically, imitation is always a response (Uzgiris, 1981). Neonatal imitation can provide strong evidence of motor resonance and offer an explanation of how the infant recognizes the similarity between actions observed and actions produced. However, regardless of one’s take on it as imitation, these exchanges provide even stronger evidence of the neonate’s interest in engaging with actions directed towards them and in ongoing dialogic interactions with others. 3 One could object that preparing for the pickup or the approaching spoon is not really understanding of the action, but a mere prediction of what will happen, and an associated response. But notice that the highly studied ‘target prediction’ – usually cast in terms of ‘goal understanding’ – is just as easily explained in terms of predictions and associations (see Uithol & Paulus, 2013; Uithol, van Rooij, Bekkering, & Haselager, 2011). Engagement beyond the mirror 7 We offer one further example as evidence of appropriate responses by infants to acts outside their own repertoire: infant laughter to deliberate, infant-directed acts of absurdity, and silliness by adults. Adults engage in a range of acts to amuse babies – and the earliest successful acts tend to be absurd facial expressions and abruptly ending tones or repetitive sounds like ‘brr-ing’ the lips’ (Sroufe & Wunsch, 1972). From around 3 or 4 months of age, infants can produce deep belly laughs to a range of adult acts directed towards them. Adult attempts at clowning tend to be wrapped up in a framework of smiling and laughter (Mireault et al., 2012) which is likely to be emotionally reassuring. Nonetheless, the adult acts that elicit such reactions and anticipations from infants can be entirely outside the infant’s own motor repertoire. The intention and ‘meaning’ of these adult acts comes from the infant’s own affective response to them and the ensuing dialogue, not from the motor resonance of the acts themselves (see also the affective embeddedness of joint attentional acts, Leavens et al., 2014). Engagement affords a broader and relational level of explanatory framework of current findings than does mirroring. It accounts for learning from the experience of observed actions that are outside the infant’s own repertoire as well as from those within; it accounts for perceived meaningfulness of observed actions under situations of motor constraint (Csibra, 2007); it accounts for findings that observational experience sensitizes infants – they attend more specifically to specific features of pets if they have experience of pets at home, a finding that does not extend to human faces (Hurley & Oakes, 2015); and it accounts for findings showing selective mirroring (Caggiano et al., 2012). Arguments about the primacy of engagement in the development of mind knowledge (Gallagher, 2001, 2008; Reddy, 2003, 2008, 2011) claim that being directly addressed by ‘other minds’ is necessary for typical social cognition. They posit a gradually expanding awareness of aspects of mind emerging from direct experience with them in second-person interactions. Thus, another’s attention directed to her allows the infant to experience attention and response by being the object of attention, before she can grasp the meaning of others’ attention directed to other objects. From a simple grasp of attention directed to self, the infant comes to understand attention to other objects – to parts of the infant’s body within the first 6 months (such as in action games on the infant’s hands and feet) to the infant’s own actions from around 7 or 8 months of age (such as repeating funny faces or sounds or movements to re-elicit adult attention or laughter) before attention to objects in distal space between 9 and 12 months of age (such as following gaze to distal targets and pointing). Similarly, understanding the goal directedness of intentional actions, it is predicted, emerges