Show Summary Details
Page of

Attachment, affect-regulation, and mentalization: the developmental origins of the representational affective self 

Attachment, affect-regulation, and mentalization: the developmental origins of the representational affective self
Attachment, affect-regulation, and mentalization: the developmental origins of the representational affective self

György Gergely

and Zsolt Unoka

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 15 August 2018

In this chapter we propose an evolutionary-based social cognitive theory of the early development of the representational affective self in humans and its role in emotional self-regulation and control. We first identify species-unique properties of human caregiver–infant interactions and critically discuss alternative proposals concerning the functional role of the human attachment system in the development of mentalizing on the one hand, and emotional self-awareness and affective self-control on the other. We propose that the human-specific features of early caregiver–infant interactions provide the necessary input conditions for specialized representation-building and attention socialization mechanisms (such as contingency detection, social biofeedback, and natural ‘pedagogy’). These mechanisms play a crucial role in establishing primary self–other affective relationship representations that capture the characteristic causal structure of contingent reactivity of early attachment relationships. They also set up cognitively accessible second-order emotion representations with associative links to the self's procedurally represented prewired basic emotions. We then characterize different levels of primary and secondary emotion-control systems and show how the developmentally established affective self–other representations subserve these emotion-regulative mechanisms. Finally, we argue that the socially constructed second-order emotion representations support the ontogenetic extension of the domain of mentalizing to include the self's own causal mental states (apart from those of others) and show how this enables the functional use of mentalization for the purposes of emotional self-regulation and control in affectively charged interactions and relationships in humans.

11.1 Species-unique properties of human attachment interactions and their functions: challenging current views on the development of mentalizing

Humans are just one of the many social animal species with an attachment instinct system where the quality of early caregiving (Harlow 1961; Bowlby 1969; Hofer 1995; Suomi 1995, 1999; Polan and Hofer 1999) has formative effects on the organism's later capacity to function adaptively in affiliative relationships. In contrast, humans may be unique in their ability to rely on a specialized inferential and representational system for mentalizing, or ‘theory of mind’ (ToM) (Dennett 1978, 1987; Baron-Cohen et al. 1985; Wellman 1990; Perner 1991; Fodor 1992; Baron-Cohen 1995; Leslie 2000) that has evolved to predict, interpret, and manipulate others’ actions by inferring and attributing causal intentional mind states (such as desires, intentions, and beliefs) to them.

Arguably, attachment and mentalizing are two independent adaptations selected to serve qualitatively different evolutionary functions. This is suggested, for example, by the fact that many social species, such as rodents or rhesus monkeys (Hofer 1995; Polan and Hofer 1999; Suomi 1999), that apparently lack the capacity for mentalizing, nevertheless possess an innate infant–caregiver attachment system. According to Bowlby's (1969) original proposal, the basic evolutionary function of the attachment instinct is to provide a specialized interactive mechanism to ensure predator evasion for the immature offspring through a prewired goal-directed signal system serving proximity seeking and maintenance between caregiver and infant. This function certainly seems independent and qualitatively different from the primary function served by the specialized social cognitive adaptation for mentalizing, which is to predict and interpret the actions of conspecifics during adult competitive or cooperative situations through attributing causal mental states to them.

However, it is noteworthy that the structural organization of early attachment interactions in humans has become qualitatively enriched by species-unique evolved features that are conspicuously absent from mother–infant attachment interactions of other social species (including our closest primate relatives). Thus human mother–infant interactions show a unique ‘proto-conversational’ turn-taking contingency structure (Brazelton et al. 1974; Trevarthen 1979; Brazelton and Tronick 1980; Stern 1985; Sander 1988; Tronick 1989; Tronick and Cohn 1989; Trevarthen and Aitken 2001). Human infants exhibit early sensitivity and preference for highly response-contingent stimulus events characteristic of the interactive style of infant-attuned social partners (Watson 1972, 1985, 1994; Bahrick and Watson 1985; Lewis et al. 1990; Rochat and Morgan 1995; Bigelow 1999; Bigelow and De Coste 2003; Bigelow and Rochat 2006). They also show special sensitivity to so-called ostensive-communicative cues (Csibra and Gergely 2006) such as eye contact (Farroni et al. 2002), infant-directed speech or ‘motherese’ (Fernald 1985, 1992; Cooper and Aslin 1990), and contingent reactivity (Watson 1994, 1995, 2001; Floccia et al. 1997; Johnson et al. 1998; Movellan and Watson 2002). Young human infants spontaneously attend to and gaze-follow others’ referential cues (such as gaze shift), but only if these are preceded by direct eye contact or infant-contingent reactivity (Johnson et al. 1998; Farroni et al. 2002; Movellan and Watson 2002; reviewed by Csibra and Gergely 2006). Unlike apes, by 12 months humans show a communicative and referential understanding and use of pointing in triadic joint attention interactions (Liszkowski et al. 2004; Behne et al. 2005; Southgate et al. 2007b; Tomasello et al. 2007). Human mother–infant interactions also involve frequent exchanges of an increasingly differentiated repertoire of basic emotion expressions (Izard and Malatesta 1987; Cohn and Tronick 1988; Malatesta et al. 1989; Tronick 1989; Camras 1992; Gergely and Watson 1996, 1999; Sroufe 1996; Gergely 2002, 2007a; Bennett et al. 2004, 2005). This contrasts with the small range and low frequency of emotion displays in non-human infant–caregiver attachment interactions. Even in primates mother-directed affect displays by infants are rare and restricted to expressions of distress and protest triggered by loss of proximity (e.g. De Marco and Visalberghi 2007, Ferrari et al. 2006). Empathic affect-mirroring emotion displays by the caregiver seem unique to human mother–infant interactions (Cohn and Tronick 1988; Malatesta et al. 1989; Tronick 1989; Uzgiris et al. 1989; Gergely and Watson 1996, 1999; Bigelow 1999; Gergely 2004, 2007a; Fonagy et al. 2002, 2007). In this chapter we will explore the question of whether these apparently species-unique properties of human attachment interactions are related functionally or developmentally to the emergence of the capacity to mentalize, or whether they serve other basic functions that are independent of, or at least not directly related to, mindreading.

In recent years several attachment theorists and developmental psychologists have put forward different versions of the general hypothesis that, in the case of humans, there is an inherent causal and functional link between the quality of early infant attachment on the one hand, and the development of the ability for mentalizing on the other. In particular, the acquisition of mindreading skills by about 4 years of age, as indicated by passing the standard verbal false-belief attribution tasks (Wimmer and Perner 1983; Wellman et al. 2001), has been suggested to be facilitated by the security of early attachment relationships (e.g. Fonagy and Target 1997; Fonagy et al. 1997a,b; Meins et al. 1998, 2003; Hobson 2002). It has been proposed that secure attachment and early parental interactive behaviours which correlate with infant attachment security at 12 months, such as maternal mind-mindedness at 6 months (Meins et al. 2001), play an important causal role in fostering the development of mentalizing as they facilitate the early appearance of explicit mindreading skills in preschoolers (e.g. Fonagy and Target 1997; Fonagy et al. 1997a,b; Meins et al. 1998, 2002, 2003; Fonagy 2001). The supporting evidence consists mainly of correlational findings showing that secure infant attachment at 1 year (as well as its earlier correlates such as maternal mind-mindedness at 6 months) predict earlier ability to pass the standard verbal false-belief tasks at around 4 years (e.g. Fonagy and Target 1997; Fonagy et al. 1997a,b; Meins et al. 1998, 2003; de Rosnay and Harris 2002; Ontai and Thompson 2002).

However, there are a number of empirical and theoretical reasons to question the strong form of the hypothesis which postulates a specific evolutionary role and/or a direct ontogenetic causal developmental and functional link between security of infant attachment and its early interactive predictors on the one hand, and the development of explicit mindreading abilities in later childhood on the other. First, one problem lies in the fact that the correlational findings linking infant attachment security during the first year to earlier success in passing the explicit ToM tasks have not been supplemented with specific causal developmental models. Ideally, such models should spell out the particular ways in which the secure quality of preverbal attachment and/or factors such as parental talk involving mental state attributions to the baby at 6 months exert their facilitating effects on the acquisition of the cognitive skills involved in inferring, attributing, representing, and reasoning about the epistemic mind states of others (such as beliefs). Numerous plausible causal linking factors have been suggested in this regard which, however, exert their effect later in development. These include the frequency of perspective taking in caregiver–child linguistic exchanges (e.g. Harris 2005; Lohmann et al. 2005; de Rosnay and Hughes 2006), the amount of complementary role play in the family, the degree of family talk about conflicting emotions and differential conflict- and emotion-regulation strategies (Dunn et al. 1991; Sabbagh and Callanan 1998; Cutting and Dunn 1999; Ruffman et al. 1999; Vinden 2001), and the number of references to thoughts and beliefs and the relationship specificity of children's real-life accounts of negative emotions (Hughes and Dunn 2002). The independent effects of these (and other) factors on the earlier appearance of explicit ToM performance around 4 years are well documented (reviewed by Fonagy et al. 2007). Other facilitating factors that are unlikely to be related to infant attachment security include having older siblings (Perner et al. 1994; Ruffman et al. 1998), the relative frequency and precocity of using complex syntactic complement structures (Astington and Jenkins 1999; de Villiers and de Villiers 2000, de Villiers and Plyers 2002), and the maturational development of executive function, selective attention, and response inhibition capacities (Leslie 2000; Perner and Lang 2000; Leslie et al. 2004).

Secondly, the demonstrated facilitating effects of early secure attachment on ToM development are not strong. Moreover, in several studies they only show up in ToM tasks which involve reasoning about both beliefs and emotions (such as belief–desire tasks) in contrast with tasks probing purely epistemic states such as false beliefs (e.g. Fonagy et al. 1997a; reviewed by Fonagy et al. 2007). Furthermore, there are strong recent arguments and supporting evidence showing that the standard use of explicit verbal false-belief tasks (Wellman et al. 2001) to diagnose the developmental achievement of ‘full’ mindreading skills around age 4 involves several complex linguistic, pragmatic interpretational, and executive function requirements. Thus it is possible that young children have difficulty passing false-belief tasks not because they still lack the capacity to mentalize (or at least some components of ToM), but rather because of deficits with some of the non-ToM-related component skills that these tasks require (Bloom and German 2000; Leslie et al. 2004). This may have seriously distorted the validity of the generally accepted developmental timetable for the ontogenetic establishment of mentalizing abilities. For these reasons, some researchers have recently argued explicitly for the abandonment of the standard false-belief tasks as a general diagnostic tool for establishing the attainment of ToM skills (e.g. Bloom and German 2000).

Complementary to these methodologically based doubts, there is converging recent evidence (based on paradigms not plagued with the methodological problems outlined above) demonstrating mentalization much earlier in human ontogeny. For example, Tomasello and his colleagues have shown that infants exhibit a variety of ToM abilities as early as 12 months. These include spontaneously monitoring others’ perceptual access to transformations of relevant aspects of reality, inferring and attributing to them the corresponding epistemic mental states of informedness or ignorance, and relying on such mental attributions to modulate one's own goal-directed behaviours in the presence of the other (e.g. Tomasello and Haberl 2003; Tomasello et al. 2005). Similarly, there is convergent new evidence based on implicit non-verbal versions of the standard false-belief tasks using violation-of-expectation looking-time paradigms which show that infants as young as 13 months of age monitor others’ perceptual access to reality, infer, attribute, and represent their true or false beliefs about the situation accordingly, and predict the other's subsequent behaviour by relying on the representational content of such attributed mental states (Onishi and Baillargeon 2005; Csibra and Southgate 2006; Onishi et al. 2007; Southgate et al. 2007a; Surian et al. 2007).

