Symposium
Chairs: Vasudevi Reddy and Jacqueline Nadel
Discussant: Giannis Kugiumutzakis
Mirroring has been argued to be a critical feature of intersubjective communication, especially in its early manifestations in infancy, but also in the communication of mature adults. Indeed, it may be argued that without some form of mirroring between interactants, intersubjectivity can neither be expressed nor obtain the experience to develop. However, we know little about the phenomenon of mirroring: of what forms it might take, of what modalities it might occur in, of whom it might occur between or of how mirroring differs from imitation. This symposium presents four papers which present data on different aspects of this question. The first paper shows that mothers who do more affect mirroring of their 3 month-old infants during natural face to face interactions have infants who are more socially competent and show more signs of primary intersubjectivity. These infants are better able to discrimnate between normal and replay conditions of their mothers than infants of mothers who do less affect mirroring. This strongly supports the view that affect and the mirroring of affect in mutual engagement is vital for social development. Does mirroring and being positively affected by it imply that the infant recognises it as imitation? The second paper looks at the phenomenon of the recognition of being imitated. Nine-week-olds in a televised face to face interaction with their mothers do not display clear observable indices of imitation recognition, although they were able to imitate. Being able to imitate facial movements therefore does not imply de facto the capacity to recognize being imitated. This finding leads to the discussion of different definitions and levels of imitation recognition, in relationship with intersubjective communication. Comparative data from young chimpanzees while interacting with themselves in a mirror and while being taught to imitate actions, is used in a third paper to raise questions about different developmental pathways in intersubjective communication and in the process of imitation. The data suggest that there may be unique aspects of mirroring behaviors, especially as seen in intersubjective interactions with mirror images, that may differ from those involved in intersubjective interactions with social partners that occurs in imitation. The final paper extends the question of how we deal with completely synchronous mirroring 96 interacting with the self in a mirror. Children with autism often love mirrors but nonetheless use the mirror not to interact with the self but to explore its reflective properties including properties of behaviour. Children with Down's Syndrome on the other hand appear to enjoy engaging the self in 'interaction'. Using the self as a partner for interaction or as an audience for performance is observable in self- directed mirror behaviour and disturbances in this ability may be a sensitive index of disturbances in the development of the self and of intersubjectivity.
Details of individual items:
paper
The role of maternal affect mirroring (MAM) on the development of primary intersubjectivity was assessed in 41 2- 3-month-old infants. Mothers and infants were seen twice, approximately one week apart. During visit 1, mothers and infants interacted over television monitors for 3 minutes. During visit 2, infants engaged in a natural 3-minute interaction with their mothers over television monitors and they viewed a replay of the videotape of their mothers' interaction from the preceding week. The order of conditions was counterbalanced. Maternal affect mirroring was measured according to level of attention maintaining, level of warm sensitivity, and level of social responsiveness (imitation of vocal and facial gestures, and modulation of negative affect). A natural split was observed in these scores, with 58% if the mothers ranking high on MAM and 42 % ranking low on MAM. Infants in the HMAM group ranked high on primary intersubjectivity (co-regulation of smiles, gazes and vocalizations with their mothers) and discriminated between the natural and replay conditions with their smiles and gazes. Infants in the LMAM group ranked low on primary intersubjectvity and did not discriminate between the natural and replay televised conditions. These findings are interpreted as the demonstration of a developmental link between maternal affect mirroring and primary intersubjectivity in infants.
paper
In this paper, we discuss different definitions and levels of imitation recognition. Starting from our findings that some low- functioning children with autism are able to imitate very simple movements but do not provide any observable indices of imitation recognition, we will address the following question: do very young infants recognize when they are imitated? Knowing that neonates are able to reproduce facial movements that they see, does it follow that they can ipso facto establish correspondence between what they do and what they see somebody else doing? To explore this question, we designed an experiment in an interactive context. Dyads of mothers and their 9-week-old infants interacted through audio-video TV monitors. The microanalyses of the observable effect of being imitated revealed poor evidence of imitation recognition, while imitations were observed. This lack of evidence stresses the interest to consider imitation recognition not simply as the other facet of imitation but rather as a seperate though linked mechanism in play in intersubjective communication.
paper
We propose to discuss the particular skills that are necessary for young chimpanzees to demonstrate mirror self- recognition and imitation of arbitrary actions. It has been suggested that imitation of novel actions is a necessary prerequisite for mirror self-recognition. An alternative hypothesis is that mirror-guided behavior (of which self- directed is one type) is independent of imitation. In terms of delayed imitation and mark-directed mirror-guided behavior (skills that develop between 18 and 24 months in human infants), both skills may be based on mental representation. To study mirror self- recognition and imitation in a comparative developmental perspective is valuable for a number of reasons. First, the learning of each behavioral competency may be different for chimpanzees than for human infants. Studying the development of similar competencies through different developmental pathwayswill add to our understanding of the cognitive and social foundations of each behavior. Second, the pathways may be similar but the learning of imitation may be disassociated from the learning of mirror behavior in chimpanzees. Thus, comparative studies help us to understand the skills necessary for each of these competencies which may be complementary or independent. We will present data demonstrating that young chimpanzees may direct some social behaviors to mirror images (such as play invitations and aggressive displays), but they do not engage in proximal social behaviors (with a tactile component), such as tickling or grooming. This is true even when the chimpanzee is marked: there are no attempts to groom the mirror image. Since the chimpanzees are tested in their normal social groups, we have been able to observe social grooming of marked individuals, so it is quite notable that no social grooming of the mirror image has been observed, either in self- recognizing or in non-self-recognizing individuals. In terms of imitation, chimpanzees demonstrate neonatal imitation (at a rate similar to that in human newborns), but when tested in imitation tasks at 3BD years of age, initially chimpanzees had no idea of the required behavior.Training to imitate at this older age involved teaching the chimpanzees how to play the imitation game: watch the demonstrator, wait until the demonstration is over before taking your turn, and try to copy the same action as observed. After training, the chimpanzees were testedon novel actions. New analyses suggest that over the course of the three attempts, the chimpanzees improved their copying of the action. This comparison of chimpanzees with humans suggests that there may be unique aspects of mirroring behaviors, especially as seen in intersubjective interactions with mirror images, that may differ from those involved in intersubjective interactions with social partners that occurs in imitation.
paper
The Self too may be treated as an audience for performance or as a partner for interaction. This paper explores behaviour towards the self in a mirror in pre-school children with autism and pre-school children with Down's Syndrome (DS). Children in both groups were reported to be and observed to be interested in mirrors, and often fascinated by them. Clear group differences emerged in the extent to which the mirror was treated primarily as a reflective surface versus primarily as the provider of an interpersonal context. The children with autism spent significantly more time attending to parts of the room in the mirror, while the children with DS spent significantly more time attending to the self in the mirror. The mean duration of bouts of attention to the mirror were significantly longer in the DS group. The primary reason for interest in the mirror in the autism group was judged to be the exploration of the surface of the mirror and the manipulation of reflections, while in the DS group the main interest in the mirror was in exploration of the Self. While the children with autism used the mirror to experiment upon the world, the children with DS often used it to 'perform' to the self as an audience. Nearly two thirds of the children in both groups passed the mirror self-recognition task (MSR) and there was thus no relation between success on the MSR and the population group. Within groups the differences in response to the self in the mirror were not significantly related to success or failure in the MSR. Given that problems with both Self and Other are well documented in children with autism (and not in children with DS) it is possible that behaviour towards the self in the mirror may be a more sensitive index of disturbances in self development than is MSR. This paper raises questions about the recognition of the self as a partner for the study of intersubjectivity.