from being the goal of others’ intentional actions from the start of life. This experience, we argue, forms the basis upon which understanding goal directedness to other objects emerges – grasping actions directed towards other objects and grasping adults’ intentions to modify the infant’s own actions (Heimann & Uithol, 2015; Reddy, 2015; Reddy et al., 2013). These views predict that difficulties in atypical development stem partly from earlier difficulties in second-person interactions, and afford predictions about the primacy of second-person engagements in typical development. Neonatal imitation, for instance, should be easier to obtain when the model addresses the infant with the act than when the act is simply witnessed, addressed to no one. Understanding the goal of a reach (when witnessing someone reach for an object, for instance) should not be easy to grasp if you have never been reached for yourself. Typical conceptual and inferential understanding of attention and goal directedness must come from these experiences of being the object or goal. 8 Vasudevi Reddy and Sebo Uithol Action understanding is a dynamic process In her critique of mirror system explanations of action understanding, Southgate (2013) argues that action production experience may provide better learning opportunities than action observation alone (suggesting that the absence of learning opportunities in a merely observed action, rather than absence of motor activation, is key to not understanding it). She argues that action production might lead to better learning about the goal object of the action (such as object names; Yu & Smith, 2012), better retention of its information consequences (Begus, Gliga, & Southgate, 2014) and possibly greater arousal. While her departure from mirror-based explanations of action understanding is a great leap forward, we believe her alternative still suffers from the spectatorial ghost that haunts the debate, framing her explanation in terms of cues and prediction. What her alternative does not quite touch on is that most infant action production experiences occur within pragmatic frames and long engagement histories, not only sensitizing infants to the specific actions of others but also drawing them into the contexts in which these actions make sense. In seeking a processual rather than a mechanistic explanation, we can draw upon various strands of cognitive science as sources for an alternative interpretation of the finding of mirror neurons and motor resonance. First, field theory (Lewin & Cartwright, 1951) and Gibsonian approaches have offered strong arguments for thinking of psychological phenomena in terms of process rather than mechanism. In their rejection of an animal–environment dualism, they posited action fields with pre-existing or ongoing sociocultural tuning processes, avoiding the more typical analytic focus on representations, and on moments of ‘here and now givenness’ (Kadar & Effken, 2005). Such a processual view steers interpretation away from the infant as analyser of observed action, towards a framework in which infant and actor engage in an action dialogue, through which understanding and meaning is created. Second, dynamic systems theory offers a non-representational account of cognitive processes, focusing on processes that are dynamically linked to other processes (Beer, 2000; Thelen & Smith, 1994). Instead of postulating an internal mental state that is causally responsible for the subsequent behaviour, this approach seeks the causes of behaviour in the interaction of various control processes both within and outside the central nervous system, in dynamic interaction with external factors (Schurger & Uithol, 2015; Shenoy, Kaufman, Sahani, & Churchland, 2011). Action understanding would consist, not in inferring a mental state that could be responsible for the observed action (see Uithol, Burnston, & Haselager, 