Finally, recent evidence from comparative animal literature calls into doubt the notion that the basic ability for implicit mindreading is an exclusively human-specific capacity. These studies suggest that some rudimentary and possibly highly domain-restricted forms of mentalizing seem to have also evolved independently and convergently in a number of non-human social species, such as apes, goats, and avian species including crows, ravens, and scrab-jays (Hare et al. 2001; Tomasello et al. 2003; Emery and Clayton 2004; Bugnyar and Heinrich 2005; Kaminski et al. 2006).

Based on this brief review, we argue that the currently popular view endorsed by a number of attachment theorist and infant researchers which assumes a (possibly evolutionary-based and human-specific) direct causal and functional link between early security of infant attachment on the one hand, and the acquisition of explicit mentalizing skills on the other, must be significantly revised. The relevant developmental and comparative evidence of recent years seems more compatible with the alternative view that an implicit and automatic capacity for mentalizing about others is not a developmental achievement, but an innate social cognitive evolutionary adaptation implemented by a specialized and prewired mindreading mechanism which seems to be active and functional at least as early as 12 months in humans. Furthermore, this capacity for implicit mentalizing is unlikely to be a uniquely human competence as, in its basic form, it also seems to have evolved independently and convergently in a number of non-human social species which live in highly competitive niches.

The above does not imply that certain types of early interactive patterns characteristic of specific attachment relationships of human infants would have no developmental effect whatsoever on later ability to use the capacity for mentalizing functionally and efficiently in coping with interpersonal interactions and relationships. There are, in fact, good empirical reasons to believe that certain dysfunctional types of early attachment relations, involving severe neglect, abuse, or dissociative, highly intrusive, or grossly unpredictable patterns of parental reactivity, have significant and long-term detrimental and disruptive effects on later capacity to use the innate competence for on-line mentalization functionally as an adaptive interpersonal coping strategy (e.g. Fonagy et al. 2000, 2002, 2003, 2007; Gergely et al. 2002b).

In this chapter we advance the argument that the specialized capacity for mentalization is likely to be an independent social cognitive adaptation whose primary evolutionary function is separate and unrelated to the basic evolutionary function for which the attachment instinct system has been selected. If so, it would not be surprising if security of early infant attachment in humans did not turn out to play a direct causal functional role in the ontogenetic development of our mentalizing capacity. Nevertheless, we propose that when maladaptive patterns of affective parental reactivity become the dominant features of the recurring interactive structure of the infant's primary attachment relationships, they can and do play a significant causal role in pathologically undermining the developing self's potential to rely on its innate mentalizing capacity as its dominant social cognitive strategy to cope with interpersonal situations and intimate and affiliative relationships during later life. Elsewhere we have argued, and analysed a specific clinical example as illustration (Gergely and Unoka 2007), that the ensuing inhibitory or distorting effects on the self's later capacity to mentalize stem from the continued influence that the self's non-conscious and procedurally represented primary affective self–other relationship representations exert on perceiving, anticipating, and interpreting the behavioural and affective reactions of interactive partners. In the rest of this chapter, we describe an evolutionary-based social developmental model about the structural nature and ontogenetic origins of such primary affective relationship representations. In particular, we shall specify the role of representation-building and attention-socialization mechanisms which result in the early establishment of such affective self–other representations that capture the causal properties of the recurring patterns of specific attachment interactions during early development.

11.2 The social construction of the representational affective self in humans

11.2.1 The ‘constitutional self’: the evolutionary function of basic emotions, their modular organization, and prewired causal properties

We assume that the infant's innate ‘constitutional self’ (Gergely and Watson 1996, 1999; Fonagy et al. 2002, 2007; Gergely 2007a) has a rich prewired affective structural organization. Apart from genetically based individual temperamental differences (Rothbart 1989; Kagan 1994), we assume, in line with current evolutionary theories of emotions (Ekman et al. 1972; Izard and Malatesta 1987; Ekman 1992; Tomkins 1995; see also Darwin 1872), that the constitutional self also contains innate specifications of the core physiological, motor, and functional components that make up the organizational structure of basic emotion programs (such as fear, anger, joy/interest, disgust, and sadness). Current cross-cultural and experimental research has identified about six basic categorical emotions that are characterized by specific and differential patterns of physiological bodily arousal states on the one hand and/or motor programs of emotion expressive facial display patterns on the other. These are universally recognized and cross-culturally shared by adult humans around the world (Ekman et al. 1972; Ekman 1992). The basic emotions are types of adaptations that originate from similar automatic emotional coping systems present in several non-human social animal species such as our primate ancestors (Panksepp 1998; Suomi 1999).

Basic emotion programs are prewired stimulus-driven and procedurally represented behavioural automatisms (Ekman 1992). They have evolved to generate rapid and adequate automatic coping responses (e.g. fight or flight) to deal with the specific types of environmental challenges that their causal triggering conditions represent (e.g. predator threat, territorial conflicts, mating competition, etc.). Such largely invariant, stereotypic, and recurring environmental stimulus conditions became represented through evolution as innately specified causal inputs which automatically activate the basic emotions. These primary emotion programs consist of two major structural components: (a) a specific pattern of physiological bodily arousal state, and (b) prewired emotion-specific motor routines. The latter component consists of two types of fixed behavioural automatisms: (i) stereotypic action tendencies (approach/avoidance, fight/flight, etc.) and (ii) expressive emotion-specific fixed facial–vocal display patterns and body postures. The evolutionary function of both components is to provide specific coping responses with the built-in goal to change (i.e. to eliminate, modify, or maintain) the environmental triggering conditions adaptively so that the challenges represented by these triggers are altered and thereby resolved.

In non-human animal species the activation of basic emotions is largely stimulus-driven and automatic, and under the external control of pre-specified inputs. Furthermore, the range of causal consequences that the execution of their fixed behavioural components (e.g. attack, threat displays, courting behaviour, distress calls) can bring about in the animal's environment is also highly restricted to a small set of stereotypic outcomes. This is so because the scope of their social environmental effects is restricted to the equally small number of innately pre-specified stereotypic reactions that they can induce in other conspecifics (e.g. flight/fight, submission, acceptance/refusal of mating approaches). Therefore the ensuing stereotyped emotional interactions among conspecifics transform the original causal environmental emotion-triggering conditions in a relatively small number of largely invariant ways. The consequent environmental changes then induce either automatic termination of the organism's basic emotional arousal, or inhibition of its behavioural expressions (by modifying its original triggering conditions), or activation of a different basic emotion (by establishing new environmental triggering conditions).

Clearly, this type of innately specified, automatic, and stimulus-controlled modular organization of basic emotions can fulfill its primary function described above only in evolutionary niches that are characterized by stable, recurring, and largely invariant social environmental conditions. Indeed, the specialized and modularly organized behavioural coping systems provided by the basic emotion programs must have evolved under the selective pressure that such invariant and stable conditions of the species’ evolutionary environment represented.

11.2.2 The emergence of context-sensitive modifiability of the control mechanisms of basic emotion programs in humans

Such highly stable evolutionary conditions have ceased to characterize the increasingly complex and changing cultural, social, and technological environment of early humans. In human social interactions the same expressions of a basic emotion can evoke largely variable responses from different individuals as a function of their personality traits, social status, current dispositions, informedness about relevant aspects of the situation, and cognitive evaluation of the significance of the emotion-inducing context. During ontogenetic development, individuals can also acquire variably large and flexible repertoires of non-stereotypic emotional response alternatives whose context-sensitive selective activation comes to be, at least partially and some of the time, under their cognitive voluntary control. This results in large variability and significant individual differences in the range and types of reactions that the automatic behavioural expressions of a conspecific's basic emotional arousal can induce in others under different situational and social contexts.

We suggest that the evolutionary emergence of such significantly growing variability and increase in the types of relevant social environmental conditions must have co-evolved with correspondingly significant modifications in the inhibitory and response selective control mechanisms of basic emotion programs in humans to ensure their continued role as functional coping systems. For the basic emotion programs to remain adaptive in the face of such conditions of situational and individually variable social environmental reactivity, their control systems must have evolved to incorporate mechanisms (a) to evaluate, anticipate, and represent the variable causal consequences that specific emotional reactions are likely to result in under different interpersonal and situational contexts, and (b) to selectively inhibit or flexibly modify their automatic emotion-induced behaviours and expressive displays as a function of their anticipated situation-specific causal effects.

11.2.3 Early maternal emotion-regulation of the infant's primary affective states within the human attachment system

In early phases of ontogenetic development the affective behaviour of young infants provides no indication of the availability of such flexible and context-sensitive emotion-control mechanisms. Infants seem largely unable to modulate or inhibit the automatic motor expressions of their basic emotional arousal states. Their affective state regulation is largely under the external control of the attachment figure's adequate reactions (Tronick 1989; Gergely and Watson 1996, 1999; Sroufe 1996; Fonagy et al. 2002, 2007; Gergely 2007a). If the caregiver can sensitively monitor and correctly interpret the infant's automatic affective state signals, she can react to them with adequate contingent emotional expressions or instrumental coping responses to regulate the baby's emotional arousal. This can be achieved by at least two different mechanisms: first, through the prewired direct physiological state-modulating effects of certain maternal attachment reactions, for example through the innate soothing effect that gentle bodily contact or the specific intonation pattern of ‘motherese’ induces in babies (Fernald 1985, 1992; Cooper and Aslin 1990; Hofer 1995; Polan and Hofer 1999); secondly, through instrumental actions transforming the baby's proximal environment, thereby appropriately changing the external conditions that have triggered (or would continue to maintain) the infant's emotional arousal.

By this process the infant's basic emotion programs start to become modified so that the young child is increasingly able to anticipate and cope with the characteristic situational variability of the habitual causal consequences to which his/her specific emotional reactions typically lead. This requires the establishment of experience-based representations of the expected causal consequences that different emotion responses tend to induce from particular attachment figures under specific situational contexts. However, it should be noted that, to enable the construction of such context-sensitive emotion representations during different person- and situation-specific affective interactions, the infant must possess a suitable event-monitoring and representation-building mechanism which can register, analyse, and represent the characteristic causal dependencies between the self's particular emotion responses and the specific reactions they invoke in others.

11.2.4 The causal contingency detection mechanism and its role in building context-specific primary emotion representations

Elsewhere we have proposed that human infants are, indeed, equipped with a specialized innate information-processing and representation-building device, the so-called ‘contingency detection mechanism’ (Watson 1994; Gergely and Watson 1996, 1999) which, when applied to the domain of emotional interactions, can construct the required type of experience-based and context-sensitive primary emotion representations. This mechanism is specialized for the detection and representation of causal contingency relations between the infant's specific responses and consequent environmental outcomes by registering the degree of causal dependency (in terms of conditional probability of co-occurrence over time) between responses and stimulus outcomes within particular situational contexts; for technical details and reviews of supporting evidence see Watson (1985, 1994, 1995, 2001), Gergely and Watson (1996, 1999), and Gergely (2002, 2004, 2007a). One basic developmental function of the contingency detection device is to identify and represent the degree of the infant's causal control over different responses of the social environment and the specific contextual factors modulating these causal effects (Watson 1995, 2001). The resulting experience-based primary representations of the degree of causal efficacy of the infant's responses form the basis of the young child's developing primary sense of self-efficacy and social self-agency (Gergely and Watson 1996; Gergely 2002; 2007a).

