Authors: Rosa M. García Pérez, Anthony Lee y R. Peter Hobson.
Abstract: Does autism involve a deficit in intersubjective engagement with other persons? We studied nonverbal communication in children and adolescents with and without autism (n = 12 per group), group matched for chronological age and verbal mental age, during 3 min of a videotaped interview. In keeping with previous studies, there were only subtle but potentially revealing group differences on behavioral ratings. Participants with autism made fewer headshakes/nods (but not smiles) when the interviewer was talking, and the interviewer made fewer head-shakes/ nods when participants were talking. Yet there were marked group differences on reliable ‘subjective’ ratings of (a) affective engagement and (b) the smoothness of reciprocal interaction. We interpret the findings in terms of a group difference in identification between conversational partners.
Keywords: Autism, Intersubjectivity, Identification, Communication, Conversation.
Introduction
In his original description of the syndrome of early childhood autism, Kanner (1943) proposed that the children had ‘disturbances of affective contact’ with other people.
Yet until recently, theories concerning the pathogenesis of autism have tended to focus upon cognitive aspects of the disorder (for recent restatements and overviews, see Bailey, Phillips, & Rutter, 1996; Hill & Frith, 2003).
Against this tide of opinion, Hobson (1989, 1993) has long championed an elaboration of Kanner’s view that a profound disruption of patterned intersubjective engagement between the child and others is basic to autism, and a substantial number of researchers have highlighted aspects of the clinical picture that are affective/relational in nature (e.g., Loveland & Landry, 1986; Rogers & Pennington, 1991; Klin, Volkmar, & Sparrow, 1992; Mundy, 1995, 2003; Sigman & Capps, 1997; Dawson, Meltzoff, Osterling, Rinaldi, & Brown, 1998; Charman, 2003).
These perspectives do not gainsay the characteristic quality of cognitive impairments in autism. Yet in its more radical form (e.g., Hobson, 2002), the intersubjective perspective ascribes the source of many of the children’s abnormalities in interpersonal understanding (‘theory of mind’) and symbolic functioning to their lesser propensity to perceive, respond to and engage with the bodily-expressed attitudes of other people.
As one critical aspect of the disorder in interpersonal relations, Hobson (1993, 2002), Hobson and Lee (1999), Hobson and Meyer (2005), and Meyer and Hobson (2004) have stressed the significance of the children’s lesser propensity to identify with the actions and attitudes of others, that is, to register and assimilate the bodily-anchored psychological stance of another person so that this becomes a potential stance for themselves.
The present study is an attempt to explore this hypothesis as it applies to face-to-face conversations between adolescents with autism and an interested adult.
Our aim was to see whether, despite earlier studies suggesting only minor abnormalities in nonverbal communication between individuals with autism and a conversational partner, there might be previously undisclosed evidence of severe disruption in intersubjective connectedness between the two parties.
Already there are several strands of evidence that point to the plausibility of this suggestion.
First, clinicians have written of the unique ‘feel’ to one’s own experience of relating to a person with autism, even to the extent of feeling treated as if one were a piece of furniture (e.g., Kanner, 1943; Bosch, 1970; Klein, 1975; Alvarez, 1992). Although such evidence is sometimes discounted, it remains the case that the appropriate measure for assessing intersubjectivity is one person’s experience in relation to another—and often it does feel that something essential is missing in one’s engagement with a person with autism.
Hobson and Lee (1998) conducted a study of nonverbal communication in situations of greeting and departure toward an adult, and supplemented behavioral measures with reliable ‘subjective ratings’ of interpersonal engagement.
The results were that children and adolescents with autism were not only less likely than matched control participants to offer spontaneous verbal and nonverbal gestures, to establish eye contact, and to smile or to wave goodbye, but also they were significantly less likely than those without autism to receive subjective ratings of being ‘strongly engaged’ with the adult.
These findings indicate that in certain circumstances, at least, limited interpersonal engagement is apparent among children with autism in both behavioral and ‘subjective’ assessments.
If one considers this aspect of autism in developmental perspective, there is evidence from videotape studies (e.g., Adrien et al., 1992; Eriksson & de Chateau, 1992; Osterling & Dawson, 1994; Baranek, 1999) and direct observations of infants (Charman et al., 1997), as well as from retrospective parental reports (e.g., Wing, 1969; Hoshino et al., 1982; Dahlgren & Gillberg, 1989; Stone & Hogan, 1993; Vostanis et al., 1998; Wimpory, Hobson, Williams, & Nash, 2000), that even very young children with autism have characteristic impairments in nonverbal communication of a kind that might reflect and/or lead to disruption in intersubjective engagement.
For example, Wimpory et al. (2000) devised a semi-structured interview which was administered to mothers of matched 2–3-year-old children with and without autism.
There were significant group differences in what the mothers reported to have occurred during the first 2 years of the children’s lives, both in person-to-person interactions (such as the frequency or intensity of eye contact, greetings, and turn-taking) and person–person-world interactions (such as referential looking and pointing to share).
Studies of toddlers and older children with autism have also reported specific impairments in nonverbal communication. These have included abnormalities in coordinating expressions of affect and/or eye contact with other people, for example in contexts involving joint action and attention (Curcio, 1978; Kasari, Sigman, Mundy, & Yirmiya, 1990; Phillips, Baron-Cohen, & Rutter, 1992), face-to-face interaction (Snow, Hertzig, & Shapiro, 1987; Dawson, Hill, Spencer, Galpert, & Watson, 1990), requests (Phillips, Gomez, BaronCohen, Laa, & Riviere, 1995), empathy and social referencing (Sigman, Kasari, Kwon, & Yirmiya, 1992), and self-consciousness (Neuman & Hill, 1978; Dawson & McKissick, 1984).
Stone, Hoffman, Lewis, and Ousley (1994) reported that preschoolers with autism showed deficits not only in imitation and social play, but also in responsiveness to others and expressions of interest in things through eye contact or pointing. Studies by Lord (1984; Lord & Magill-Evans, 1995) have highlighted the paucity of initiations and coordinated expressions in the peer interactions of children with autism.
Each of the above studies highlights one or another aspect of nonverbal communication that may play a role in establishing and maintaining subjective engagement between persons.
A further aspect of intersubjectivity at the focus of the present study—the propensity to identify with someone else—has been the explicit topic of only three published studies to date.
Hobson and Lee (1999) reported striking differences between matched adolescents with and without autism in the propensity to imitate the style with which actions were carried out. This appeared to reflect a failure to ‘link in with’ aspects of the expressive behavior of someone else.
Moreover, participants without autism tended to imitate how the demonstrator used an object held against his shoulder by holding the same object against their shoulder, whereas those with autism tended to position the object in front of themselves on the table.
In further studies of this latter phenomenon, Meyer and Hobson (2004) and Hobson and Meyer (2005) report further evidence that children with autism have a reduced propensity to imitate the self/other orientation of another person’s actions. These results suggest that individuals with autism might lack a propensity to adopt the self-anchored stance-in-acting of someone else, that is, they may be less inclined to identify with someone else and assume the other person’s style of action and self-orientation as their own.