2014 for problems with this framework), but in ongoing prediction and response generation within interaction. Third, enactivism highlights the fact that cognition did not evolve for creating faithful representations about the external world, but to allow the organism to act appropriately. That is, mind is seen not as inhering in the individual, but as emerging and existing dynamically in the relationship between organisms and their surroundings (including other agents; McGann, De Jaegher, & Di Paolo, 2013). It is recognized that the interaction process itself forms an irreducible domain of dynamics which can be constitutive of individual agency (De Jaegher & Froese, 2009). According to Di Paolo and De Jaegher’s ‘Interactive Brain Hypothesis’, (2012) when an individual interacts with others, the interaction processes would not function merely as perceptual input to ready-made mechanisms but would also play a role in shaping brain mechanisms for social interaction. Each of these approaches enables interpreting motor activation during action observation not merely in terms of ‘resonance’ (i.e., evoking the same motor representation in the observer), but in terms of interaction (Dumas, 2014) and participation in the Engagement beyond the mirror 9 creation of meaning. This mutual adaptation results in interactional synchrony to which both members contribute (Dumas, Nadel, Soussignan, Martinerie, & Garnero, 2010). Next, these approaches predict that motor activation during action observation is not automatic and stable, but can be influenced by a variety of factors, most notably the interaction and the actor. This prediction is indeed supported by a volley of findings on both infants and adults. Hoehl et al. (2014) found desynchronization in the alpha frequency range in nine-month-old infants when looking at an object together with an adult during a social interaction involving eye contact. Interestingly, when infant and experimenter only looked at the object without engaging in eye contact, no such effect was observed. In adults, Oberman, Pineda, and Ramachandran (2007) showed that the mu frequency is modulated by the degree of social interaction. Being the addressee of an action results in the largest suppression. Finally, motor resonance is response dependent. Newman-Norlund, Van Schie, Van Zuijlen, and Bekkering (2007) found greater activation in the right inferior frontal gyrus and the bilateral intraparietal lobules (areas typically associated with action mirroring) when participants had to perform a complementary action compared to an imitative action. We interpret these findings as strong support for a process view of action understanding. Thinking in process terms forces us to recognize something ongoing rather than a snapshot, something changing and developing rather than static, and something open to experience rather than pre-determined. Our point here is that the phenomena of action understanding (both at behavioural and neural levels) are fundamentally emergent, contextual, open to experience, and subject to change within engagement. Thus, thinking of action understanding itself as a process is different from thinking of it as a mechanism and opens us up to different methods, evidence, and explanations. Conclusion We have argued that the current focus on mirroring as the route to explaining the development of action understanding is misleading and problematic. Instead, we argue that the predominant state of being is relational and involved and that any complete account of (the development of) action understanding must take this engagement into account. If we treat interaction as a special case of actions, we are missing out on grasping a key source of experiences for infants’ understanding of others’ actions. When we move beyond the mirror and start viewing interaction as the more basic setting for social cognition and exposure to