We shall not describe the details of the structural and functional characteristics of this specialized causal representation-building mechanism here; the interested reader is referred to Watson (1985, 1994, 1995; 2001), Gergely and Watson (1996, 1999), or Fonagy et al. (2002). For our present purposes we call attention to only one of its crucial properties: To discover and represent the actual degree of causal control that a specific infant behaviour has over a particular response by the caregiver requires the monitoring, assessment, and representation of two different (and independent) aspects of the causal dependency relation that are obtained between these two target events over time. In other words, the contingency detection mechanism must construct two separate representations. The first must monitor the occurrences of a specific target behaviour of the self and represent the degree of causal likelihood that it is followed by a particular monitored response of the other. The second must monitor the occurrences of the target response of the other and represent the degree of causal likelihood that it was preceded by the self's target behaviour that was monitored. Only the aggregate of these two values can provide a correct estimate and representation of the actual degree of causal relatedness of the two target events monitored (Watson 1994; Gergely and Watson 1996).

Let us clarify this point with an example (cf. Watson 1995). Imagine a frightened boy whose negative emotional arousal automatically activates a fear-expressive facial display towards his mother. Assume that she promptly responds to the infant's emotion-expression by picking him up. By monitoring the causal structure of this specific type of attachment interaction over recurring episodes, the infant's contingency detection mechanism will register that his fear-expressive facial–vocal display results in being picked up by his mother, say, 80 per cent of the time. On this basis alone, it would seem safe to conclude that the degree of causal efficacy of this specific attachment behaviour in inducing the desired maternal response is, in fact, quite high. But would this provide a correct estimate of the actual degree of the infant's causal control over the mother's specific response? Not necessarily. Imagine that this mother is over-controlling, or physically intrusive, or has an anxious, worried, and fearful disposition (projecting these emotions onto her baby), or is emotionally unstable and insecure with a constant need for physical proximity herself. Any of these maternal dispositions can result in highly frequent occurrences of picking her baby up even when this was not preceded (and evoked) by her infant's fear-expression. To take into account such occasions when the maternal response was not brought about by the infant's fear-expressive behaviour, the contingency detection mechanism needs to, and does (see Watson 1985, 1994; Gergely and Watson 1996, 1999), separately monitor the occurrences of the maternal target behaviour, checking the conditional probability with which her reaction was actually preceded (and evoked) by the infant's fear-expressive display. Assume that this second type of assessment indicates that only 20 per cent of the mother's target behaviours were, in fact, preceded by her infant's fear-expression. Clearly, under such circumstances it would be grossly misleading to represent the actual degree of causal control that the infant's fear-expression has over being picked up by his mother as being very high.

In fact, as the evidence indicates (e.g. Watson 1985, 1994), the design structure of the contingency detection mechanism is sensitive to this problem and solves it by separately monitoring and representing the two types of target events and their respective degrees of (prospective and retrospective) causal dependency in relation to each other. This results in setting up two separate representations for the two behavioural events: one representing the monitored target behaviour of the infant and its degree of causal efficacy in bringing about the particular target response of the other (the ‘sufficiency index’), and one representing the monitored target response of the other and its degree of causal dependence on the previous occurrence of the specific target behaviour of the infant (the ‘necessity index’) (Watson 1994; Gergely and Watson 1996).

11.2.5 Two types of primary emotion representation of early affective attachment interactions and their causal properties: self-to-other versus other-to-self emotion schemes

Let us examine what kind of experience-based primary emotion representations are constructed when the contingency detection mechanism is applied to recurring affective infant–caregiver attachment interactions. There are three features of the resulting primary affective self–other representations that we think are of central importance because of the functional role they will play in supporting the infant's emerging capacity for emotional self-regulation and control during affectively charged attachment interactions. First, they will represent crucial information about the characteristic causal properties of the specific emotion responses that are monitored (representing both their causal triggering conditions and their expectable consequences). Secondly, the characteristic profile of these causal conditions will be separately represented for different situational and person-contexts in which the specific target responses tend to occur. Thirdly, separate primary emotion schemes will be established to represent the causal and situational properties of the self's own emotion responses on the one hand, and those of the monitored target responses of different caregivers on the other.

Clearly, it can be highly adaptive for the infant to actively monitor and separately represent the causal and situational conditions under which different caregivers tend to produce specific types of responses. This is particularly relevant when such target behaviours of the other represent significant challenges for the self's physical well-being and emotional state of arousal. Such representations provide vital information for the infant to anticipate emotionally highly charged and potentially self-endangering interactions with particular caregivers in specific attachment situations and therefore can facilitate the generation of adaptive anticipatory coping reactions to deal with them.

Thus we propose that the contingency detection mechanism creates two basic kinds of experience-based primary emotion representations that capture the causal properties of recurring types of affective interactions with particular attachment figures under specific situational contexts.

  • Self-to-Other causal emotion schemes These represent two types of predictable causal consequences that a specific emotion reaction is likely to lead to when performed in the presence of a particular attachment figure in a given situational context:

    1. (i) the relative likelihood that the infant's emotion-expression will induce a particular response by the caregiver in the situation;

    2. (ii) the probable consequences that the infant-induced reaction of the caregiver will give rise to in that situation, i.e. the ways in which it is likely to modify the environmental conditions that had triggered the emotional reaction in the first place.

  • Other-to-Self causal emotion schemes These represent the characteristic causal properties of specific target behaviours that are likely to be performed by the caregiver under certain attachment contexts. These behaviours of specific caregivers become active targets of monitoring either because they belong to the likely consequences of some of the infant's own monitored emotion-expressive responses (by her Self-to-Other emotion schemes), or because their occurrence is likely to trigger a basic emotional reaction in the infant (thus representing a significant environmental challenge for the self).

Such Other-to-Self emotion schemes will include causal information about the types of environmental triggering conditions that are likely to induce (and therefore can be used to anticipate) the other's monitored target behaviours. Importantly, such causal conditions may also include the infant's representation of his/her own specific emotion-expressive responses as having triggered the caregiver's monitored reactions in the past during particular situations and with differential likelihoods. In such cases the infant can use this represented causal information to control, modify, or pre-empt the occurrence of the other's anticipated target reaction by inhibiting its potential cause, i.e. by suppressing or modifying the behavioural expression of his/her own emotion reactions that are represented as likely to induce the other's, potentially self-endangering, target behaviour whose occurrence the infant is trying to avoid.

11.2.6 Primary response inhibition and substitution mechanisms of automatic emotional self-control

Two basic types of emotion control processes (involving primary automatic versus secondary cognitive control mechanisms) serve affective self-regulation and coping in humans during emotionally charged interactions. We discuss primary automatic response inhibition first.

In our view, the basic developmental function of the infant's experience-based primary emotion schemes is to provide him/her with representational means to anticipate and adaptively influence the course of affective attachment interactions with particular caregivers in the service of emotional self-control and coping. By extracting and representing the context- and person-specific causal patterns of affective interactions, these primary representations equip the infant with a repertoire of specific emotion schemes which can be used to anticipate, and to some degree actively modulate, the emotional consequences that particular types of attachment interactions will bring about.

There are two types of mechanism through which the infant's primary emotion control system can utilize the representational contents of the Self-to-Other and Other-to-Self emotion schemes for these purposes. The first is a primary response inhibitory control mechanism which can automatically suppress the behavioural expression of a basic emotional arousal state. One condition that leads to the triggering of such a response-inhibitory process is when the activation of the self's basic emotion reaction primes a specific Self-to-Other emotion scheme which represents the behavioural expression of the activated emotion state as likely to lead to severe negative consequences for the self in the given attachment context. In such cases, the anticipatory activation of the represented negative consequences exerts a direct and automatic inhibitory effect, blocking the impending motor expression of the activated emotion and thereby avoiding the realization of the negative consequences anticipated. This type of primary response-inhibitory emotion control mechanism is automatic and procedural in nature.

Evidence from attachment research (Cassidy and Shaver 1999) indicates that this primary response-inhibitory coping mechanism becomes operational rather early. An example is the separation-induced response pattern observed in the Strange Situation Test (Ainsworth et al. 1978) in a subgroup of 1-year-olds who show avoidant attachment to their caregiver. Unlike other (securely or insecurely attached) infants, avoidantly attached babies typically do not exhibit the behavioural-expressive signs of distress or protest that would indicate negative physiological arousal induced by separation from their caregiver (Ainsworth et al. 1978; Goldberg et al. 1995; Cassidy and Shaver 1999). Despite this apparent lack of emotion-expressive reactivity to separation, physiological measures (heart rate and cortisol elevation) indicate that, similarly to infants belonging to other attachment categories, separation from the caregiver induces clear bodily indices of high negative emotional arousal and stress in such infants (e.g. Spangler and Grossmann 1993; Cassidy and Shaver 1999).

Importantly, this automatic inhibition of the expression of separation-induced distress is relationship specific; no strong correlation has been found between the types of attachment infants exhibit towards different caregivers (reviewed by Cassidy and Shaver 1999; Fonagy 2001). There is evidence (Ainsworth et al. 1978; Cassidy and Shaver 1999) that the establishment in avoidant infants of the primary response-inhibitory coping strategy towards a particular caregiver is related to (and is best explained as an adaptation to cope with) the predictable and systematically negative emotional reactivity habitually exhibited by the caregiver in response to the infant's expressions of negative affect across situations (Watson 2001).

Note that the same automatic response-inhibitory control mechanism can also be triggered by the activation of some of the infant's Other-to-Self emotion schemes which represent the causal properties of those target behaviours of a given caregiver that are likely to induce negative emotional consequences for the infant in the given attachment context. This can happen when the situation triggers in the baby a basic emotional arousal whose behavioural expression is represented in the infant's Other-to-Self scheme as one of the potential causes that are likely to invoke the self-endangering target behaviour of the caregiver. In such cases the infant's emotional arousal activates the Other-to-Self scheme in question, leading to the anticipatory evocation of its representational content: the potential occurrence of the caregiver's target behaviour and its expectable negative consequences. This will trigger the automatic response-inhibition control mechanism to suppress the expression of the infant's basic emotion arousal and thereby avoid the realization of its anticipated negative consequences by blocking their causal triggering event from occurring.

However, it should be pointed out that, when exercised in isolation, this kind of automatic response-inhibitory coping reaction is of rather limited adaptive value for the infant's emotion regulation for two reasons. First, while it automatically suppresses the expression of the activated emotion, it does so without affecting the physiological arousal reaction itself. One of the undesirable consequences is that while the inhibition of the emotion expression removes the source of likely further escalation of the infant's emotional reaction, the negative physiological arousal already activated remains unmodified and unmodulated. Secondly, the inhibition of the emotion expression also leaves unchanged the environmental triggering conditions that led to the activation of the basic emotion reaction in the first place. Therefore the infant's emotional arousal will be maintained as long as its environmental triggering source continues to be present. In short, such an automatic inhibitory ‘freezing’ of the emotion response system can leave the self helplessly overwhelmed by its uncontrolled and continued state of heightened physiological arousal and stress which, if occurring frequently, can lead to deleterious long-term toxic consequences, resulting in the development of a rigid and dysfunctional organization of the physiological stress-regulation system (Francis and Meaney 1999; Francis et al. 1999; Fonagy et al. 2002; Pruessner et al. 2004; Wismer Fries et al. 2005).

The danger of toxic consequences can be coped with only by applying the second type of primary emotion control mechanism which involves response substitution. This mechanism exploits the possibility provided by the representational contents of primary Self-to-Other and Other-to-Self emotion schemes to automatically activate a suitable substitute behaviour which is represented as having a high likelihood to change the environmental conditions that had activated the infant's original basic emotional arousal in the first place. Clearly, this can be accomplished only if the infant possesses a sufficiently rich and differentiated repertoire of response alternatives suitable to serve as adequate substitute responses in the given situation.