actions, we open our eyes to different methods, evidence, and explanations. Action understanding takes place in a context of interaction in which meaning and sense-making emerge from an ongoing dialogic engagement. Acknowledgements The writing of this paper was supported by grant number eRS#16834 from St. Louis University and the John Templeton Foundation to the first author and a Marie SklodowskaCurie grant 657605 of the European Union’s Framework Programme Horizon 2020 to the second author. The opinions expressed in this publication are those of the authors and do not reflect the views of the funders. 10 Vasudevi Reddy and Sebo Uithol References Ambrosini, E., Reddy, V., de Looper, A., Costantini, M., Lopez, B., & Sinigaglia, C. (2013). Looking ahead: Anticipatory gaze and motor ability in infancy. PLoS ONE, 8, e67916. doi:10.1371/ journal.pone.0067916 Bateson, M. C. (1979). The epigenesist of conversational interaction: A personal account of research development. In M. Bullowa (Ed.), Before speech the beginning of human communication (pp. 63–77). Cambridge, UK: Cambridge University Press. Beer, R. (2000). Dynamical approaches to cognitive science. Trends in Cognitive Sciences, 4, 91–99. doi:10.1016/S1364-6613(99)01440-0 Begus, K., Gliga, T., & Southgate, V. (2014). Infants learn what they want to learn: Responding to infant pointing leads to superior learning. PLoS ONE, 9, e108817. doi:10.1371/journal.pone. 0108817 Bigelow, A. E. (1998). Infants’ sensitivity to familiar imperfect contingencies in social interaction. Infant Behavior and Development, 21(1), 149–162. Brisson, J., Warreyn, P., Serres, J., Foussier, S., & Adrien, J.-L. (2012). Motor anticipation failure in infants with autism: A retrospective analysis of feeding situations. Autism: The International Journal of Research and Practice, 16, 420–429. doi:10.1177/1362361311423385 Bruner, J. S. (1974). From communication to language – A psychological perspective. Cognition, 3, 255–287. doi:10.1016/0010-0277(74)90012-2 Buccino, G., Binkofski, F., Fink, G. R., Fadiga, L., Fogassi, L., Gallese, V., . . . Freund, H.-J. (2001). Action observation activates premotor and parietal areas in a somatotopic manner: An fMRI study. European Journal of Neuroscience, 13, 400–404. Caggiano, V., Fogassi, L., Rizzolatti, G., Casile, A., Giese, M. A., & Thier, P. (2012). Mirror neurons encode the subjective value of an observed action. Proceedings of the National Academy of Sciences of the USA, 109, 11848–11853. doi:10.1126/science.1166818 Cannon, E. N., & Woodward, A. L. (2012). Infants generate goal-based action predictions. Developmental Science, 15, 292–298. doi:10.1111/j.1467-7687.2011.01127.x Cochin, S., Barthelemy, C., & Roux, S. (1999). Observation and execution of movement: Similarities demonstrated by quantified electroencephalography. European Journal of Neuroscience, 11, 1839–1842. Costall, A. (2013). Things that help us make what we are. In G. Sammut, P. Daanen & F. M. Mofhaddam (Eds.), Understanding the self and others explorations in intersubjectivity and interobjectivity (pp. 66–67). Abingdon, UK: Routledge. Csibra, G. (2007). Action mirroring and action interpretation: An alternative account. In P. Haggard, Y. Rossetti & M. Kawato (Eds.), Sensorimotor foundations of higher cognition. Attention and performance XXII (pp. 427–451). Oxford, UK: Oxford University Press. Daum, M. M., Prinz, W., & Aschersleben, G. (2011). Perception and production of object-related grasping in 6-month-olds. Journal of Experimental Child Psychology, 108, 810–818. doi:10.1016/j.jecp.2010.10.003 De Jaegher, H., & Froese, T. (2009). On the role of social interaction in individual agency. Adaptive Behavior, 17, 444–460. doi:10.1177/1059712309343822 Di Paolo, E. A., & De Jaegher, H. (2012). The interactive brain hypothesis. Frontiers in Human Neuroscience, 6(163), 1–16. doi:10.3389/fnhum.2012.00163/abstract Di Pellegrino, G., Fadiga, L., Fogassi, L., Gallese, V., & Rizzolatti, G. (1992). Understanding motor events: A neurophysiological study. Experimental Brain Research, 91(1), 176–180. doi:10.1007/BF00230027 Dumas, G. (2014). Towards a two-body neuroscience. Communicative & Integrative Biology, 4, 349–352. doi:10.4161/cib.4.3.15110 Dumas, G., Martinerie, J., Soussignan, R., & Nadel, J. (2012). Does the brain know who is at the origin of what in an imitative interaction? Frontiers in Human Neuroscience, 128, 1–11. doi:10.3389/ fnhum.2012.00128 Engagement beyond the mirror 11 Dumas, G., Nadel, J., Soussignan, R., Martinerie, J., & Garnero, L. (2010). Inter-brain synchronization during social interaction. PLoS ONE, 5(8), e12166. doi:10.1371/ journal.pone.0012166 Fadiga, L., Fogassi, L., Pavesi, G., & Rizzolatti, G. (1995). Motor facilitation during action observation: A magnetic stimulation study. Journal of Neurophysiology, 73, 2608–2611. Farroni, T., Csibra, G., Simion, F., & Johnson, M. H. (2002). Eye contact detection in humans from birth. Proceedings of the National Academy of Sciences of the USA, 99, 9602–9605. doi:10.1073/pnas.152159999 Farroni, T., Massaccesi, S., Menon, E., & Johnson, M. H. (2007). Direct gaze modulates face recognition in young infants. Cognition, 102, 396–404. doi:10.1016/j.cognition.2006.01.007 Farroni, T., Massaccesi, S., Pividori, D., & Johnson, M. H. (2004). Gaze following in newborns. Infancy, 5(1), 39–60. doi:10.1207/s15327078in0501_2 Friedman, M. S. (1955). Martin Buber: The life of dialogue. Chicago, IL: Chicago University Press. Gallagher, S. (2001). The practice of mind. Theory, simulation or primary interaction. Journal of Consciousness Studies, 8(5–7), 83–108. Gallagher, S. (2008). Inference or interaction: Social cognition without precursors. Philosophical Explorations, 11(3), 163–174. doi:10.1080/13869790802239227 Gallese, V., Fadiga, L., Fogassi, L., & Rizzolatti, G. (1996). Action recognition in the premotor cortex. Brain, 119, 593–610. doi:10.1093/brain/119.2.593 Gallese, V., & Goldman, A. (1998). Mirror neurons and the simulation theory of mind-reading. Trends in Cognitive Sciences, 2, 493–501. doi:10.1016/S1364-6613(98)01262-5 Gallese, V., & Sinigaglia, C. (2011). What is so special about embodied simulation? Trends in Cognitive Sciences, 15, 512–519. doi:10.1016/j.tics.2011.09.003 Goldman, A. (2006). Simulating minds: The philosophy, psychology, and neuroscience of mindreading. Philosophy of mind series. New York, NY: Oxford. Grossmann, T., Johnson, M. H., Farroni, T., & Csibra, G. (2007). Social perception in the infant brain: Gamma oscillatory activity in response to eye gaze. Social Cognitive and Affective Neuroscience, 2, 284–291. doi:10.1093/scan/nsm025 Grossmann, T., Parise, E., & Friederici, A. D. (2010). The detection of communicative signals directed at the self in infant prefrontal cortex. Frontiers in Human Neuroscience, 4, 201. doi:10.3389/fnhum.2010.00201 Hamilton, A. F. de. C. (2009). Research review: Goals, intentions and mental states: Challenges for theories of autism. Journal of Child Psychology and Psychiatry, 50, 881–892. doi:10.1111/ j.14697610.2009.02098.x Hari, R., Forss, N., Avikainen, S., Kirveskari, E., Salenius, S., & Rizzolatti, G. (1998). Activation of human primary motor cortex during action observation: A neuromagnetic study. Proceedings of the National Academy of Sciences of the USA, 95, 15061–15065. Heimann, K., & Uithol, S. (2015). Participation over observation: The roots of action understanding in attunement. Journal of Consciousness Studies, 22(1–2), 45–48. Heyes, C. M. (2010). Where do mirror neurons come from? Neuroscience and Biobehavioral Reviews, 34, 575–583. doi:10.1016/j.neubiorev.2009.11.007 Hickok, G. (2009). Eight problems for the mirror neuron theory of action understanding in monkeys and humans. Journal of Cognitive Neuroscience, 21, 1229–1243. doi:10.1162/jocn.2009.21189 Hoehl, S., Michel, C., Reid, V. M., Parise, E., & Striano, T. (2014). Eye contact during live social interaction modulates infants’ oscillatory brain activity. Social Neuroscience, 9, 300–308. doi:10.1080/17470919.2014.884982 Hunnius, S., & Bekkering, H. (2014). What are you doing? How active and observational experience shape infants’ action understanding. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 369, 1–11. doi:10.1098/rstb.2013.0490 Hurley, K. B., & Oakes, L. M. (2015). Experience and distribution of attention: Pet exposure and infants’ scanning of animal images. Journal of Cognition and Development, 16(1), 11–30. doi:10.1080/15248372.2013.833922 12 Vasudevi Reddy and Sebo Uithol Hutto, D. D. (2013). Action understanding: How low can you go? Consciousness and Cognition, 22 (3), 1142–1151. doi:10.1016/j.concog.2013.01.002 Iacoboni, M., & Dapretto, M. (2006). The mirror neuron system and the consequences of its dysfunction. Nature Reviews: Neuroscience, 7, 942–951. doi:10.1038/nrn2024 Jones, S. S. (2009). The development of imitation in infancy. Philosophical Transactions of the Royal Society B-Biological Sciences, 364, 2325–2335. doi:10.1098/rstb.2009.0045 Kadar, E. E., & Effken, J. A. (2005). From discrete actors to goal-directed actions: Toward a processbased methodology for psychology. Philosophical Psychology, 18, 353–382. doi:10.1080/ 09515080500177358 Kampe, K. K. W., Frith, C. D., & Frith, U. (2003). “Hey John”: Signals conveying communicative intention toward the self activate brain regions associated with ‘mentalizing’, regardless of modality. Journal of Neuroscience, 23, 5258–5263. Kanakogi, Y., & Itakura, S. (2011). Developmental correspondence between action prediction and motor ability in early infancy. Nature Communications, 2, 341. doi:10.1038/ncomms1342 Kanner, L. (1943). Autistic disturbances of affective contact. Nervous Child, 2, 217–250. Kourtis, D., Knoblich, G., & Sebanz, N. (2013). History of interaction and task distribution modulate action simulation. Neuropsychologia, 51, 1240–1247. doi:10.1016/j.neuropsychologia.2013. 04.001 Kourtis, D., Sebanz, N., & Knoblich, G. (2010). Favouritism in the motor system: Social interaction modulates action simulation. Biology Letters, 6, 758–761. doi:10.1098/rsbl.2010.0478 Kugiumutzakis, G. (1998). Neonatal imitation in the intersubjective companion space. In S. Braten (Ed.), Intersubjective communication and emotion in early ontogeny (pp. 63–88). Cambridge, UK: Cambridge University Press. Leavens, D. A., Sansone, J., Burfield, A., Lightfoot, S., O’Hara, S., & Todd, B. K. (2014). Putting the “Joy” in joint attention: Affective-gestural synchrony by parents who point for their babies. Frontiers in Psychology, 879, 1–11. doi:10.3389/fpsyg.2014.00879 Leudar, I., & Costall, A. (2009). Against theory of mind. Basingstoke, UK, New York, NY: Palgrave Macmillan. Lewin, K., & Cartwright, D. (1951). Field theory in social science: Selected theoretical papers. New York, NY: Harper. Markova, I. (1982). Paradigms, thought and language. Chichester, UK: Wiley. Markova, G., & Legerstee, M. (2006). Contingency, imitation, and affect sharing: Foundations of infants’ social awareness. Developmental Psychology, 42(1), 132–141. doi:10.1037/00121649.42.1.132 McGann, M., De Jaegher, H., & Di Paolo, E. A. (2013). Enaction and psychology. Review of General Psychology, 17, 203–209. doi:10.1037/a0032935 McQuaid, N., Bibok, M., & Carpendale, J. (2009). Relation between maternal contingent responsiveness and infant social expectations. Infancy, 14, 390–401. doi:10.1080/15250 000902839955 Meltzoff, A. N., & Keith Moore, M. (1994). Imitation, memory, and the representation of persons. Infant Behavior and Development, 17(1), 83–99. doi:10.1016/0163-6383(94)90024-8 Mireault, G., Poutre, M., Sargent Hier, M., Dias, C., Perdue, B., & Myrick, A. (2012). Humour perception and creation between parents and 3- to 6-month-old infants. Infant and Child Development, 21, 338–347. doi:10.1002/icd.757 Mukamel, R., Ekstrom, A., Kaplan, J., Iacoboni, M., & Fried, I. (2010). Single-neuron responses in humans during execution and observation of actions. Current Biology, 20, 750–756. doi:10.1016/j.cub.2010.02.045 Nadel, J., Carchon, I., Kervella, C., Marcelli, D., & R