11.2.7 Secondary cognitive appraisal mechanisms and top-down voluntary processes of emotional self-control

We have argued that the automatic primary emotion control mechanisms, when applied over a sufficiently differentiated repertoire of primary emotion representations, allow the infant to exert a certain measure of adaptive behavioural control in the service of emotional self-regulation during early affective interactions. However, this primary emotional coping system is severely limited in both its scope and its flexibility. This is because the Self-to-Other and Other-to-Self emotion schemes are stimulus-driven and procedurally represented emotion representations which are automatically activated whenever the stimulus situation is sufficiently similar to their represented environmental triggering conditions or their associated situational and person-contexts. By necessity, this restricts their scope of adaptive applicability to the narrow domain of the recurring types of affective attachment interactions with caregivers, whose features are represented by the primary emotion schemes in the first place.

However, as children develop, they encounter an increasing variety of new interpersonal contexts of affective interactions, often involving unfamiliar persons and situations. This results in two basic problems of adaptation that the primary emotion representations and their automatic emotion control mechanisms cannot satisfactorily cope with and that necessitate the establishment of an additional type of secondary top-down cognitive mechanism of emotion control (e.g. Lazarus 1991; Lazarus and Lazarus 1994; Thompson 1994; LeDoux 2000; Posner and Rothbart 2000; Ellsworth and Scherer 2003; Ochsner and Gross 2005).

The first problem arises when the new interactive situations encountered represent significant challenges for the self that would call for adaptive emotional coping responses, but which, because of their unfamiliarity and lack of sufficient similarity to the triggering conditions represented in the child's primary emotion schemes, do not automatically trigger a basic emotion reaction. To recognize the realistic nature of the challenge that such situations represent for the self (and consequently to generate appropriate emotional coping reactions to them), the child has to rely on secondary cognitive appraisal mechanisms. By having access to an increasingly large pool of relevant knowledge and being able to derive relevant inferences from situational cues, these top-down cognitive appraisal processes can generate realistic evaluations of the significance that these new types of input condition represent for the self. The interpretation that such secondary cognitive appraisals provide can then be used by top-down executive action control systems to activate relevant emotion responses to cope with the situational challenge thus recognized.

The second type of problem arises when the surface features of the current interactive context automatically trigger the self's primary and procedural emotion representations leading to non-adaptive and inappropriate emotion reactions. This can occur whenever the environmental situation shows sufficient similarity with the triggering conditions or associated situational and person-contexts that are represented by some of the self's primary Self-to-Other or Other-to-Self emotion schemes. Since the input specifications of such primary emotion schemes represent the recurring patterns specific to the self's early attachment interactions, their ‘spurious’ triggering by a featurally partially overlapping current context is highly unlikely to generate an automatic emotional reaction that is adaptive in the present situation. Therefore, if left cognitively unchecked, such automatic and similarity-driven contextual activation of the procedurally organized primary emotion schemes are likely to result in inappropriate and reality-distorting projective constructions of the emotional significance of the current interpersonal situation. In order to control such dysfunctional automatic activation of primary emotion schemes, the self must again rely on top-down processes of secondary cognitive reappraisal of current situational cues (e.g. Lazarus 1991; Thompson 1994; Ellsworth and Scherer 2003; Ochsner and Gross 2005) to inhibit and override the impending execution of the spuriously activated primary emotions. This ontogenetically developing capacity for top-down control over one's primary emotional automatisms relies on two related aspects of cognitive processes of secondary evaluation. Both result in a reinterpretation of events and the generation of alternative response options.

First, the secondary reappraisal processes can reinterpret the actual significance for the self of the causal evoking conditions that have automatically activated the primary emotional reaction in the first place. Based on stored or situationally inferred relevant information, one can reappraise and revise the interpretation of what the emotion-triggering events signify for the self. This top-down cognitive re-evaluation can override the automatic output of the ‘quick-and-dirty’ stimulus-driven primary perceptual appraisal processes (LeDoux 1995, 1996, 2000; Ochsner and Gross 2005) which led to the original automatic triggering of one's primary emotion reaction. Secondly, the secondary cognitive evaluative processes can also reappraise the realistic significance of the expectable causal consequences to which the automatic expression of the spuriously activated primary emotional reaction would lead.

However, several developmental, representational, and cognitive preconditions need to be established before such secondary cognitive reappraisal processes can be successfully applied to serve their emotion-regulative function adaptively. First, top-down cognitive processes of emotional self-control need to be ‘alerted’ when a specific emotional arousal state has been activated. Therefore the cognitive emotion control system must be able to monitor, detect, and introspectively access internal state cues that signal the onset of a basic emotional reaction in the self. Secondly, this introspective emotion recognition must be based on sufficiently differentiated internal state signals to allow for the correct identification and self-attribution of the particular emotion category that has been triggered. Thirdly, the cognitive processes of secondary reappraisal must have access to sufficient and relevant (stored or inferred) information in the light of which adaptive re-evaluation of the realistic significance of the causal conditions of the basic emotion reaction can be successfully performed. Importantly, these processes of representational access and situational reappraisal must be implemented under time pressure to deliver their output before the execution of the stereotypic behavioural components of the activated basic emotion program is automatically implemented.

11.3 The establishment of second-order emotion representations: developmental mechanisms and social environmental preconditions

11.3.1 The development of the subjective sense of a differentiated affective self

Many currently popular developmental and attachment theories consider the human-specific characteristics of early caregiver–infant attachment interactions (see section 11.1) as evidence supporting a ‘rich’ mentalistic interpretation of such interactions in terms of primary ‘intersubjectivity’. These theories assume that during the first months of life human infants already have introspective access to a variety of differential mental states (such as emotions, goals, intentions, or motives), can recognize and attribute such internal states to the other's mind, and experience subjective and mutual awareness of ‘being in’ and ‘sharing’ such internal states with the caregiver (Meltzoff and Moore 1977, 1989, 1998; Trevarthen 1979, 1993; Stern 1985; Braten 1988, 1992, 1998; Giannino and Tronick 1988; Meltzoff and Gopnik 1993; Hobson 1993, 2002; Trevarthen and Aitken 2001; Meltzoff 2002). Elsewhere we have criticized this primary intersubjectivist interpretation on a number of grounds, arguing that it attributes to very young infants overly ‘rich’ mentalizing capacities that are not necessary to explain the characteristics of early mother–infant interactions in humans (for details see Gergely and Watson 1996, 1999; Fonagy et al. 2002, 2007; Gergely 2002, 2007a). We have proposed an alternative model of early socio-emotional self development (Gergely and Watson 1996, 1999; Fonagy et al. 2002) within the framework of contingency detection and attachment theory (Watson 1985, 2001). We hypothesized that infants’ subjective awareness of differentiated emotional self-states has important social interactional origins with an individually variable developmental course of ontogenetic unfolding. In this view, subjective self-awareness and sense of differential affective self-states is viewed as a (relatively hard-won) developmental achievement rather than being a universally given prewired capacity. Similarly, we see the establishment and ‘intersubjective’ recognition of mutual ‘sharing’ of affective states with others as an emergent property of the subjective self, rather than an initial and universal starting state of human infants’ mental life.

In particular, we have proposed that the subjective sense of discrete emotional self-states (e.g. awareness of being ‘angry’, rather than just experiencing an undifferentiated negative state of tension) is established as a result of infants’ repeated experience with the pattern of contingent parental reactivity and social ‘mirroring’ feedback reactions that the automatic expressions of these, initially non-conscious, basic emotional arousal states evoke from the attachment environment. Our model is based on two central assumptions.

  1. 1. The introspectively inaccessible constitutional affective self We assume (together with numerous others) that the infant's innate constitutional self has a richly structured internal organization at birth. While on the one hand the constitutional self is characterized by significant genetically based individual temperamental differences, it also contains a basic set of prewired universal categorical emotion programs that are primary biological adaptations (Ekman et al. 1972; Ekman 1992; Gergely and Watson 1996). As argued earlier, these basic emotions are best understood as prewired stimulus-driven procedural physiological and motor automatisms that are initially not accessible to conscious awareness and over which the baby has no voluntary control at first. Early affect-regulation is carried out mainly by the attachment environment, as the caregiver, reading the infant's automatic emotion expressions, reacts to them with appropriate affect-modulating interactions and emotion displays (Gergely and Watson 1996, 1999; Fonagy et al. 2002; Gergely 2007a). Thus, while infants may be sensitive to the general (positive versus negative) hedonic quality of their affective states, we assume that they have no initial introspective awareness of their discrete basic emotion states as categorically distinct subjective self-states (Gergely and Watson 1996). It is in this sense that we consider the constitutional affective self to be initially introspectively inaccessible; while it contains discrete and categorical basic emotional states, these prewired primary automatisms are at first not subject to introspective awareness and so they do not present themselves as subjectively experienced differential internal self-states. We further hypothesize, contrary to classical approaches such as Bruner et al. (1966) and Mahler et al. (1975) (cf. Gergely 2000), that in human infants, because of a prewired directionality bias, the attention system is initially dominantly externally oriented showing a primary sensitivity bias towards exteroceptive stimulation. Therefore it is assumed that very little introspectively directed attentional monitoring of internal self-states takes place in the earliest months of life (for supporting arguments see Gergely and Watson 1996, 1999; Fonagy et al. 2002).

  2. 2. The social interactive origins of affective subjectivity: the development of introspectively accessible emotional self-representations We assume that for the categorical basic emotion states of the constitutional self to become introspectively accessible and subjectively differentially experienced, two basic conditions must be established developmentally. On the one hand, the primary and procedural basic emotion programs need to become associated with second-order representations which, when activated, are cognitively accessible to introspectively oriented attentional self-monitoring processes. In other words, we suggest that the initially non-conscious primary emotion states become introspectively detectable differentiated states of subjective awareness through the activation and cognitive access of their associated second-order representations only. For this to happen, it is also necessary that the primary orientation bias of the young infant's attention system towards external stimulus events should become modified. The infant's attentional system must become partially directed towards the self's internal states to allow the active introspective monitoring and detection of the activation of the self's second-order emotion representations. Introspective emotion recognition then becomes possible whenever the second-order emotion representations are triggered by the stimulus-induced automatic arousal of the self's basic emotions to which they are associatively linked. This leads to the introspective self-perception of ‘being in’ a particular differentiated emotion state, giving rise to the discrete subjective awareness of that state.

The cognitively accessible and introspectively monitored second-order representations of the primary emotion states can then provide the basis for the establishment of a (continuously updated) representation of the dynamically changing affective states of the self. This makes it possible to generate self-predictions about what kind of emotion-induced behaviour one is about to perform by anticipating the impending execution of the emotion-specific action schemes that the self's activated basic emotional arousal state is about to trigger. To become subjectively aware of one's current dispositional affective states before the automatic activation of the primary emotion-induced action tendencies would take effect is a central precondition for the self's emerging capacity to exercise online affective self-control (Gergely and Watson 1996). The introspective detection of one's emotional arousal state allows the self to foresee and cognitively evaluate the likely consequences to which the anticipated primary emotion reactions would lead in the given interpersonal context. The self can then avoid such non-adaptive outcomes by inhibiting or modifying the anticipated primary emotional reactions in time (i.e. before they are executed) rather than being automatically driven to act them out.

Our developmental proposal is that both of these preconditions for the establishment of the introspectively accessible subjective self are fulfilled as a result of, and to the degree to which, the attachment environment provides systematic contingent feedback reactions to the automatic expressions of the initially non-conscious primary emotional arousal states of the infant. In other words, it is the repeated experience that one's current internal states are externally ‘mirrored’ or ‘reflected’ through the infant-attuned contingent reactions of the social environment that leads to the development of a subjective sense and awareness of one's differentiated affective self-states.