eserbat Plantey, D. (1999). Expectancies for social contingency in 2-month-olds. Developmental Science, 2, 164–173. doi:10.1111/14677687.00065 Nagels, A., Kircher, T., Steines, M., & Straube, B. (2015). Feeling addressed! The role of body orientation and co-speech gesture in social communication. Human Brain Mapping, 365(5), 1965–1936. doi:10.1002/hbm.22746 Engagement beyond the mirror 13 Nagy, E., & Molnar, P. (2004). Homo imitans or homo provocans? Human imprinting model of neonatal imitation. Infant Behavior and Development, 27(1), 54–63. doi:10.1016/ j.infbeh.2003.06.004 Newman-Norlund, R. D., Van Schie, H. T., Van Zuijlen, A., & Bekkering, H. (2007). The mirror neuron system is more active during complementary compared with imitative action. Nature Neuroscience, 10, 817–818. doi:10.1038/nn1911 Oberman, L. M., Pineda, J. A., & Ramachandran, V. S. (2007). The human mirror neuron system: A link between action observation and social skills. Social Cognitive and Affective Neuroscience, 2(1), 62–66. doi:10.1093/scan/nsl022 Parise, E., Reid, V. M., Stets, M., & Striano, T. (2008). Direct eye contact influences the neural processing of objects in 5-month-old infants. Social Neuroscience, 3, 141–150. doi:10.1080/ 17470910701865458 Raczaszek-Leonardi, J. (2013). Young children’s dialogical actions: The beginnings of purposeful intersubjectivity. IEEE Transactions Autonomous Mental Development, 5, 210–221. doi:10.1109/TAMD.2013.2273258 Reddy, V. (2003). On being the object of attention: Implications for self-other consciousness. Trends in Cognitive Sciences, 7, 397–402. doi:10.1016/S1364-6613(03)00191-8 Reddy, V. (2008). How infants know minds. Cambridge, MA: Harvard University Press. Reddy, V. (2011). A gaze at grips with me. In A. Seemann (Ed.), Joint attention new developments in psychology, philosophy of mind, and social neuroscience (pp. 137–157). Cambridge, MA: MIT Press. Reddy, V. (2015). Joining intentions in infancy. Journal of Consciousness Studies, 22(1–2), 24–44. Reddy, V., Chisholm, V., Forrester, D., Conforti, M., & Maniatopoulou, D. (2007). Facing the perfect contingency: Interactions with the self at 2 and 3 months. Infant Behavior and Development, 30, 195–212. doi:10.1016/j.infbeh.2007.02.009 Reddy, V., Markova, G., & Wallot, S. (2013). Anticipatory adjustments to being picked up in infancy. PLoS ONE, 8, e65289. doi:10.1371/journal.pone.0065289 Reid, V. M., Striano, T., Kaufman, J., & Johnson, M. H. (2004). Eye gaze cueing facilitates neural processing of objects in 4-month-old infants. Neuroreport, 15, 2553–2555. doi:10.1097/ 00001756-200411150-00025 Rizzolatti, G., Fadiga, L., Gallese, V., & Fogassi, L. (1996). Premotor cortex and the recognition of motor actions. Cognitive Brain Research, 3, 131–142. doi:10.1016/0926-6410(95)00038-0 Rizzolatti, G., Fogassi, L., & Gallese, V. (2001). Neurophysiological mechanisms underlying the understanding and imitation of action. Nature Reviews: Neuroscience, 2, 661–670. doi:10. 1038/35090060 Rizzolatti, G., & Sinigaglia, C. (2010). The functional role of the parieto-frontal mirror circuit: Interpretations and misinterpretations. Nature Reviews: Neuroscience, 11, 264–274. doi:10.1038/nrn2805 Rossmanith, N., Costall, A., Reichelt, A. F., Lopez, B., & Reddy, V. (2014). Jointly structuring triadic spaces of meaning and action: Book sharing from 3 months on. Frontiers in Psychology, 5, 1390. doi:10.3389/fpsyg.2014.01390 Schilbach, L., Eickhoff, S. B., Mojzisch, A., & Vogeley, K. (2008). What’s in a smile? Neural correlates of facial embodiment during social interaction. Social Neuroscience, 3(1), 37–50. doi:10.1080/ 17470910701563228 Schilbach, L., Timmermans, B., Reddy, V., Costall, A., Bente, G., Schlicht, T., & Vogeley, K. (2013). Toward a second-person neuroscience. Behavioral and Brain Sciences, 36, 393–414. doi:10.1017/S0140525X12000660 Schilbach, L., Wohlschlaeger, A. M., Kraemer, N. C., Newen, A., Shah, N. J., Fink, G. R., & Vogeley, K. (2006). Being with virtual others: Neural correlates of social interaction. Neuropsychologia, 44, 718–730. doi:10.1016/j.neuropsychologia.2005.07.017 Schnell, K., Bluschke, S., Konradt, B., & Walter, H. (2011). Functional relations of empathy and mentalizing: An fMRI study on the neural basis of cognitive empathy. Neuroimage, 54, 1743– 1754. doi:10.1016/j.neuroimage.2010.08.024 14 Vasudevi Reddy and Sebo Uithol Schurger, A. A., & Uithol, S. (2015). Nowhere and everywhere: The causal origin of voluntary action. Review of Philosophy and Psychology. Advance online publication. doi:10.1007/s13164-0140223-2 Senju, A., & Csibra, G. (2008). Gaze following in human infants depends on communicative signals. Current Biology, 18, 668–671. doi:10.1016/j.cub.2008.03.059 Shenoy, K. V., Kaufman, M. T., Sahani, M., & Churchland, M. M. (2011). A dynamical systems view of motor preparation: Implications for neural prosthetic system design. Progress in Brain Research, 192, 33–58. doi:10.1016/B978-0-444-53355-5.00003-8 Sommerville, J. A., Woodward, A. L., & Needham, A. (2005). Action experience alters 3-month-old infants” perception of others” actions. Cognition, 96(1), B1–B11. doi:10.1016/j.cognition. 2004.07.004 Southgate, V. (2013). Does infant behaviour provide support for the mirror neuron theory of action understanding? Consciousness and Cognition, 22, 1114–1121. doi:10.1016/j.concog.2013. 04.008 Sroufe, L. A., & Wunsch, J. P. (1972). The development of laughter in the first year of life. Child Development, 43, 1326–1344. Striano, T., Kopp, F., Grossmann, T., & Reid, V. M. (2006). Eye contact influences neural processing of emotional expressions in 4-month-old infants. Social Cognitive and Affective Neuroscience, 1(2), 87–94. doi:10.1093/scan/nsl008 Thelen, E., & Smith, L. B. (1994). A dynamic systems approach to the development of cognition and action. Cambridge, MA: MIT Press. Trevarthen, C. (1977). Descriptive analyses of infant communication behaviour. In H. R. Schaffer (Ed.), Studies on mother–infant interaction (pp. 227–270). London, UK: Academic Press. Uithol, S., Burnston, D., & Haselager, W. F. G. (2014). Why we may not find intentions in the brain. Neuropsychologia, 56, 129–139. doi:10.1016/j.neuropsychologia.2014.01.010 Uithol, S., & Paulus, M. (2013). What do infants understand of others’ action? A theoretical account of early social cognition. Psychological Research, 78, 609–622. doi:10.1007/s00426-013-0519-3 Uithol, S., van Rooij, I., Bekkering, H., & Haselager, W. F. G. (2011). Understanding motor resonance. Social Neuroscience, 6, 388–397. doi:10.1080/17470919.2011.559129 Uzgiris, I. C. (1981). Two functions of imitation during infancy. International Journal of Behavioral Development, 4(1), 1–12. doi:10.1177/016502548100400101 Woodward, A. L. (1998). Infants selectively encode the goal object of an actor’s reach. Cognition, 69(1), 1–34. doi:10.1016/S0010-0277(98)00058-4 Woodward, A. L., & Gerson, S. A. (2014). Mirroring and the development of action understanding. Philosophical Transactions of the Royal Society B-Biological Sciences, 369, 20130181. doi:10.1098/rstb.2013.0181 Woodward, A. L., & Guajardo, J. J. (2002). Infants’ understanding of the point gesture as an objectdirected action. Cognitive Development, 17(1), 1061–1084. doi:10.1016/S0885-2014(02) 00074-6 Yu, C., & Smith, L. B. (2012). Embodied attention and word learning by toddlers. Cognition, 125, 244–262. doi:10.1016/j.cognition.2012.06.016 Received 15 March 2015; revised version received 26 June 2015