Of course, no amount of social ‘mirroring’ could result in the establishment of these two (representational and attentional) preconditions unless infants were equipped with an adequate mechanism to detect and represent the contingent relatedness between their automatic state-expressive displays on the one hand, and the ensuing environmental mirroring reactions on the other. This mechanism must be able to internalize the representation of the external ‘mirroring’ feedback signal as a second-order representation of the primary self-state whose automatic expression invokes the contingent ‘mirroring’ reaction in question.

In previous work on the social biofeedback theory of parental affect-mirroring (Gergely and Watson 1996, 1999) we have proposed that the infant's contingency detection device has the necessary structural properties to establish second-order emotion representations when applied to contingent affect-mirroring interactions. As already discussed, this prewired causal contingency analyser (Watson 1985, 1994) automatically monitors and assesses the degree of contingent relatedness between the infant's responses and subsequent external stimulus events. By monitoring three different aspects of contingent relatedness in parallel (temporal contingency, spatial similarity, and relative intensity) (reviewed by Gergely and Watson 1999), the mechanism can identify those aspects of the social environment that are under (some specific degree of) causal control of the infant's state expressive displays. One important developmental consequence of this causal contingency analysis is that the resulting discovery of the degree of contingent control that the self exercises over aspects of the social environment generates a subjective sense of causal efficacy and self-agency in the infant.

11.3.2 The role of ‘markedness’ of affect-mirroring expressions in the internalization of second-order representations of affective self-states

One of the human-specific characteristics of infant–caregiver interactions (see section 11.1) is that infant-attuned caregivers are inclined to produce ‘marked’ affect-mirroring expressions that are induced by their baby's automatic emotion displays. Such ‘marked’ affect-reflecting expressions are saliently transformed versions of the stereotypic motor displays of categorical emotions which normally express the caregiver's actual affective states (Gergely and Watson 1996, 1999; Fonagy et al. 2002). ‘Markedness’ involves motor transformations that result in an exaggerated, slowed down, schematic, abbreviated, or only partially executed normative display pattern of basic emotion expressions. Furthermore, marked affect-mirroring is typically accompanied by ostensive-communicative cues (such as direct eye contact, eyebrow raising, and vocal intonation pattern characteristic of ‘motherese’) that are also among the human-specific features of early caregiver–infant interactions (Csibra and Gergely 2006).

Our original functional interpretation of ‘markedness’ (Gergely and Watson 1996) emphasized its importance as a cue to signal that the displayed emotion ‘is not for real’ and should therefore be ‘decoupled’ from the mother, i.e. its attribution as her actual and real emotion state should be inhibited. Here we further propose that ‘marked’ affect-mirroring interactions constitute a specific form of ‘pedagogical’ manifestation of cultural knowledge transfer (Csibra and Gergely 2006) in the domain of emotion socialization. This new theoretical perspective allows us to extend the original functional account of ‘marked’ affect-mirroring interactions in two ways.

  • We shall show why ‘marked’ affect-mirroring displays become interpreted self-referentially by the infant, leading to their referential ‘anchoring’ as second-order emotion representations of the constitutional self's primary emotions.

  • We shall explicate how the ‘markedness’ of contingent affect-mirroring expressions contributes to the modification of the primary directional bias of the infant's initially dominantly externally oriented attention system, resulting in its eventual internally directed deployment to introspectively monitor the self's internal affectivef states.

11.3.3 Natural ‘pedagogy’: the functions of cues of ‘ostensive communication’ and ‘referential knowledge manifestation’ in communicative knowledge transfer

Recently, Csibra and Gergely (2006) developed a new evolutionary-based theory of cultural transmission, postulating a human-specific learning system that they call natural ‘pedagogy’ (Gergely 2007b; Gergely et al. 2007). They propose that humans have become adapted to spontaneously manifest relevant cultural knowledge to conspecifics who are naturally inclined to attend to and learn from such knowledge manifestations through a specialized cue-driven social communicative learning mechanism of mutual design. Human pedagogy ensures efficient intergenerational transfer of relevant cultural knowledge in a variety of knowledge domains including words, gestural symbols, artefact functions and novel means actions, valence properties of object kinds, social habits, rituals, etc. (Gergely et al. 2002a, 2007; Gergely and Csibra 2005, 2006; Csibra and Gergely 2006; Gergely 2007b).

Here we argue that natural pedagogy is also employed in the domain of emotion socialization to identify those categorical emotions that are culturally universal and shared among humans, to sensitize the infant to them, and to transfer relevant knowledge about them to the infant. In this view, early infant–caregiver affective interactions (involving ostensively cued ‘marked’ forms of contingent emotion-mirroring) constitute a special case of pedagogical knowledge transfer whereby sensitive caregivers establish second-order representations in infants that identify and encode the categorical emotions that are culturally universal and can be shared with and communicated to other humans.

The built-in mutual design structure of natural pedagogy involves biological preparedness for both providing and receiving relevant cultural information among humans. Knowledgeable caregivers show a natural inclination to express their ‘communicative intention’ (cf. Sperber and Wilson 1986) to transfer new and relevant knowledge to ignorant juveniles by addressing them through specific ‘ostensive’ communicative cues (such as eye contact, eyebrow flashing, contingent reactivity, or the prosodic pattern of ‘motherese’) to which infants show prewired sensitivity and preferential orientation (for a review of the supporting evidence see Csibra and Gergely 2006). Ostensive cues are followed by referential cues (such as gaze-shift) to help the infant identify the referent object about which new and relevant information is going to be demonstrated through the caregiver's ‘marked’ forms of knowledge manifestations. These ostensive and manifestative pedagogical cues trigger a specially receptive learning attitude in infants which induces fast mapping of the manifested information to their representation of the referent.

Cues of ‘referential knowledge manifestations’ help the infant to identify and extract the content of the adult's ‘referential intention’ (cf. Sperber and Wilson 1986), i.e. the new and relevant knowledge manifested about the referent, which the infant then encodes as a property of the referent kind. Such knowledge manifestations are cued by being performed in a specially ‘marked’ manner which involves a modified and saliently transformed version of the normative and primary form of application of the novel cultural knowledge skill that is being demonstrated for the infant to learn. (Imagine, for example, the difference between the ‘marked’ manifestative form of demonstrating to a novice ‘how to’ hammer in a nail versus the standard motor execution of the same routine when it is performed simply to fulfill its primary function of driving in a nail.) The features of ‘markedness’ which characterize knowledge manifestations serve several pedagogical functions. They cue the infant that the demonstrated content conveys relevant and culturally shared information about some essential property of the referent to be identified and represented. They also help the infant infer what aspect of the manifested knowledge should be encoded by foregrounding and selectively emphasizing the new and relevant information that should be extracted and internalized as part of the referent representation (for details see Gergely 2007b).

11.3.4 The role of cues of ‘ostension’ and ‘markedness’ in inducing the self-referential interpretation of ‘marked’ emotion-mirroring displays

We propose that the cues of ‘markedness'of emotion-mirroring parental expressions constitute a special instance of the ‘marked’ forms of manifestative knowledge displays which characterize communicative acts of pedagogical knowledge transfer. ‘Marked’ affect-mirroring displays are also accompanied by ostensive cues that signal the adult's communicative intent during pedagogical interactions (such as direct eye contact, raised eyebrows, the vocal pattern of ‘motherese’, and addressing the infant by name). We argued above (Gergely and Watson 1996, 1999; Fonagy et al. 2002) that as a consequence of these cues the infant inhibits the attribution of the expressed emotion to the caregiver as her ‘real’ feeling by referentially ‘decoupling’ the emotion display from her. We can now add that the cues of ostension and ‘markedness’ of emotion manifestation during affect-mirroring interactions also function to trigger the referential interpretive and receptive learning attitude of the infant's ‘pedagogical stance’. In other words, these pedagogical cues induce the infant to search for and identify the intended referent of the caregiver's ‘marked’ emotion manifestation and to ‘anchor’ the mirroring display to some referent other than the caregiver's actual emotion state (from which it has been ‘decoupled’). In trying to establish the referent of the ‘marked’ emotion display, the infant will rely on cues of referent identification (such as eye-gaze direction) which accompany the caregiver's affect-mirroring expression. Since the caregiver is looking at the infant while displaying her emotion-reflecting expressions, the infant's attention will be directed towards his/her own face and body as the likely spatial locus of the referent to which the ‘marked’ (and ‘decoupled’) emotion-mirroring display should be referentially ‘anchored’. Note that the infant's automatic facial–vocal emotion expressions that are being mirrored by the caregiver emanate from the very same locus as well. The infant's contingency detection mechanism identifies these emotion expressive responses of the self (and their proprioceptive bodily correlates) as the source of contingent control that the infant exercises over the caregiver's mirroring response (Gergely and Watson 1996). These two sources of information converge to identify the infant's own internal emotional arousal state and its motor expressions as the referent of the caregiver's ‘marked’ mirroring display. As a result, the infant will referentially ‘anchor’ the representation of the ‘marked’ mirroring display to his/her own primary and procedural emotion program and will internalize it as its second-order representation. In sum, this is how, through activating the infant's pedagogical stance, repeated experience with ostensively cued ‘marked’ affect-mirroring feedback reactions of infant-attuned caregivers can ‘teach’ the infant about the self's differential categorical emotion states which become represented as universal and cognitively accessible emotion categories shared by other humans. This is achieved by the referential ‘anchoring’ and internalization of the caregiver's ‘marked’ mirroring displays as second-order emotion representations which become associated with the infant's primary and procedural emotional self-states. Through this process the infant's attentional system becomes introspectively sensitized and internally directed to detect and monitor the activation of the self's affective states. This, together with the sensitization effects of the contingency-based social biofeedback mechanism as described by Gergely and Watson (1996), also results in heightened sensitivity to the internal proprioceptive cues that accompany emotional self-expressions. A further consequence is the development of subjective awareness and emergent voluntary induction of the instrumental effects of contingent control that the self exerts through its emotion responses over the infant-attuned reactivity of the social environment.

Through these processes the infant's social mirroring environment contributes significantly to the establishment of the introspectively detectable subjective self by populating it with cognitively accessible second-order representations of internal affective self-states. This provides the representational basis which, together with the infant's introspectively socialized attentional system, make the developmental emergence of top-down cognitive emotional self-monitoring and secondary reappraisal processes possible.

11.4 Mentalizing and its role in secondary cognitive reappraisal processes of emotional self-control

We now turn to the important role that the ability to mentalize can play in enhancing the self's developing competence to achieve top-down cognitive emotional self-regulation and control by increasing the efficiency and scope of the secondary cognitive reappraisal processes involved. We speculate that the proper evolutionary domain (cf. Sperber and Hirschfeld 2004) of the cognitive adaptation for mentalizing may have originally been restricted to infer and represent the causal intentional mental states of other minds only. This was sufficient to confer significant evolutionary benefits on the mindreader by allowing him/her to anticipate the likely actions of other conspecifics in primarily competitive situations. However, we hypothesize that with the human-specific establishment during early social and emotional development of cognitively accessible second-order representations of internal self-states, the proper domain of the mentalizing capacity becomes ontogenetically extended to include in its actual domain the mind of one's own self as well. This is likely to result in significantly better coping strategies in the realm of competitive and cooperative interactions and affective relationships by enabling the self to anticipate its own actions and reactions, as well as those of others, through introspectively detecting, self-attributing, representing, and reflecting about one's own internal mental states that are the causal sources underlying one's actions.

So how is the inferential system for mentalizing applied introspectively in the domain of emotional self-regulation and control to make coping with affectively charged interpersonal interactions more effective? Clearly, one's mindreading abilities can often provide highly useful and relevant information to instruct the self's secondary cognitive emotion reappraisal mechanisms. This would be the case, for example, when these top-down cognitive processes are applied to re-evaluate the realistic significance for the self of the causal input conditions that have induced one's primary and automatic emotional arousal reactions in situations where these triggering conditions involve another person's actions, reactions, or emotion expressions directed towards the self. In such cases mentalizing processes can support secondary cognitive reappraisals by accessing and reasoning about stored knowledge concerning the other person's enduring mental dispositions, self-related attitudes, long-term aspirations, temperamental and personality traits, or recent life events that are relevant for the reconstruction of the mental reasons behind the other's self-directed actions. Similarly, one's mentalizing ability can be recruited to detect and interpret current situational or behavioural cues that are relevant for inferring what particular present desires, intentions, and (possibly false) beliefs about the situation and/or one's own self may have induced the other's responses towards the self that led to the triggering of one's automatic basic emotional arousal response.

Mentalizing can also significantly facilitate the generation of secondary cognitive evaluations of the likely causal consequences that automatically acting out or expressively displaying one's basic emotional arousal reactions would lead to in the given interpersonal context. The validity and scope of the predictions generated by such secondary cognitive evaluations can be qualitatively enriched by mentalistic reasoning about what kind of causal mental states and reactions would be expected to be induced in the other person by the self's automatic emotion-induced responses. Such mentalizing inferences can take into consideration a range of accessible or inferable information about relevant situation- and person-specific motivational factors such as the other's temperamental features, personality traits, childhood history, current dispositional states, his/her quality of relationship with and available relevant knowledge about one's own self, etc. Importantly, in generating predictions about the other's likely reactions to one's potential emotional outburst a good mentalizer may also be able to take into account available knowledge about the quality of the other's mentalizing skills that the latter can be counted on to apply when evaluating the reasons behind one's own emotional reactions towards him/her in the given situation.

11.5 Conclusions

One of the, arguably, species-unique abilities of humans is the exercise of online introspective control over their primary emotional impulses in affectively charged interpersonal situations and relationships. However, individuals differ significantly in the degree to which this remarkable capacity is available to them or can be put to functional use. In this chapter we have outlined a social cognitive developmental theory specifying the early social environmental preconditions and cognitive representation-building and attention socialization mechanisms that underlie the ontogenetic unfolding of the introspectively accessible representational affective self. This approach relates the individual variability in the capacity for introspective access to affective self-states and emotional self-control to the differences in the availability of a contingently reactive infant-attuned attachment environment in which the infant's automatic emotion expressive responses induce contingent external feedback reactions and ‘marked’ forms of affect-mirroring displays by caregivers. We have argued that these human-specific features of early caregiver-infant interactions constitute necessary environmental input conditions for the infant's representation-building and attention socialization mechanisms to establish second-order cognitively accessible representations for the initially non-conscious, stimulus-controlled, and procedural basic emotional automatisms of the innate constitutional self. Our approach sheds new light on the functional role of the human-specific features of early affective attachment interactions (such as contingent turn-taking reactivity, ‘marked’ forms of affect-mirroring, or ostensive-communicative cues of ‘pedagogical’ knowledge manifestations) in early social cognitive development. It interprets ostensively cued ‘marked’ affect-reflective interactions as a special case of ’ knowledge manifestations (Csibra and Gergely 2006) which function to transfer relevant cultural knowledge about universally shared emotion categories by establishing second-order representations for the infant's primary emotion states. We have contrasted this view with currently popular primary ‘intersubjectivist’ interpretations which consider the species-unique features of caregiver–infant interactions as characteristics of early secure attachment relationships that function as developmental precursors and facilitating conditions fostering the acquisition of mentalizing by early childhood. Finally, we have argued that the developmentally established second-order emotion representations allow the self to extend the original domain of the innate capacity for mentalizing about other minds to include introspective mentalizing about one's own internal affective mental states and anticipated emotion-induced actions as well. As a result the introspectively accessible representational affective self can employ its innate mentalizing ability in the service of emotional self-regulation and control by anticipating and adaptively modifying its own emotion-induced action tendencies to cope more efficiently with affectively charged interpersonal interactions.


Ainsworth, M.D.S., Blehar, M.C., Waters, E., and Wall, S. (1978). Patterns of attachment: a psychological study of the Strange Situation. Hillsdale, NJ, Lawrence Erlbaum.Find this resource:

    Astington, J.W. and Jenkins, J.M. (1999). A longitudinal study of the relation between language and theory-of-mind development. Developmental Psychology, 35, 1311–20.Find this resource:

    Bahrick, L.R. and Watson, J.S. (1985). Detection of intermodal proprioceptive–visual contingency as a potential basis of self-perception in infancy. Developmental Psychology, 21, 963–73.Find this resource:

    Baron-Cohen, S. (1995). Mindblindness:an essay on autism and theory of mind. Cambridge, MA, Bradford–MIT Press.Find this resource:

      Baron-Cohen, S., Leslie, A.M., and Frith, U. (1985). Does the autistic child have a ‘theory of mind’? Cognition, 21, 37–46.Find this resource:

      Behne, T., Carpenter, M., and Tomasello, M. (2005). One-year-olds comprehend the communicative intentions behind gestures in a hiding game. Developmental Science, 8, 492–9.Find this resource:

      Bennett, D.S., Bendersky, M., and Lewis, M. (2004). On specifying specificity: facial expressions at 4 months. Infancy, 6, 425–9.Find this resource:

      Bennett, D.S., Bendersky, M., and Lewis, M. (2005). Does the organization of facial expression change over time? Facial expressivity from 4 to 12 months. Infancy, 8, 167–87.Find this resource:

      Bigelow, A.E. (1999). Infants’ sensitivity to imperfect contingency in social interaction. In Early social cognition (ed. P Rochat), pp 137–154. Hillsdale, NJ: Lawrence Erlbaum.Find this resource:

        Bigelow, A.E. and De Coste, C. (2003). Infants’ sensitivity to contingency in social interactions with familiar and unfamiliar partners. Infancy, 4, 111–40.Find this resource:

        Bigelow, A.E. and Rochat, P. (2006). Two-month-old infants’ sensitivity to social contingency in mother–infant and stranger–infant interaction. Infancy, 9, 313–25.Find this resource:

        Bloom, P. and German, T.P. (2000). Two reasons to abandon the false belief task as a test of theory of mind. Cognition, 77, B25–31.Find this resource:

        Bowlby, J. (1969). Attachment and loss: Vol. 1: Attachment. London: Hogarth Press and Institute of Psycho-Analysis.Find this resource:

          Braten, S. (1988). Dialogic mind: the infant and the adult in protoconversation. In Nature, cognition, and system, Vol. 1 (ed. M Carvallo), pp. 187–205. Dordrecht: Kluwer Academic.Find this resource:

          Braten, S. (1992). The virtual other in infants’ minds and social feelings. In The dialogical alternative (ed. H Wold), pp. 77–97. Oslo: Scandinavian University Press.Find this resource:

            Braten, S. (ed.) (1998). Intersubjective communication and emotion in early ontogeny. Paris: Cambridge University Press–Editions de la Maison des Science de l’Homme.Find this resource:

              Brazelton, T.B. and Tronick, E. (1980). Preverbal communication between mothers and infants. In The social foundations of language and thought (ed. D.R. Olson), pp. 299–315. New York: Norton.Find this resource:

                Brazelton, T.B., Koslowski, B., and Main, M. (1974). The origins of reciprocity: the early mother–infant interaction. In The effect of the infant on its caregiver (ed. M. Lewis and L. Rosenblum), pp. 49–76. New York: John Wiley.Find this resource:

                  Bruner, J.S., Olver, P.R., and Greenfield, P.M. (1966). Studies on cognitive growth. New York: John Wiley.Find this resource:

                    Bugnyar, T. and Heinrich, B. (2005). Ravens, Corvus corax, differentiate between knowledgeable and ignorant competitors. Proceedings of the Royal Society of London, 272, 1641–6.Find this resource:

                    Camras, L.A. (1992). Expressive development and basic emotions. Cognition and Emotion, 6, 269–83.Find this resource:

                    Cassidy, J. and Shaver, P.R. (1999). Handbook of attachment: theory, research and clinical implications. New York: Guilford Press.Find this resource:

                      Cohn, J.F. and Tronick, E.Z (1988). Mother–infant face-to-face interaction: influence is bidirectional and unrelated to periodic cycles in either partner's behavior. Developmental Psychology, 24, 386–92.Find this resource:

                      Cooper, R.P. and Aslin, R.N. (1990). Preference for infant-directed speech in the first month after birth. Child Development, 61, 1584–95.Find this resource:

                      Csibra, G. and Gergely, G. (2006). Social learning and social cognition: the case of pedagogy. In Processes of change in brain and cognitive development. Attention and Performance XXI, (ed. M.H. Johnson and Y.M. Munakata), pp. 249–74. Oxford University Press.Find this resource:

                      Csibra, G. and Southgate, V. (2006). Evidence for infants understanding false beliefs should not be dismissed. Trends in Cognitive Sciences, 10, 4–5.Find this resource:

                      Cutting, A.L. and Dunn, J. (1999). Theory of mind, emotion understanding, language, and family background: individual differences and interrelations. Child Development, 70, 853–65.Find this resource:

                      Darwin, C. (1872). The expression of emotions in man and animals. New York: Philosophical Library.Find this resource:

                      De Marco, A. and Visalberghi, E. (2007). Facial displays in young tufted capuchin monkeys (Cebus apella): appearance, meaning, context and target, Folia Primatologica, 78, 118–37.Find this resource:

                      Dennett, D.C. (1978). Beliefs about beliefs. Behavioral and Brain Sciences, 4, 568–70.Find this resource:

                      Dennett, D.C. (1987). The intentional stance. Cambridge, MA: MIT Press.Find this resource:

                        de Rosnay, M. and Harris, P.L. (2002). Individual differences in children's understanding of emotion: the roles of attachment and language. Attachment and Human Development, 4, 39–54.Find this resource:

                        de Rosnay, M. and Hughes, C. (2006). Conversation and theory of mind. Do children talk their way to socio-cognitive understanding? British Journal of Developmental Psychology, 24, 7–37.Find this resource:

                        de Villiers, J.G. and de Villiers, P. (2000). Linguistic determinism and the understanding of false belief. In Children's reasoning and the mind (ed. P Mitchell and K Riggs), pp. 191–228. Hove: Psychology Press.Find this resource:

                          de Villiers, J.G. and Pyers, J.E. (2002). Complements to cognition: a longitudinal study of the relationship between complex syntax and false-belief-understanding. Cognitive Development, 17, 1037–60.Find this resource:

                          Dunn, J., Brown, J., and Beardsall, L. (1991). Family talk about feeling states and children's later understanding of others’ emotions. Developmental Psychology, 27, 448–55.Find this resource:

                          Ekman, P. (1992). Facial expressions of emotion: new findings, new questions. Psychological Science, 3, 34–8.Find this resource:

                          Ekman, P., Friesen, W.V., and Ellsworth, P. (1972). Emotion in the human face. New York: Pergamon Press.Find this resource:

                            Ellsworth, P.C. and Scherer, K.R. (2003). Appraisal process in emotion. In Handbook of affective sciences (ed. R.J. Davidson, K.R. Scherer, and H.H. Goldsmith), pp. 572–95. Oxford University Press.Find this resource:

                              Emery, N.J. and Clayton, N.S. (2004). The mentality of crows: convergent evolution of intelligence in corvids and apes. Science, 306, 1903–7.Find this resource:

                              Farroni, T., Csibra, G., Simion, F., and Johnson, M.H. (2002). Eye contact detection in humans from birth. Proceedings of the National Academy of Sciences of the USA, 99, 9602–5.Find this resource:

                              Fernald, A. (1985). Four-month-old infants prefer to listen to motherese. Infant Behavior and Development, 8, 181–95.Find this resource:

                              Fernald, A. (1992). Human maternal vocalizations to infants as biological signals: an evolutionary perspective. In The adapted mind: evolutionary psychology and the generation of culture (ed. J.H. Barkow, J. Tooby, and L.C. Cosmides), pp. 391–428. Oxford University Press.Find this resource:

                                Ferrari, P.F., Visalberghi, E., Paukner, A., Fogassi, L., Ruggiero, A., and Suomi, S.T. (2006). Neonatal limitation in rhesus monkeys. PLoS Biology, 4(9), 302.Find this resource:

                                Floccia, C., Christophe, A., and Bertoncini, J. (1997). High-amplitude sucking and newborns: the quest for underlying mechanisms. Journal of Experimental Child Psychology, 64, 175–89.Find this resource:

                                Fodor, J, (1992). A theory of the child's theory of mind. Cognition, 44, 283–96.Find this resource:

                                Fonagy, P. (2001). Attachment theory and psychoanalysis. New York: Other Press.Find this resource:

                                  Fonagy, P. and Target, M. (1997). Attachment and reflective function: their role in self-organization. Development and Psychopathology, 9, 679–700.Find this resource:

                                  Fonagy, P., Redfern, S., and Charman, T. (1997a). The relationship between belief–desire reasoning and a projective measure of attachment security (SAT). British Journal of Developmental Psychology, 15, 51–61.Find this resource:

                                  Fonagy, P., Steele, H., Steele, M., and Holder, J. (1997b). Attachment and theory of mind: overlapping constructs? Association for Child Psychology and Psychiatry Occasional Papers, 14, 31–40.Find this resource:

                                    Fonagy, P., Target, M., and Gergely, G. (2000) Attachment and borderline personality disorder: a theory and some evidence. Psychiatric Clinics of North America, 23, 103–23.Find this resource:

                                    Fonagy, P., Gergely, G., Jurist, E. and Target, M. (2002). Affect-regulation, mentalization and the development of the self. New York: Other Press.Find this resource:

                                      Fonagy, P., Target, M., Gergely, G., Allen, J.G. and Bateman, A. (2003). The developmental roots of borderline personality disorder in early attachment relationships: a theory and some evidence. Psychoanalytic Inquiry, 23, 412–59.Find this resource:

                                      Fonagy, P., Gergely, G. and Target, M. (2007). The parent–infant dyad and the construction of the subjective self. Journal of Child Psychology and Psychiatry, 48, 288–328.Find this resource:

                                      Francis, D.D. and Meaney, M.J. (1999) Maternal care and the development of stress responses. Current Opinions in Neurobiology, 9, 128–34.Find this resource:

                                      Francis, D.D., Diorio, J., Liu, D., and Meaney, M.J. (1999). Variations in maternal care form the basis for a non-genomic mechanism of inter-generational transmission of individual differences in behavioral and endocrine responses to stress. Science, 286(5442), 1155–8.Find this resource:

                                      Gergely, G. (2000). Reapproaching Mahler: new perspectives on normal autism, normal symbiosis, splitting and libidinal object constancy from cognitive developmental theory. Journal of the American Psychoanalytic Association, 48(4), 1197–1228.Find this resource:

                                      Gergely, G. (2002). The development of understanding self and agency. In Blackwell handbook of childhood cognitive development (ed. U. Goshwami), pp. 26–46. Oxford: Blackwell.Find this resource:

                                      Gergely, G. (2004). The role of contingency detection in early affect-regulative interactions and in the development of different types of infant attachment. Social Development, 13, 468–88.Find this resource:

                                      Gergely, G. (2007a). The social construction of the subjective self: the role of affect-mirroring, markedness, and ostensive communication in self development. In Developmental science and psychoanalysis (ed. L. Mayes, P. Fonagy, and M. Target). London: Karnac.Find this resource:

                                        Gergely, G. (2007b). Learning ‘about’ versus learning ‘from’ other minds: human pedagogy and its implications. In Innateness. Vol. III: Foundations and mechanisms (ed. P Carruthers). Oxford University Press.Find this resource:

                                          Gergely, G. and Csibra, G. (2005). The social construction of the cultural mind: imitative learning as a mechanism of human pedagogy. Interaction Studies, 6, 463–81.Find this resource:

                                          Gergely, G. and Csibra, G. (2006). Sylvia's recipe: the role of imitation and pedagogy in the transmission of cultural knowledge. In Roots of human sociality: culture, cognition, and human interaction (ed. S. Levenson and N. Enfield), pp. 229–255. Oxford: Berg.Find this resource:

                                            Gergely, G. and Unoka, Z. (2007). The development of the unreflective self. In Mentalization: theoretical considerations, research findings, and clinical implications (ed. F.N. Bush). Hillsdale, NJ: Analytic Press.Find this resource:

                                              Gergely, G. and Watson, J.S. (1996). The social biofeedback theory of parental affect-mirroring: the development of emotional self-awareness and self-control in infancy. International Journal of Psycho-Analysis, 77, 1–31.Find this resource:

                                              Gergely, G. and Watson, J.S. (1999). Early social-emotional development: contingency perception and the social biofeedback model. In Early social cognition (ed. P Rochat), pp. 101–37. Hillsdale, NJ: Lawrence Erlbaum.Find this resource:

                                                Gergely, G., Bekkering, H., and Király, I. (2002a). Rational imitation in preverbal infants. Nature, 415, 755.Find this resource:

                                                Gergely, G., Fonagy, P., and Target, M. (2002b). Attachment, mentalization, and the etiology of borderline personality disorder. Self Psychology, 3, 73–82.Find this resource:

                                                  Gergely, G., Király, I., and Egyed, K. (2007). On pedagogy. Developmental Science, 10, 139–46.Find this resource:

                                                  Giannino, A. and Tronick, E.Z (1988). The mutual regulation model: the infant's self and interactive regulation and coping and defensive capacities. In Stress and coping across development (ed. T.M. Field, P.M. McCabe, and N. Schneiderman), pp. 47–68. Hillsdale, NJ: Lawrence Erlbaum.Find this resource:

                                                    Goldberg, S., Muir, R., and Kerr, J. (1995). Attachment theory: social, developmental and clinical perspectives. Hillsdale, NJ: Analytic Press.Find this resource:

                                                      Hare, B., Call, J., and Tomasello, M. (2001). Do chimpanzees know what conspecifics know? Animal Behavior, 61, 139–151.Find this resource:

                                                      Harlow, H.F. (1961). The development of affectional patterns in infant monkeys. In Determinants of infant behaviour (ed. BM Foss), Vol. 1, pp. 75–88. London: Methuen.Find this resource:

                                                        Harris, P. (2005). Conversation, pretence and theory of mind. In Why language matters for theory of mind (ed. J. Astington and J. Baird), pp. 70–83. New York: Oxford University Press.Find this resource:

                                                        Hobson, R.P. (1993). Autism and the development of mind. Hove: Lawrence Erlbaum.Find this resource:

                                                          Hobson, P. (2002). The cradle of thought: exploring the origins of thinking. Oxford University Press.Find this resource:

                                                            Hofer, M.A. (1995). Hidden regulators: implications for a new understanding of attachment, separation and loss. In Attachment theory: social, developmental and clinical perspectives (ed. S. Goldberg, R. Muir, and J. Kerr), pp. 203–30. Hillsdale, NJ: Analytic Press.Find this resource:

                                                              Hughes, C. and Dunn, J. (2002). ‘When I say a naughty word.’ Children's accounts of anger and sadness in self, mother and friend: longitudinal findings from ages four to seven. British Journal of Developmental Psychology, 20, 515–35.Find this resource:

                                                              Izard, C.E. and Malatesta, C.Z. (1987). Perspectives on emotional development. I: Differential emotions theory of early emotional development. In Handbook of Infant Development (2nd edn) (ed. J.D. Osofsky), pp. 494–554. New York: John Wiley.Find this resource:

                                                                Johnson, S.C., Slaughter, V., and Carey, S. (1998). Whose gaze will infants follow? The elicitation of gaze-following in 12-month-olds. Developmental Science, 1, 233–8.Find this resource:

                                                                Kagan, J. (1994). Galen's prophecy: temperament in human nature. New York: Basic Books.Find this resource:

                                                                  Kaminski, J., Call, J., and Tomasello, M. (2006). Goat's behaviour in a competitive food paradigm: evidence for perspective taking? Behaviour, 143, 1341–56.Find this resource:

                                                                  Lazarus, R.S. (1991). Emotion and adaptation. New York: Oxford University Press.Find this resource:

                                                                    Lazarus, R.S. and Lazarus, B.N. (1994). Passion and reason: making sense of our emotions. New York: Oxford University Press.Find this resource:

                                                                      LeDoux, J.E. (1995). Emotion: clues from the brain. Annual Review of Psychology, 46, 209–35.Find this resource:

                                                                      LeDoux, J.E. (1996). The emotional brain: the mysterious underpinnings of emotional life. New York: Simon & Schuster.Find this resource:

                                                                        LeDoux, J.E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience, 23, 155–84.Find this resource:

                                                                        Leslie, A.M. (2000). ‘Theory of Mind’ as a mechanism of selective attention. In The new cognitive neurosciences (2nd edn) (ed. M.S. Gazzaniga), pp. 1235–47. Cambridge, MA: MIT Press.Find this resource:

                                                                          Leslie, A.M., Friedman, O., and German, T.P. (2004). Core mechanisms in ‘theory of mind’. Trends in Cognitive Sciences, 8, 528–33.Find this resource:

                                                                          Lewis, M., Allessandri, S.M., and Sullivan, M.W. (1990). Violation of expectancy, loss of control and anger expressions in young infants. Developmental Psychology, 26, 745–51.Find this resource:

                                                                          Liszkowski, U., Carpenter, M., Henning, A., Striano, T., and Tomasello, M. (2004). Twelve-month-olds point to share attention and interest. Developmental Science, 7, 297–307.Find this resource:

                                                                          Lohmann, H., Tomasello, M., and Meyer, S. (2005). Linguistic communication and social understanding. In Why language matters for theory of mind (ed. J. Astington and J. Baird), pp. 245–65. Oxford University Press.Find this resource:

                                                                          Mahler, M., Pine, F., and Bergman, A. (1975). The psychological birth of the human infant: symbiosis and individuation. New York: Basic Books.Find this resource:

                                                                            Malatesta, C.Z., Culver, C., Tesman, R.J., and Shepard, B. (1989). The development of emotion expression during the first two years of life. Monographs of the Society for Research in Child Development, 54, No. 219.Find this resource:

                                                                            Meins, E., Fernyhough, C., Russel, J., and Clark-Carter, D. (1998). Security of attachment as a predictor of symbolic and mentalising abilities: a longitudinal study. Social Development, 7, 1–24.Find this resource:

                                                                            Meins, E., Fernyhough, C., Fradley, E., and Tuckey, M. (2001). Rethinking maternal sensitivity: Mothers’ comments on infants mental processes predict security of attachment at 12 months. Journal of Child Psychology and Psychiatry, 42, 637–48.Find this resource:

                                                                            Meins, E., Fernyhough, C., Wainwright, R., Das Gupta, M., Fradley, E., and Tuckey, M. (2002). Maternal mind-mindedness and attachment security as predictors of theory of mind understanding. Child Development, 73, 1715–26.Find this resource:

                                                                            Meins, E., Fernyhough, C., Wainwright, R., et al. (2003). Pathways to understanding mind: construct validity and predictive validity of maternal mind-mindedness. Child Development, 74, 1194–1211.Find this resource:

                                                                            Meltzoff, A.N. (2002). Imitation as a mechanism of social cognition: origins of empathy, theory of mind, and the representation of action. In Blackwell handbook of childhood cognitive development (ed. U. Goshwami), pp. 6–25. Oxford: Blackwell.Find this resource:

                                                                            Meltzoff, A.N. and Gopnik, A. (1993). The role of imitation in understanding persons and developing a theory of mind. In Understanding other minds: perspectives from autism (ed. S. Baron-Cohen, H. Tager-Flusberg, and D.J. Cohen), pp. 335–65. Oxford University Press.Find this resource:

                                                                              Meltzoff, A.N. and Moore, M.K. (1977). Imitation of facial and manual gestures by human neonates. Science, 198, 75–8.Find this resource:

                                                                              Meltzoff, A.N. and Moore, M.K. (1989). Imitation in newborn infants: exploring the range of gestures imitated and the underlying mechanisms. Developmental Psychology, 25, 954–62.Find this resource:

                                                                              Meltzoff, A.N. and Moore, M.K. (1998). Infant intersubjectivity: broadening the dialogue to include imitation, identity and intention. In Intersubjective communication and emotion in early ontogeny (ed. S. Braten), pp. 47–62. Paris: Cambridge University Press–Editions de la Maison des Science de l’Homme.Find this resource:

                                                                                Movellan, J.R. and Watson, J.S. (2002). The development of gaze following as a Bayesian systems identification problem. UCSD Machine Perception Laboratory Technical Reports, 2002(1).Find this resource:

                                                                                  Ochsner, K.N. and Gross, J.J. (2005). The cognitive control of emotion. Trends in Cognitive Sciences, 9, 242–9.Find this resource:

                                                                                  Onishi, K.H. and Baillargeon, R. (2005). 15-month-old infants understand false beliefs. Science, 308, 255–8.Find this resource:

                                                                                  Onishi, K.H., Baillargeon, R., and Leslie, A.M. (2007). 15-month-old infants detect violations in pretend scenarios. Acta Psychologica, 124, 106–28.Find this resource:

                                                                                  Ontai, L.L. and Thompson, R.A. (2002). Patterns of attachment and maternal discourse effects on children's emotion understanding from 3 to 5 years of age. Social Development, 11, 433–50.Find this resource:

                                                                                  Panksepp, J. (1998). Affective neuroscience: the foundations of human and animal emotions. Oxford University Press.Find this resource:

                                                                                    Perner, J. (1991). Understanding the representational mind, Cambridge, MA: MIT Press.Find this resource:

                                                                                      Perner, J. and Lang, B. (2000). Theory of mind and executive function. Is there a developmental relationship? In Understanding other minds: perspectives from autism (ed. S. Baron-Cohen, H. Tager-Flusberg, and D.J. Cohen), pp. 150–81. New York: Oxford University Press.Find this resource:

                                                                                        Perner, J., Ruffman, T., and Leekman, S.R. (1994). Theory of mind is contagious: You catch it from your sibs. Child Development, 65, 1228–38.Find this resource:

                                                                                        Polan, H.J. and Hofer, M.A. (1999). Psychobiological origins of attachment and separation responses. In Handbook of attachment: theory, research and clinical implications (ed. J. Cassidy and P.R. Shaver), pp. 162–80. New York: Guilford Press.Find this resource:

                                                                                          Posner, M.I. and Rothbart, M.K. (2000). Developing mechanisms of self-regulation. Development and Psychopathology, 12, 427–41.Find this resource:

                                                                                          Pruessner, J.C., Champagne, F., Meaney, M.J., and Dagher, A. (2004). Dopamine release in response to a psychological stress in humans and its relationship to early life maternal care: a positron emission tomographystudy using [11C]raclopride. Journal of Neuroscience, 24, 2825–31.Find this resource:

                                                                                          Rochat, P. and Morgan, R. (1995). Spatial determinants in the perception of self-produced leg movements in 3- and 5-month-old infants. Developmental Psychology, 31, 626–36.Find this resource:

                                                                                          Rothbart, M.K. (1989). Temperament and development. In Temperament in childhood (ed. G.A. Kohnstamm, J.E. Bates, and M.K. Rothbart), pp. 187–247. New York: John Wiley.Find this resource:

                                                                                            Ruffman, T., Perner, J., Naito, M., Parkin, L., and Clements, W. (1998). Older (but not younger) siblings facilitate false belief understanding. Developmental Psychology, 34, 161–74.Find this resource:

                                                                                            Ruffman, T., Perner, J., and Parkin, L. (1999). How parenting style affects false belief understanding. Social Development, 8, 395–411.Find this resource:

                                                                                            Sabbagh, M.A. and Callanan, M.A. (1998). Metarepresentation in action: 3-, 4-, and 5-year-olds’ developing theories of mind in parent–child conversations. Developmental Psychology, 34, 491–502.Find this resource:

                                                                                            Sander, L. (1988). The event-structure of regulation in the neonate–caregiver system as a biological background for early organisation of psychic structure. In Frontiers in self psychology (ed. A. Goldberg), pp. 64–77. Hillsdale, NJ: Analytic Press.Find this resource:

                                                                                              Spangler, G. and Grossmann, K.E. (1993). Bio-behavioral organization in securely and insecurely attached infants. Child Development, 64, 1439–50.Find this resource:

                                                                                              Southgate, V., Senju, A., and Csibra, G. (2007a). Action anticipation through attribution of false beliefs by two-year-olds. Psychological Science, 18, 587–92.Find this resource:

                                                                                              Southgate, V., van Maanen, C., and Csibra, G. (2007b). Infant pointing: communication to cooperate or communication to learn? Child Development, 78, 735–40.Find this resource:

                                                                                              Sperber, D. and Wilson, D. (1986). Relevance: communication and cognition. Oxford: Blackwell.Find this resource:

                                                                                                Sperber, D. and Hirschfeld, L. (2004). The cognitive foundations of cultural stability and diversity. Trends in Cognitive Sciences, 8, 40–6.Find this resource:

                                                                                                Sroufe, L.A. (1996). Emotional development. The organization of emotional life in the early years. Cambridge University Press.Find this resource:

                                                                                                Stern DN. (1985). The Interpersonal Word of the Infant. New York: Basic Books.Find this resource:

                                                                                                  Suomi, S.J. (1995). Influence of Bowlby's attachment theory on research on non-human promate biobehavioral development. In Attachment theory: social, developmental and clinical perspectives (ed. S. Goldberg, R. Muir, and J. Kerr), pp. 185–201. Hillsdale, NJ: Analytic Press.Find this resource:

                                                                                                    Suomi, S.J. (1999). Attachment in rhesus monkeys. In Handbook of attachment: theory, research and clinical implications (ed. J. Cassidy and P.R. Shaver), pp. 181–97. New York: Guilford Press.Find this resource:

                                                                                                      Surian, L., Caldi, S., and Sperber, D. (2007). Attribution of beliefs by 13-month-old infants. Psychological Science, 18, 580–6.Find this resource:

                                                                                                      Thompson, R.S. (1994). Emotion regulation: a theme in search of definition. Monographs of the Society for Research on Child Development, 59, 25–52.Find this resource:

                                                                                                      Tomasello, M. and Haberl, K. (2003). Understanding attention: 12- and 18-month-olds know what's new for other persons. Developmental Psychology, 39, 906–12.Find this resource:

                                                                                                      Tomasello, M., Call, J., and Hare, B. (2003). Chimpanzees understand psychological states—the question is which ones and to what extent. Trends in Cognitive Sciences, 7, 153–6.Find this resource:

                                                                                                      Tomasello, M., Carpenter, M., Call, J., Behne, T., and Moll, H. (2005). Understanding and sharing intentions: the origins of cultural cognition. Behavioral and Brain Sciences, 28, 675–91.Find this resource:

                                                                                                      Tomasello, M., Carpenter, M., and Liszkowski, U. (2007). A new look at infant pointing. Child Development, 78, 705–22.Find this resource:

                                                                                                      Tomkins, S. (1995). Exploring affect: the selective writings of Silvan Tomkins. Cambridge University Press.Find this resource:

                                                                                                      Trevarthen, C. (1979). Communication and cooperation in early infancy: a description of primary intersubjectivity. In Before speech: the beginning of interpersonal communication (ed. M Bullowa), pp. 321–47. New York: Cambridge University Press.Find this resource:

                                                                                                        Trevarthen, C. (1993). The self born in intersubjectivity: an infant communicating. In The perceived self (ed. U Neisser), pp. 121–73. New York: Cambridge University Press.Find this resource:

                                                                                                          Trevarthen, C. and Aitken, K.J. (2001) Infant intersubjectivity: research, theory, and clinical applications. Journal of Child Psychology and Psychiatry, 42, 3–48.Find this resource:

                                                                                                          Tronick, E.Z (1989). Emotions and emotional communication in infants. American Psychologist, 44, 112–19.Find this resource:

                                                                                                          Tronick, E.Z and Cohn, J.F. (1989). Infant–mother face-to-face interaction: age and gender differences in coordination and the occurrence of miscoordination. Child Development, 60, 85–92.Find this resource:

                                                                                                          Uzgiris, I.C., Benson, J.B., Kruper, J.C., and Vasek, M.E. (1989). Contextual influences on imitative interactions between mothers and infants. In Action in social context: perspectives on early development (ed. J.J. Lockman and N.L. Hazen), pp. 103–27. New York: Plenum Press.Find this resource:

                                                                                                          Vinden, P.G. (2001). Parenting attitudes and children's understanding of mind: a comparison of Korean American and Anglo-American families. Cognitive Development, 16, 793–809.Find this resource:

                                                                                                          Watson, J.S. (1972). Smiling, cooing, and ‘the game’. Merrill-Palmer Quarterly, 18, 323–39.Find this resource:

                                                                                                            Watson, J.S. (1985). Contingency perception in early social development. In Social perception in infants (ed. T.M. Field and N.A. Fox), pp. 157–76. Norwood, NJ: Ablex.Find this resource:

                                                                                                              Watson, J.S. (1994). Detection of self: the perfect algorithm. In Self-awareness in animals and humans: developmental perspectives (ed. S.T. Parker, R.W. Mitchell, and M.L. Boccia), pp. 131–48. New York: Cambridge University Press.Find this resource:

                                                                                                              Watson, J.S. (1995). Mother–infant interaction: dispositional properties and mutual designs. In Perspectives in ethology. Vol. 11: Behavioral design (ed. N.S. Thompson). New York: Plenum Press.Find this resource:

                                                                                                                Watson, J.S. (2001). Contingency perception and misperception in infancy: some potential implications for attachment. Bulletin of the Menninger Clinic, 65, 296–321.Find this resource:

                                                                                                                Wellman, H. (1990). The child's theory of mind. Cambridge, MA: Bradford Books–MIT Press.Find this resource:

                                                                                                                  Wellman, H.M., Cross, D., and Watson, J. (2001). Meta-analysis of theory-of-mind development: the truth about false belief. Child Development, 72, 655–84.Find this resource:

                                                                                                                  Wimmer, H. and Perner, J. (1983). Beliefs about beliefs: representation and constraining function of wrong beliefs in young children's understanding of deception. Cognition, 13, 103–28.Find this resource:

                                                                                                                  Wismer Fries, A.B., Ziegler, T.E., Kurian, J.R., Jacoris, S., and Pollak, S.D. (2005). Early experience in humans is associated with changes in neoropeptides critical for regulating social behavior. Proceedings of the National Academy of Sciences of the USA, 102, 17237–40.Find this resource: