Anticipatory Concern, page 1
Running Head: ANTICIPATORY CONCERN: A STUDY IN AUTISM
Anticipatory concern: A study in autism
Jessica A Hobson, Ruth Harris, Rosa García-Pérez, and R Peter Hobson
Institute of Child Health, University College London and
Tavistock Clinic, London
Authors for Correspondence: Dr Jessica A Hobson
Address: Behavioural and Brain Sciences Unit
Institute of Child Health
30 Guilford Street
London WC1N 1EH
Telephone: +44 (0) 207 905 2162
Fax: +44 (0) 207 831 7850
Word count in body of text (incl Abstract: max allowed 12,000 words): 9,439 words
Anticipatory Concern, page 2
There has been substantial research on children’s empathic responsiveness towards
distressed people, and on the limited responsiveness of children with autism. To date,
however, there have not been experimental studies to test how far children show concern
towards someone who might be expected to feel badly, when that person has not (yet)
expressed any negative feelings. We tested matched groups of children with autism and
learning disability, and typically developing children of similar verbal mental age
(approximately six years), with a novel procedure in which participants witnessed one
person (E1) tearing the drawing of another (E2). In a comparison condition, a blank card
was torn. In the torn-drawing condition, as predicted, fewer participants with autism
orientated towards E2 with an immediate look, and as a group, they were rated as
showing less concern for, and fewer concerned looks towards, E2. We discuss possible
implications for theoretical perspectives on the early development of empathy in typically
as well as atypically developing children.
Anticipatory Concern, page 3
Anticipatory concern: A study in autism
There is a rich tradition of research into young children’s emerging capacities to
show empathy and concern towards other people (see, for example, contributions to
Eisenberg & Strayer, 1987a; also Barrett, Zahn-Waxler, & Cole, 1993; Cole, Barrett, &
Zahn-Waxler, 1992; Zahn-Waxler & Robinson, 1995). Although there have been a
variety of theoretical perspectives concerning the nature and origins of empathy (well
reviewed from an historical perspective by Wispé, 1987, and more recently expressed in
peer commentaries on Preston & de Waal, 2002), as well as dispute over the criteria
appropriate for judging when ‘true’ empathy or sympathy is observed among young
children (Hoffman, 2007; Thompson, 1987), there has been little empirical research to
evaluate children’s manifest affective responsiveness to the plight of someone else whose
(potential) distress is not perceptible in the person’s bodily expressions of emotion. For
the present set of studies, we devised a novel method to explore this specific aspect of
empathic concern from the perspective of developmental psychopathology. Through an
investigation of anticipatory concern among children with autism and matched
individuals without autism – one group with mental retardation and another typically
developing – our aim was to assess the plausibility of the hypothesis that human fellow-
feeling is structured by the propensity to identify with the attitudes of other people.
There are several themes that characterize contemporary theorizing about the
nature and early development of empathy among children. In part, the debates reflect
longstanding controversies about the role of cognitive appraisals in the genesis of
emotions (as expressed, for example, in the exchanges between Zajonc,1984, and
Lazarus, 1984), and more specifically about the cognitive abilities that need to have
Anticipatory Concern, page 4
developed in order for young children to have specific kinds of feeling. In the case of
empathy and sympathy, there is a case for arguing that, as Sroufe (1995, p 127) expresses
it, ‘In many ways true empathy/altruism and hostile aggression draw on the same
cognitive advances – namely, the child’s understanding of the feelings of the other
person’. In particular, it appears that the newfound ability to think about oneself and
others as individual selves around the middle of the second year of life is critical for the
new forms of sympathetic role-taking that are observed at this time (Barresi & Moore,
1996; Hobson, 1993a; Hoffman, 1982; Kagan, 1982; Lewis, 2003, 2004; Moore, 2007).
Among other cognitive accomplishments that have been highlighted as important for
empathy and concern are the ability to take the role of someone else (Feshbach & Roe,
1968), the exercise of the imagination (Harris, 1989), hypothetical thinking and the
availability and use of mental models of alternative psychological stances (Perner, 1990),
and theory of mind or mentalizing abilities (Frith, 2003).
The question arises whether the development of these understandings of self and
other is founded upon earlier forms of affective responsiveness to the feelings of another
person in which infants register the otherness of the other (Hobson, 1993a, b; Hobson,
Chidambi, Lee, & Meyer, 2006; Hoffman, 2007; Strayer, 1987). As Thompson (1987)
has discussed, the dominance of cognitive-developmental perspectives on emotional
awareness and responsiveness, coupled with an emphasis on relatively detached and
intellectually demanding methods to assess empathy, may have underestimated infants’
capacities for feeling towards others who are apprehended, but not conceptualized, as
separate beings. Here it is relevant to note a distinction emphasized by Eisenberg
Anticipatory Concern, page 5
(Eisenberg, 2002; Eisenberg & Strayer, 1987b) concerning empathy, sympathy, and
personal distress. Eisenberg and Strayer (1987b, pp. 5-7) state:
In our view, empathy involves sharing the perceived emotion of another –
“feeling with” another. This vicarious affective reaction may occur as a
response to overt perceptible cues indicative of another’s affective state
(e.g., a person’s facial expressions), or as the consequence of inferring
another’s state on the basis of indirect cues (e.g., the nature of the other’s
situation)…Sympathy is “feeling for” someone, and refers to feelings of
sorrow, or feeling sorry, for another. That is to say, sympathy often
involves feelings of concern, although the conscious cognitive realization
that one is concerned about another’s welfare is an outcome, rather than a
part, of sympathizing. Often sympathy is the consequence of empathizing,
although it may be possible for sympathy (as well as empathy) to result
from processes such as cognitive perspective taking. Whether or not
empathy always mediates sympathizing is an open question.
As this quotation makes clear, the distinction between empathy and sympathy may be
important for some purposes, yet there might be an intimate developmental relation
between the two. If we are to explore this issue, as well as to reconcile differences in
emphasis concerning the relative primacy of emotional and cognitive factors in the
genesis and elaboration of empathy in early development, it will be important to specify
what is already inherent in the structure of basic forms of human social-emotional
experience, and what are the processes through which cognitively articulated
understandings of, as well as responsiveness to, other persons are achieved. The present
Anticipatory Concern, page 6
study of sympathetic concern from the perspective of developmental psychopathology is
intended to contribute to this domain of research.
The Case of Autism
One approach to uncovering the grounding and developmental implications of
young children’s propensity to empathize with others, is to study conditions in which
there appears to be a diminished capacity for such responsiveness to other people’s
feelings. Early childhood autism represents the most celebrated case in point. When
Kanner (1943) first identified the syndrome, he characterized his cohort of 11 children as
having ‘inborn autistic disturbances of affective contact’ with other people (p. 250).
Kanner attempted to capture the children’s stance in relation to others by writing of their
‘profound aloneness’, and conveyed how ‘people, so long as they left the child alone,
figured in about the same manner as did the desk, the bookshelf, or the filing cabinet’ (p.
246). Such descriptions have been amplified by other clinical accounts such as that of
Bosch (1970), who concluded that in autism, a “delay occurs in the constituting of the
other person in whose place I can put myself” (Bosch, 1970, p. 89).
These observations highlight how empathic human relatedness towards others – a
pervasive feature of interpersonal engagement, not reducible to the more specific
potential to show concern toward people in distress – appears to be striking for its relative
lack among many children with autism. In the last two decades, there has been
substantial empirical and theoretical progress in tracing how impairments in
intersubjective person-with-person co-ordination and communication might be pivotal for
a range of the children’s intellectual as well as social difficulties.
Anticipatory Concern, page 7
The sources of empirical evidence are several. Parental reports (e.g. Dahlgren
and Gillberg, 1989; Hobson et al., 2006; Lord, Storoschuk, Rutter, & Pickles, 1993;
Stone & Lemanek, 1990; Vostanis et al., 1998; Wimpory, Hobson, Williams, & Nash,
2000; Wing, 1969) reveal that young children with autism are relatively unresponsive to
other people’s non-verbal as well as verbal communication, and that such abnormalities
occur in one-to-one affective engagement as well as in joint attention and other forms of
co-reference towards a shared world. In a study by Wimpory and colleagues (2000), for
example, where parents of matched young children with and without autism were asked
to describe their offspring in the first two years of life, not one of the infants with autism
were reported to have shown frequent and intense eye contact or engaged in turn-taking
with adults, and there were also fewer infants with autism who greeted or waved to their
parents, or who directed feelings of anger and distress towards people. When
interviewed in a recent study by Hobson et al. (2006), parents described how their
children with autism showed jealousy towards others and were affected by others’
moods, but far fewer were reported to show concern or guilt towards other people than
were matched children without autism. As one parent responded when asked if her son
showed concern if she were upset: “He might be worried but he doesn’t have that
empathy sort of concern – he doesn’t show that at all… Empathetic sadness isn’t there”
(Hobson et al., 2006, p 67). Hobson and colleagues concluded that it is especially in
expressing ‘person-centred’ feelings – that is, feelings for and in relation to other people
– that children with autism are unusual.
Systematic observational and experimental studies of children and adolescents
with autism provide complementary evidence for limitations in their affective relatedness
Anticipatory Concern, page 8
towards other people (e.g., Attwood, Frith, & Hermelin, 1988; Dawson et al., 1990;
Hobson & Lee, 1998; Lord, 1995; Snow, Hertzig, & Shapiro, 1987). Sigman, Kasari,
Kwon, & Yirmiya (1992) conducted direct tests of young children’s empathic
responsiveness and concern in relation to other people’s expressions of distress or other
negative feelings. Participants were 30 young autistic children with a mean age under
four years, closely matched non-autistic children with learning disabilities and typically
developing children. The technique was to code these children's behaviour when an adult
(in some conditions the child’s parent) pretended to hurt herself by hitting her finger with
a hammer, simulated fear towards a remote-controlled robot, and pretended to be ill by
lying down on a couch for a minute, feigning discomfort. In each of these situations,
children with autism were unusual in rarely looking at or relating to the adult (also
Charman, Swettenham, Baron-Cohen, Cox, Baird, & Drew, 1997, for similar studies with
20-month-olds). In one of a set of quasi-experimental studies, Hobson et al. (2006)
investigated how children with autism responded to a situation that might be expected to
elicit feelings of guilt. Although similar numbers of children in each of the matched
groups made attempts to repair an object that broke in their hands, and similar numbers
showed a negative emotional response, there was a highly significant group difference in
ratings of guilt. For example, only 2 out of 12 participants with autism but 10 out of 12
participants without autism showed “guilty looks” involving a gaze pattern towards
and/or away from the tester with accompanying anxiety and reassurance-seeking that
tended to elicit the raters’ sympathy.
These studies of real-life exchanges are complemented on the one hand by a range
of experimental studies suggesting autism-specific impairments in the perception and
Anticipatory Concern, page 9
expression of emotion (as reviewed by Hobson, 2005), and on the other, interviews with
more able children and adolescents with autism that provide telling detail of what the
children appear to experience in relation to others (Bauminger & Kasari, 2000; Lee &
Hobson, 1998; Yirmiya, Sigman, Kasari, & Mundy, 1992; and Baron-Cohen, Richler,
Bisarya, Gurunathan, & Wheelright, 2003, for a questionnaire approach). Kasari,
Chamberlain, and Bauminger (2001) described how high-IQ children with autism could
report feeling guilt, but compared with control children they provided fewer self-
evaluative statements and fewer instances of empathic guilt, and were more likely to
describe situations in terms of rule-breaking, disruptiveness and damage to property,
rather than those of causing physical or emotional harm to others
The hypothesis that underpins the present study focuses upon the quality of
intersubjective impairment that might account for the contrast between individuals with
and without autism in the structuring of interpersonal relations. One critical aspect of
empathy and related feelings of concern and guilt (and arguably, other emotions that
entail engagement with other people’s attitudes such as embarrassment, coyness, shame,
and mutual joy) is that a person does more than respond to someone emotionally: he or
she engages with the other person’s feelings as the other’s feelings. Our hypothesis is that
(most) individuals with autism have specific limitations in experiencing and manifesting
‘person-centred’ feelings such as concern for the reason that such relations are configured
by the propensity to identify with the psychological stance of another person (a theoretical
position partly founded upon Freud, 1955/1921, and elaborated in Hobson, 1993; Hobson
et al., 2006; Hobson, 2007). The important thing about this process is that one has
feelings that are congruent with those of the person identified-with, but one also retains
Anticipatory Concern, page 10
an affective relation towards this other as experienced. For example, if one identifies
with a person’s suffering or anger, one might still feel concern or fear towards the person
identified-with; and the concern or fear is felt in virtue of the other being experienced as
suffering or angry, through identification. In keeping with phenomenological approaches
to the nature of interpersonal responsiveness (Merleau-Ponty, 1964; Stein, 1989), it is
postulated that such modes of apprehending the emotional stance of others underpin
(rather than presuppose) one’s concepts of people’s minds, or so-called Theory of Mind,
and are the basis for (not the result of) cognitively articulated role-taking. In addition,
this form of self/other structuring of social experience leads one to anticipate - and where
appropriate, orientate towards and engage with – feelings that belong to someone else,
even in the absence of perceptual evidence of the other person’s state. According to this
hypothesis, then, identifying-with is a basic unit of analysis of social relatedness, and
limitations in this process are pivotal for an adequate account of the developmental
psychopathology of autism.
The Present Study
Whereas previous quasi-experimental studies have addressed children’s potential
for empathy through their manifestations of concern when someone else shows distress,
we were interested in evaluating whether they would show concern in anticipation of the
person having a negative experience, prior to that person showing any explicit reaction to
potentially distressing circumstances. Here any group differences could not be attributed
to children’s failure to perceive or affectively respond to another person’s expressions or
other manifestations of a state of distress, for none were shown. Rather, this was a test of
whether participants apprehended the situation as one that would hurt another person’s
Anticipatory Concern, page 11
feelings as it would their own. If they were to do so, and if they had the potential to care
about the person, then they might express and/or convey their concern in anticipation of
the reaction they expected.
Thus our aim was to test whether the children would identify with another
person’s situation and experiences, and in so doing, express sympathy or direct other
expressions of fellow-feeling towards the person in question. We predicted that when
participants saw an (unreactive) person’s drawing torn by someone else, children with
autism would be less inclined than matched children without autism to a) orientate
towards that person immediately and b) show ‘person-centred’ sympathy or concern. We
employed two measures of such empathic responsiveness, complementing global ratings
of the children’s expressions of concern for the person, with ratings of the qualities of
looks towards the person.
Many of the quasi-experimental studies cited above on the social relatedness of
children with autism, as well as research on sympathy among typically developing
children, do not include control conditions. For many intents and purposes, for instance in
demonstrating children’s marked or limited responsiveness to other people’s expressions
of feeling, a control condition is necessary only to determine the specificity of
participants’ reactions to any particular context. For the present purposes, too, we
considered that the torn-drawing task was a stand-alone condition, and being central to
the study and the focus of our predictions (and potentially influenced if children had had
prior experience of a similar set of events), it was administered first to all participants.
Subsequently, as a subsidiary procedure, we also administered a comparison condition in
Anticipatory Concern, page 12
which there was one critical difference from the Empathy condition, namely that a blank
card instead of a person’s drawing was torn.
We tested three groups of children, comprising a) 20 children (16 boys, 4 girls)
with autism, b) 18 matched children (13 boys, 5 girls) without autism but with learning
disabilities, mild to moderate mental retardation, and/or developmental delays, and c) 14
children (5 boys, 9 girls) who were typically developing, and whose mean verbal mental
age (6 years) was similar to that of the other two groups.
Participants with autism displayed impairments in social interaction and
communication, coupled with repetitive or stereotyped interests and activities,
characteristic of the disorder. We confirmed the clinical diagnosis by completing a DSM-
IV criteria checklist (American Psychiatric Association, 1994) on the basis of systematic
interviews with teachers, and by rating classroom observations of the children using the
Childhood Autism Rating Scale (CARS; Schopler, Reichler, & Renner, 1988). The
CARS covers such domains as social relatedness, verbal and nonverbal communication,
repetitive behaviour, sensory abnormalities, and emotion regulation. Scores of 30 or
above are in keeping with a diagnosis of autism. On the CARS, children with autism
received scores ranging from 26.5 – 54.5 (M = 37.2, SD = 7.3). The one participant who
scored below 30 was an adolescent girl. In order to clarify her diagnostic status, we
administered the Autism Diagnostic Observation Schedule (ADOS; Lord, Rutter,
Dilavore, & Risi, 1999), Module 3. Her scores on the ADOS – a series of semi-structured
presses designed to assess for the presence of social and communication difficulties –
Anticipatory Concern, page 13
were consistent with the clinical diagnosis of autism and therefore she was retained in the
The children with autism were group-matched with those in the learning disability
(LD) group for chronological age and language performance on the British Picture
Vocabulary Scales (BPVS; Dunn, Dunn, & Whetton, 1982), the British version of the
Peabody Picture Vocabulary Scale. The BPVS is a standardised, widely used measure of
receptive vocabulary, assessing a cognitive ability that is relatively impaired in persons
with autism (Jarrold, Boucher, & Russell, 1997; Lockyer & Rutter, 1970). The typically
developing children were selected on the basis that their mean chronological age was
similar to the mean verbal mental age of the children with autism and those with learning
disabilities. To confirm their levels of language ability, they were assessed using the
Preschool Language Scale (PLS–3–UK, Zimmerman et al., 1997) which has been
developed for use with children of this age and offers standard scores, percentile ranks,
and language age equivalents.
Participant characteristics appear in Table 1.
Table 1 about here.
There were two parts to the procedure. The first part was the Empathy condition
in which a person’s drawing was torn. The second part was a Comparison condition in
which there was a similar sequence of events, except that a blank piece of paper was torn
instead of the person’s drawing. As already discussed, we decided to administer the
Anticipatory Concern, page 14
Empathy condition first, for the reason that this was the critical condition for studying
participants’ expressions of concern, and all the principal analyses of group differences
involved data from this condition and did not involve analyses of group by task
interactions. We considered that children who had already received the Comparison
condition might have had an altered reaction to what was supposed to be a novel (and
potentially shocking) event if many aspects of the set-up were already familiar. In order
to lessen the chance of interference effects on participants’ responses in the Comparison
condition – which was not intended as a strict control condition, but an evaluation of
participants’ orientation and emotional reaction towards E2 when the tearing of a piece of
paper should have no special meaning for E2 - we ensured that for the children with
autism and those with learning disabilities, there was a gap of six months between
administration of the Empathy and Comparison conditions. This was not possible in the
case of the typically developing children due to constraints on their availability, and for
these participants we conducted the two procedures in the same testing session, with an
additional set of activities interpolated between the two.
Empathy condition. Two female testers, E1 and E2, each familiar to the children
as regular visitors to their classroom, invited the children to participate individually in
videotaped sessions that took place in quiet rooms in their own schools. E1 (the
perpetrator of the potentially hurtful act) sat beside the child at a table, directly across
from E2 (the victim). For most of the time, all three testers appeared on videotape, but
the camera was facing the participant and E1 in order to capture the participant’s
reactions. The arrangement was such that it was always clear when the participant
looked to E2’s face or elsewhere.
Anticipatory Concern, page 15
Partly in order to establish a setting in which empathic responses might be
anticipated, the testers conducted the procedure in an atmosphere of joint enterprise and
mutual involvement. E2 distributed blank 4” x 6” index cards to each person present, and
placed a box of colored pens in the center of the table. She said: “Let’s each draw an
animal.” E1 drew a rabbit, E2 drew a turtle, and the child drew an animal of his or her
own choice. In order to engage the children with the testers, as well as to familiarize the
children with the drawings of each, E1 announced that she was drawing a purple rabbit,
and E2 showed her turtle to the child when it was almost complete (with the exception of
a tail) and asked the child, “Do turtles have tails?” before adding a tail to the turtle. She
also complimented E1 on her rabbit. Meanwhile, both testers asked the child about his or
her drawing as it was being drawn, and gave praise when it was completed.
Once the drawings were completed, E2 gathered them together and placed them
into transparent sleeves with a zip across the top. As a brief game, these folders were
mixed around and the child was asked to identify his or her own drawing. Then E2
placed the three folders into a basket which she gave to E1, and asked E1 to remove the
drawings from their sleeves and put them away. E2 turned away and prepared materials
for another purpose.
At this point E1, seated beside the child, secured the child’s attention, and one by
one removed the drawings from the sleeves. First she held up her own drawing and said,
“This is the rabbit that I drew.” Next she held up the child’s drawing and stated, “This is
the [named animal] that you drew.” Finally, she held up E2’s drawing and, once again
having checked that the child was attending, stated, “and this is the turtle that [name of
E2] drew, isn’t it?”
Anticipatory Concern, page 16
Once the child affirmed that this was the case, E2 looked up to observe what was
happening. At this moment, with a neutral expression and without acknowledgement
towards E2, E1 matter-of-factly tore E2’s drawing of the turtle into four pieces. E2 was
watching as this happened, but sat motionless and maintained a neutral expression. E1
and E2 did not look at each other, and remained still for about three seconds, so that it
was possible to capture the child’s spontaneous reactions on videotape.
At the conclusion of this very brief period following the tearing of the drawing,
the testers talked with the participant about what had happened, and offered reassurance
that E2 had told E1 that she did not want to keep her drawing, and had asked that it be
thrown away. The duration of this interaction depended upon participants’ responses, in
that in some cases the children asked for explanation, made comments, or even protected
or tore up their own drawings.
Comparison condition. All participants (with the exception of five participants
with autism and two with learning disabilities who were not available for testing on a
second visit) received a Comparison condition to confirm that any empathic reactions
they had shown in the Empathy condition were related to the potential feelings of the
person whose drawing was torn. Here the procedure was similar to that already
described, except that an additional blank index card had been added to the basket so that
this time, E1 gained the child’s attention not only for each animal drawing, but also for
the blank card. Upon discovering the blank card, E1 stated, “This one is blank”, before
tearing the blank drawing into four pieces. There was a further modification for the
typically developing children because they were given both conditions on the same day.
Anticipatory Concern, page 17
Here, the testers drew animals (a lion and a dog) that were different to those they had
Ratings of the Videotapes
The videotapes were edited into brief excerpts. Each excerpt began when E1
said, ‘This one is blank’ in the Comparison condition, or ‘This is the drawing that E2
drew’ in the Empathy condition, immediately prior to the tearing episode. We included
this part of the procedure in the excerpt, because there was plenty of indication in the
ensuing exchanges between participants and testers whether or not E2’s drawing had
been torn (see illustrative vignettes and ratings of these interactions in Tables 2-4), and so
it was not possible to achieve ‘blindness’ of raters in this respect. In addition, we wanted
to confirm that all participants were attentive to the actions of E1 at this point, which
proved to be the case in all instances. The excerpt ended following the child’s response
to the tearing and the testers’ subsequent explanation and/or reassurance. Although
certain of our ratings pertain to the moments immediately after the tearing event, others
concern affective responses that waned only after subsequent communicative exchanges
with the children had been completed, and they had been reassured that E2 was not
troubled by what had happened (again, Tables 2-4 for examples).
Participants’ responses to E2 following E1’s tearing of the paper were rated by
judges who were blind to the children’s diagnoses and the hypotheses and predictions of
the study. There were two sets of ratings by separate pairs of raters – one pair recorded
whether participants gave a spontaneous and immediate look to E2, and made a rating of
overall concern; the second pair made ratings of the occurrence and qualities of looks,
Anticipatory Concern, page 18
across the videotape excerpt. The ratings, with estimates of inter-rater reliabilities, were
Spontaneous look to E2. The first set of judges were asked to indicate whether or
not the child showed a spontaneous look to E2 during or immediately following (i.e.
within one second) the tearing of the paper. On the Comparison condition (45
videotapes), kappa = .83; and on the Empathy condition (52 videotapes) there was 100%
agreement between the two judges regarding the presence or absence of such a look.
Degree of concern for E2. The first pair of judges was also asked to rate the
degree of concern for E2 shown by the participant. The possible scores were 0 (No
concern), 1 (Limited concern) or 2 (Clear concern). The estimate of inter-rater reliability
for the Empathy condition (52 videotapes) was kappa = .78; for the Comparison
condition (45 videotapes), there was 100% agreement. Tables 2 – 4 provide illustrations
of participants’ reactions that were given scores of 0, 1, or 2. These descriptions were
written by one of the raters, while blind to the children’s diagnoses.
Tables 2 – 4 about here.
Looks to E2. The second pair of judges was asked to look at the videotape
excerpts and note each time the child looked to E2. On counts of the numbers of looks
made by each participant, for each condition there was almost perfect inter-rater
reliability (for the Empathy condition, ICC = .99; for the Comparsion condition, ICC =
.95). For those very few instances where one rater had judged a look to be present whilst
the other had not, the judges were asked to review the specific looks together and decide
Anticipatory Concern, page 19
jointly whether or not eye contact had occurred. The agreed instances of looks were used
for the ratings on quality of eye contact.
Quality of each look to E2. Here, each look was categorized either as being a
concerned look, or a non-concerned look of three possible kinds. A concerned look was
defined as expressing interpersonal contact with empathic concern for the tester’s
feelings, including a sense of discomfort on her behalf. For the present purposes, the
three remaining types of looks (looks to establish reciprocal contact and to share what
was happening, usually with positive affect; looks to ‘check’ the situation and/or E2’s
response; and looks prompted by the behaviour or comments of the testers) will not be
considered separately, because there was some ambiguity among these types of look
when the testers and participants were engaged in dialogue, and for the present purposes,
the critical feature of these looks is that they did not include expressions of concern.
In order to train the raters to arrive at a shared understanding of the definitions of
the different kinds of look, we discussed seven of the 155 looks with the raters together.
On the remaining 147 looks that were rated independently, estimates of inter-rater
reliability were kappa = .82 for concerned vs. not concerned looks.
There was a categorical yes/no judgment for the presence of an immediate
spontaneous look to E2, and scores for the degree of concern for E2 were limited in range
(0 – 2). Therefore we employed non-parametric analyses for these data, as we did for
subsequent group comparisons of the numbers of individuals showing any concerned
Anticipatory Concern, page 20
In order to analyze the data on rates of looks per minute (frequencies), for which
parametric statistics were employed, it was necessary to adjust for variability in the total
lengths of the videotape excerpts. In particular, the duration of the Comparison condition
tended to be short, principally because there was little dialogue or need to reassure the
children after the blank card was torn (for participants with autism, M = 5.9 seconds, SD
= 1.4 seconds, range = 4 to 9 seconds; for LD participants, M = 5.7 seconds, SD = 1.9
seconds, range = 4 to 11 seconds; for typically developing participants, M = 9.5 seconds,
SD = 2.1 seconds, range = 7 to 15 seconds). In the Empathy condition, the videotape
excerpts were very similar in length for participants with autism and those with learning
disabilities (for children with autism, M = 16.1 seconds, SD = 10.9 seconds, range = 5 –
48 seconds; for those with learning disabilities, the M = 15.1 seconds, SD = 5.6 seconds,
range = 6 – 24 seconds). However, for the younger, typically developing children the
mean length was substantially longer at M = 25.2 seconds, SD = 7.3 seconds, range = 17
– 45 seconds. Therefore in order to compare the three groups for the prevalence of
different kinds of looks to E2, we computed rates of looks per minute. Although we had
made directional predictions as described earlier, all significance values are reported at
Spontaneous and Immediate Look to E2
On the Comparison condition involving a blank drawing, there were only a small
number of participants (1 with autism, 4 LD, and 3 TD) who looked at E2 spontaneously
during or immediately after the blank card was torn (χ2= 1.97, df = 2, ns).
On the Empathy condition, only 2 out of 20 participants with autism gave an
immediate, spontaneous look to E2 when her drawing was torn. In contrast, 11 out of 18
Anticipatory Concern, page 21
children with LD, and 11 out of 14 TD children, gave an immediate look to E2 (χ2=
18.06, df = 2, p < .001). Children with autism were significantly different from both LD
and TD participants (Fisher’s Exact p = .002 and p = .000, respectively). Obviously, the
two groups without autism were not significantly different from one another.
Degree of Concern for E2
The results appear in Figure 1. In the blank drawing condition, not a single
participant in any group was rated as showing concern for E2. On the Empathy
Condition, by contrast, there were significant group differences (Kruskal-Wallis Test χ2=
15.67, df = 2, p < .001). For example, three-quarters of participants with autism fell into
the category of showing no concern, whereas the majority of children in the LD and TD
groups fell into the category of showing clear concern. As predicted, children and
adolescents with autism were less likely to be judged as showing concern relative to those
with learning disabilities, Mann-Whitney U = 80.5, z = 3.23, p < .01, and relative to those
who were typically developing, Mann-Whitney U = 33.0, z = 4.07, p < .001. LD and TD
participants were not different in this respect (Mann-Whitney U = 106.5, z = .88, ns).
Figure 1 about here.
Frequency of Looks to E2
In the Comparison condition with a blank card, there were significant group
differences in rates of looking per minute, F(2, 44) = 8.46, p < .001. Although there was
a non-significant trend for children with autism to look towards E2 less often than those
with learning disabilities, t(29) = 1.93, p < .10, the principal contrasts were that typically
Anticipatory Concern, page 22
developing children looked at E2 significantly less than children with autism, t(27) =
2.62, p < .05 as well as children with learning disabilities, t(28) = 3.93, p < .01.
The results for the Empathy condition appear in Figure 2. Once again, the groups
were significantly different in terms of their rates of looking to E2, after E2’s drawing
was torn, F(2, 51) = 6.44, p < .01. In this case, children with autism looked to E2 less
often than the children with learning disabilities, t(36) = 3.35, p < .01 as well as those
who were typically developing, t(32) = 2.29, p < .05. LD and TD participants were not
significantly different from one another, t(30) = 1.25, ns.
Inspection of Figure 2 reveals how the pattern of responses across the two
conditions were different for the TD children, in that they were much less likely to look
at E2 when the blank card was torn than when her drawing was torn. There is a
significant group (3) x task (2) interaction for rates per minute of looking in the
Comparison and Empathy conditions, F(2, 42) = 9.46, p < .001. For participants with
autism and LD, there was not a significant difference between rates per minute of looking
across conditions, t(14) and t(16) = 1.2 and .58, ns. For those in the TD group, there was
a significant difference, t(14) = 4.73, p < .001.
Figure 2 about here.
Although the children with autism showed significantly fewer looks than the other
participants only in the Empathy condition, it was also the case that they showed
relatively few looks in the Comparison condition. In order to assess the meaning of these
results, as well as those concerning the similarities in frequencies of looking across
Anticipatory Concern, page 23
conditions by participants with autism and LD, it was important to examine whether there
were group differences in the qualities of looks.
Quality of Looks to E2
We adopted two approaches to analyzing the quality of looks. Firstly, we
considered whether participants showed any concerned and/or other kinds of look in the
two conditions. Secondly, we analyzed the rates and proportions of different kinds of
look that were shown.
In the Comparison condition, there was only a single participant, a girl with a
learning disability, who was judged to have made a concerned look, and she made only
one. All of the remaining looks were rated as non-concerned.
In the Empathy condition, only three out of the 20 (15%) participants with autism
showed at least one concerned look, whereas 12 out of 18 (67%) of those in the LD group
and 100% of the TD children did so (χ2= 25.44, df = 2, p < .001). In this respect,
participants with autism were significantly different from those with LD (Fisher’s Exact p
= .002) and TD children (Fisher’s Exact p = .000). TD participants were also more likely
to show concerned looks than the LD children (Fisher’s Exact p = .024). This profile of
group differences appeared specific to concerned looks, because there were 11 out of 20
children with autism, 11 out of 18 LD participants, and six out of 14 TD participants who
showed at least one non-concerned look during the Empathy condition (χ2= 1.07, df = 2,
Participants with autism who showed concern. Two of the three children with
autism who showed a concerned look were also judged to show concern according to the
global rating made by the separate raters. The first of these participants was a male aged
Anticipatory Concern, page 24
13 years 4 months with a verbal mental age of 14 years 3 months and a CARS score of
32. He had clinical features of Asperger syndrome and was very sociable and friendly.
The second individual was a soft-spoken, shy boy aged 10 years 5 months with a verbal
mental age of 7 years 10 months and a CARS score of 31.5. The third child with autism
judged as showing a concerned look by one set of raters was judged by the other raters as
not showing any concern at all. She was a female aged 14 years 6 months with a verbal
mental age of 9 years 9 months and a CARS score of 35.5. The clinical description of her
response to the tearing, written by one of the raters (still blind to diagnosis) who judged
her as showing no concern, illustrates how there was ambiguity in whether the look
expressed concern for E2:
She has a neutral expression and her arms are crossed in front of
her body. She nods and says, “yeah” when shown E2’s drawing.
Her eyes widen slightly as E1 tears the drawing. She then asks
E1, “why you ripping it for?” Her tone of voice suggests that she
doesn’t approve of what E1 is doing. She stares at E1 and then at
Specificity of concerned looks. To confirm the specificity of the dearth of
concerned looks among participants with autism, we conducted a further analysis in
which we considered only those participants who showed at least one look during the
Empathy condition. There were 12 of the 20 participants with autism, all 18 of those with
learning disabilities, and all 14 of those who were typically developing who showed a
look. Here there were group differences in the percentages of looks judged to be
concerned, F(2, 43) = 13.85, p < .001. Among participants with autism, 17% of their
Anticipatory Concern, page 25
looks were rated as concerned looks, whereas this was the case for 55% of the looks for
LD participants, t(28) = 2.55, p < .05), and 87% of the looks made by TD children, t(24)
= 7.4, p < .001 (and this latter group also showed a higher proportion than the LD
participants, t(30) = 2.64, p < .05). These results indicate that even in relation to overall
levels of looking, it was specifically in relation to concerned looks that participants with
autism were atypical.
This pattern of results needs to be considered in relation to the finding that across
the Comparison and Empathy conditions, neither among participants with autism, nor
among those with LD, was there a significant difference in frequencies of looking.
Considered in isolation, this result might have suggested that the two conditions were
little different in their effects on looking, or that the two groups were affected in similar
ways by the two conditions. However, for participants with autism, the proportions of
concerned looks were 0% in the Comparison condition and 17% of the looks in the
Empathy condition; for LD participants, by contrast, the proportion of concerned looks
were 4% and 55%, respectively. This indicates how among the LD participants but less
so for those with autism, the Empathy condition produced a very different profile of
looking, even though the frequency of looks overall changed relatively little within each
This picture is filled out by analyses of rates per minute of participants’ looks.
The results on the Empathy condition appear in Figure 3. Here it may be observed that
although it was very rare for participants with autism to show relatively high rates of
concerned look, this was common for LD and TD participants. In this same condition,
the rates per minute of non-concerned looks to E2 are shown in Figure 4. Here there was
Anticipatory Concern, page 26
little to distinguish the groups. The one feature that distinguishes the typically
developing children is that none showed more than 7 non-concerned looks per minute,
whereas a minority of participants in the other two groups did so.
Figures 3 and 4 about here.
For this study we devised a novel methodological approach to assess whether
children with autism, those with learning disabilities, and others who were typically
developing, showed concern for another person in a relatively natural (albeit contrived)
situation. The setting was one in which there was no requirement - and indeed, no
opportunity – for participants to perceive and respond to another person’s bodily
expressions of distress, because there were none. Participants simply saw someone else
tear a person’s drawing, in the person’s presence. The critical question was whether they
would react to this event by turning to look at the person who might be affected, and
show concern towards her.
Our principal focus was upon the potentially limited responsiveness of the
participants with autism. The results were clear-cut, and in keeping with our predictions:
in contrast with matched children with learning disabilities on the one hand, and young
typically developing children on the other, it was rare for participants with autism to
show spontaneous, immediate looks to the person. Moreover, as judged both by reliable
‘subjective’ ratings of concern, and by counts of ‘concerned looks’ towards the other
person, participants with autism were less likely to show concern for the person. At this
Anticipatory Concern, page 27
level of description, then, the study demonstrates how children with autism were unusual
in showing little orientation towards, or concern for, another person who might be
expected to feel distress.
It was not merely the tearing of a blank piece of paper that elicited a reaction from
participants, because in the Comparison condition, unsurprisingly, very few of the
children of any group showed concern for anyone. We had decided not to counterbalance
the order of the Empathy and Comparison conditions because the Empathy condition was
critical to the group comparisons, and commonsense suggested that there would be little
reason for participants to orientate to E2 when a blank card was torn. On the other hand,
prior experience with this condition might have influenced participants’ subsequent
behaviour if the Empathy condition had followed. Especially given that the Comparison
condition was administered six months after the Empathy condition, it is most unlikely
that participants’ almost total lack of concerned looks to either adult in the Comparison
condition was due to order effects. In the case of the TD participants, who received the
two conditions on the same day, it remains possible that their earlier experience with the
Empathy condition increased their seeming disinterest when the blank card was torn; but
among the participants with LD, too, and even among a very small minority of those with
autism, there was clear evidence from the timing and quality of their looks, as well as
global ratings of concern, that the two conditions were experienced in very different
Yet this appeared to be less the case for participants with autism, who rarely
looked to E2 while or immediately after her drawing was torn, and who showed fewer
concerned looks in the subsequent phases of the procedure. This was not explicable in
Anticipatory Concern, page 28
terms of a lack of looks per se, for the reason that compared with participants without
autism, a significantly smaller proportion of their looks were concerned in quality. In
addition, global ratings of concern encompassed evidence beyond looking patterns, for
instance in what participants expressed in words, and while three-quarters of this group
showed no evidence of concern, this was the case for only a minority of the LD
participants and almost none of the typically developing children. An interesting
question remains about the motivation for the looks to E2 that were made by children
with autism, and mostly such looks appeared to be checking out what might happen next,
or to be prompted by tester-initiated interaction.
There are, of course, several ways to interpret these group differences. For
example, it might be argued that children with autism tend to be undemonstrative in their
feelings in many social and even non-social settings, so it is unjustifiable to suppose that
there was specificity to their apparent indifference in the present circumstances. There
are several reasons why this (hypothetical) explanation is implausible. Firstly, although
there is evidence to suggest that sometimes children with autism show their feelings in
atypical ways (e.g., MacDonald et al., 1989; Ricks, 1975; Yirmiya, Kasari, Sigman, &
Mundy, 1989), and that in a variety of social situations their orientation and
expressiveness towards people is diminished (as illustrated in the Introduction, and as one
would expect if their empathic engagement with others is limited), it is also the case that
children with autism of the age and ability tested here show clear manifestations of
certain feelings, even seemingly complex and socially orientated feelings such as
jealousy (Bauminger, 2004). For example, the systematic semi-structured interviews
with parents undertaken by Hobson et al. (2006) yielded evidence that children with
Anticipatory Concern, page 29
autism manifest clear signs of jealousy and the emotions of happiness, distress, anger and
fear, as well as emotional responsiveness to the moods of others; and subsequent quasi-
experimental ‘real-life’ observations of children and adolescents with autism recorded
abundant signs of self-consciousness as well as coherent behavioural responses to events
such as a doll’s leg breaking off, praise of the child’s drawing, and nuzzling by a cuddly
Teddy. Studies by Capps, Kehres, and Sigman (1998), Dawson and Adams (1984), and
García-Pérez, Lee, & Hobson (2007) provide further examples of children with autism
manifesting organized, if not wholly typical, affective responsiveness in situations of
social interaction and one-to-one conversation. In view of these findings, it is difficult to
propose that the present group differences reflect emotional ‘flatness’ per se.
An alternative, more refined position would be to argue that rather than
demonstrating how children with autism have little empathy and/or concern for others,
the present study simply shows that the children do not care about the fate of a person’s
drawings, whether one of their own or one sketched by someone else. Under these
circumstances, no wonder if they failed to show a reaction in the circumstances we had
contrived. Although ethical considerations would preclude a control condition in which
the children’s own drawings were torn in order to see if they were upset, one might have
devised further settings in which the adult might be expected to feel distressed. Yet
again, the present results are in keeping both with clinical descriptions of children with
autism, and with parental reports of lack of concern – in both of which, a variety of
circumstances have been cited as instances in which concern might have been expected,
but was unforthcoming. For example, one parent of a child with autism reported of her
son, “When I’m sad, it disturbs him, he doesn’t quite know what to do and then he just
Anticipatory Concern, page 30
looks and if I don’t say anything, he just moves away. A normal child would ask or say
what is happening, he wouldn’t.” (Hobson et al., 2006, p 67).
There is also evidence that for children with autism of the age and ability tested
here, drawings are personally meaningful. In one of the conditions devised by Hobson et
al. (2006), participants with autism were praised for a drawing they had completed. In a
control condition, they were praised for a drawing someone else had made. Here there
was a significant effect of condition, in that participants with as well as without autism
displayed more pride in the former rather than the control condition. Bauminger (2004)
reported how children with autism showed jealousy when their parents praised another
child’s drawing and not their own. Kasari et al. (1993) described how children with
autism showed pride accomplishing a puzzle, and again this attests to their experience of
pleasure in what they produce. These results speak not only to the question of whether
children with autism show any feelings, but also to the personal significance for the
children themselves of having completed a drawing.
For at least one empirical and one theoretical reason, it is important to note that
this was not so much a study of ‘affective expressiveness’ among the participants, but
rather, a study of a particular form of interpersonal relatedness, namely that of showing
concern towards a person. One reason this is important is that the measures we employed
extend beyond those that are conventionally considered instances of affective expression.
For example, we recorded immediate and spontaneous looks to the person whose drawing
was torn, regardless of its affective quality; and we took note of where participants
looked subsequently, and how they behaved. The point is that concern is manifest in a
combination of expressions and actions, and the pattern of relatedness that concern
Anticipatory Concern, page 31
involves was relatively absent among the participants with autism, in contrast to
participants from the other groups – and indeed, Vaish and Tomasello (2007) have
employed a modified version of our ‘torn drawing’ task to demonstrate prosocial
behaviour as well as sympathy towards a victim among typically developing 18- and 24-
month-olds. From a theoretical standpoint, this focus on interpersonal relatedness
highlights how we need to avoid pre-judging whether or not humans are biologically
predisposed to have forms of feeling that are organized in relation to persons from early
in development, rather than (for example) supposing that infants begin with emotions like
pleasure, distress and frustration that become focussed upon and oriented towards people
through domain-general forms of learning, and only at subsequent stages of development.
This, indeed, is at the core of the hypothesis that children with autism have a
relative lack of such person-related organization to their affective states. Evidence that
children with autism have a limited propensity to identify with another person’s attitudes
comes from a range of studies beyond those of affective responsiveness cited in the
Introduction. In particular, children with autism have diminished inclination to imitate
the self/other-orientated aspects of action (Hobson & Meyer, 2005; Meyer & Hobson,
2004) with an associated dearth of ‘sharing looks’ in joint attention (Hobson & Hobson,
2007); they rarely imitate another person’s style of executing actions (Hobson & Lee,
1999); they show a relative lack of head-nodding when another person speaks to them
(García-Pérez, Lee, & Hobson, 2007); they have inconsistent role-taking when re-casting
stories (García-Pérez, Hobson, & Lee, 2007); they show little discrimination in their
drawings of human figures but not houses (Lee & Hobson, 2006); and they are
distinctive in communicating with a relative lack of emotional engagement, sharing
Anticipatory Concern, page 32
experience in joint attention, communication of style, and shifting in communicative role
(Hobson, Lee, & Hobson, 2007).
We have argued that such abnormalities in autism reflect a relative lack of what in
children without autism is a motivational-cum-affective process through which one
person has the propensity to register and in part assimilate – and at the same time,
respond to – the bodily-anchored psychological stance of another person, either as
perceived or as imagined. We consider this to be a basic human psychological
propensity, where ‘basic’ means that on a psychological level, the process is non-
reducible to other terms, is almost specific to the social domain (although not entirely so,
as Lipps argued), and is developmentally prior to, and formative of, many aspects of
subsequent social and social-cognitive development (see Bråten, 1998; Cooley, 1902;
Fairbairn, 1952; Meltzoff & Brooks, 2001; Tomasello, 2005; Trevarthen, 1979;
andTrevarthen & Aitken, 2001, for related lines of thinking). Indeed, it would be
valuable if future studies addressed the conditions under which identification might be
facilitated or inhibited among children with autism – for example, by testing whether it
would make a difference if parents took the place of experimenters in the ‘torn drawing’
Such considerations return us to the question of how far in typical development,
sympathy and concern are dependent upon empathy on the one hand, and conceptual
development on the other. According to the present account, one manifestation of the
propensity to identify with someone else is to experience empathy, but such ‘feeling
with’ another may be just one element within a more complex relational stance, for
example that of concern; and an implication is that one is predisposed to experience
Anticipatory Concern, page 33
events in relation to a ‘virtual other’ (as Bråten, 1998, has expressed the matter) to whom
feelings are attributed, where an actual person may assume the place of this other created
in the mind. If this is the case, even such sympathetic responses as those reported in this
study need not amount to instances of ‘cognitive empathy’, for example defined as ‘a top-
down process whereby the subject effortfully tries to represent the state of the object; also
referred to as “putting oneself in the place of the other” or “imaginatively projecting
oneself into the situation of another”’ (Preston et al., 2007, p. 255, italics in original).
We do not deny that there is something like imagination involved in turning to
someone whose drawing is torn, and we do not deny that understandings or thoughts are
important for this. Yet we doubt whether the immediacy and patterning of one person’s
concerned reaction to another justifies the qualifier ‘cognitive’ for the empathic reaction,
as if thinking or effortful role-taking – or indeed, inferential reasoning based on ‘theory
of mind’ concepts - is essential to the emotional-cum-motivational relational force of
such events. Indeed, we suggest that in important respects, cognitive components of role-
taking are the developmentally elaborated distillate of cognitive aspects of an affectively
and motivationally configured process by which one is moved by and to the bodily
expressed attitudes of other persons (also Frijda, 1993). The process is early in onset, and
then evolves and endures throughout the lifespan. Such a view on the ‘primitiveness’ of
self-other relations appears to have been entertained by Darwin (1872, reprinted 1965, p
358), who wrote about his six-month-old’s expression of melancholy in response to his
nurse’s feigned distress, ‘Therefore it seems to me that an innate feeling must have told
him that the pretended crying of his nurse expressed grief; and this through the instinct of
sympathy excited grief in him’. If in this way, one person’s motivation to look to and
Anticipatory Concern, page 34
care about someone else arises out of affective engagement with feelings anticipated in or
attributed to the other, then no wonder participants with autism who are limited in these
respects rarely showed concern.
Although we consider this theoretical approach establishes a basis for the most
satisfactory account of the developmental psychopathology of autism, especially when
affected children’s early (preconceptual) abnormalities are taken into account, we
acknowledge that the results of the present study do not yield decisive evidence for or
against the theory. Given how participants were matched, one cannot exclude the
possibilities that either domain-general imaginative or generative capacities (e.g., Harris,
1989; Jarrold, Boucher, & Lewis, 1993; Minshew, Meyer, & Goldstein, 2002), or
conceptually mediated role-taking abilities (for example, built upon ‘theory of mind’
understanding), were responsible for the observed group differences. Indeed, if a
substantial part of the children’s limitations in generating thoughts and understanding
people develops in virtue of the propensity to identify with the attitudes of others, as one
of us has argued (Hobson, 1993a; 2002), then some of the conflicts among these
alternative theoretical positions might be resolved. Whether or not this proves possible,
the present investigation confirms that children with autism manifest limited anticipatory
concern towards the (expected) feelings of someone else. The findings are in keeping
with other evidence that the children have a relative lack of empathy and other-person-
centred feelings in relation to others, and arguably, that they may be limited in the
propensity to identify with other people’s attitudes. If so, then the presence of such
feelings among young typically developing children may testify to the operation of such
Anticipatory Concern, page 35
processes, initially shaping and subsequently shaped by the children’s concepts of other
selves as persons-with-minds, from early in children’s lives.
Anticipatory Concern, page 36
This project was supported financially by the Foundation for Autism Research and
Remediation (FARR), the Baily Thomas Charitable Fund, and the Tavistock and Portman
NHS Foundation Trust with NHS R&D Funding. Portions of the work were presented at
the 2007 International Meeting for Autism Research, and the 2007 Biennial Conference
of the Society for Research in Child Development. We gratefully acknowledge the
parents, staff, and students of Edith Borthwick School, Helen Allison School, Swiss
Cottage School, and Tolworth Infant School whose participation made this study
possible. We thank Kyratso Bargiota, Jacqueline Brown, Susana Caló, Jessica Clarke and
Emma Martin, for their generous assistance with ratings, videotape editing and testing.
Anticipatory Concern, page 37
American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental
Disorders: Revised Fourth Edition (DSM-IV). Washington, DC: APA.
Attwood, A., Frith, U., & Hermelin, B. (1988). The understanding and use of
interpersonal gestures by autistic and Down’s syndrome children. Journal of
Autism and Development Disorders, 18, 241 – 257.
Baron-Cohen, S., Richler, J., Bisarya, D., Gurunathan, N., & Wheelwright, S. (2003).
The systemizing quotient: an investigation of adults with Asperger syndrome or
high-functioning autism, and normal sex differences. Philosophical Transactions
of the Royal Society of London, 358, 361-374.
Barresi, J., & Moore, C. (1996). Intentional relations and social understanding.
Behavioral and Brain Sciences, 19, 107 – 154.
Barrett, K. C., Zahn-Waxler, C., & Cole, P. M. (1993). Avoiders vs. amenders: Implications
for the investigation of guilt and shame during toddlerhood? Cognition and Emotion,
7, 481 – 505.
Bauminger, N. (2004). The expression and understanding of jealousy in children with
Development and Psychopathology, 16, 157 – 177.
Bauminger, N., & Kasari, C. (2000). Loneliness and friendship in high-functioning
children with autism. Child Development, 71, 447 – 456.
Bosch, G. (1970). Infantile autism (translated by D. Jordan & I. Jordan). New York:
Anticipatory Concern, page 38
Bowler, D. M., Gardiner, J. M., & Grice, S. J. (2000). Episodic memory and
remembering in adults with Asperger syndrome. Journal of Autism and
Developmental Disorders, 30, 295 – 304.
Bråten, S. (1998). Infant learning by altercentric participation: The reverse of egocentric
observation in autism. In S. Bråten (Ed.) Intersubjective communication and
emotion in early ontogeny, pp. 105 – 126. University Press, Cambridge.
Buber, M. (1984, originally 1937). I and Thou (2nd ed.) (trans. R.G.Smith). Edinburgh:
Capps, L., Kehres, J., & Sigman, M. (1998). Conversational abilities among children with
autism and children with developmental delays. Autism, 2, 325 – 344.
Campbell, J. (2005). Joint attention and common knowledge. In N. Eilan, C. Hoerl, T.
McCormack, & J. Roessler (Eds.), Joint attention: Communication and other minds
(pp. 287-297). Oxford: Oxford University Press.
Charman, T., Swettenham, J., Baron-Cohen, S., Cox, A., Baird, G., & Drew, A. (1997).
Infants with autism: An investigation of empathy, pretend play, joint attention,
and imitation. Developmental Psychology, 33, 781 – 789.
Cole, P.M., Barrett, K.C., & Zahn-Waxler, C. (1992). Emotion displays in two-year-olds
during mishaps. Child Development, 63, 314 – 324.
Cooley, C.H. (1902). Human nature and the social order. New York: Scribner.
Dahlgren, S.O. & Gilberg, C. (1989). Symptoms in the first two years of life: A
preliminary population study of infantile autism. European Archives of Psychiatry
and Neurological Sciences, 238, 169 – 174.
Anticipatory Concern, page 39
Darwin, C. (1965, originally 1872). The expression of the emotions in man and animals.
Chicago: University of Chicago Press.
Dawson, G. Hill, D. Spencer, A. Galpert, L., & Watson, L. (1990). Affective exchanges
between young autistic children and their mothers. Journal of Abnormal Child
Psychology, 18, 335 –345.
Dunn, L. M., Dunn, L., & Whetton, C. (1982). British Picture Vocabulary Scale.
Windsor, U.K.: NFER-Nelson.
Eisenberg, N. (2002). Distinctions among various modes of empathy-related reactions: A
matter of importance in humans. Behavioral and Brain Sciences, 25, 33-34.
Eisenberg, N., & Strayer, J. (1987a). Empathy and its development. Cambridge:
Cambridge University Press.
Eisenberg, N., & Strayer, J. (1987b). Critical issues in the study of empathy. In N.
Eisenberg & J. Strayer (eds.), Empathy and its development, pp. 3-13. Cambridge:
Cambridge University Press.
Fairbairn, W.R.D. (1952). Object-relations and dynamic structure. In Fairbairn, W.R.D.,
Psychoanalytic studies of the personality, pp. 137-151. London: Routledge.
Feshbach, N., & Roe, K. (1968). Empathy in six and seven year olds. Child Development,
Freud, S. (1955/1921). Identification, in J. Strachey (Ed.), The standard edition of the
complete psychological works of Sigmund Freud (Vol. xviii, pp. 105 – 110).
Frijda, N.H. (1993). The place of appraisal in emotion. Cognition and Emotion, 7, 357-
Anticipatory Concern, page 40
Frith, U. (2003). Autism: Explaining the enigma (2nd ed). Oxford: Blackwell.
García-Pérez, R.M., Lee, A., & Hobson, R.P. (2007). On intersubjective engagement in
autism: A controlled study of nonverbal aspects of conversation. Journal of
Autism and Developmental Disorders, 37, 1310-1322.
García-Pérez, R.M., Hobson, R.P., & Lee, A. (2007). Narrative role-taking in autism.
Journal of Autism and Developmental Disorders (in press).
Harris, P. L. (1989). Children and emotion. Oxford: Blackwell.
Hobson, R.P. (1993a). Autism and the development of mind. Hove, Sussex: Erlbaum.
Hobson, R.P. (1993b). The emotional origins of social understanding. Philosophical
Psychology, 6, 227 – 249.
Hobson R.P. (2002). The cradle of thought. London: Macmillan (and 2004: New York,
Oxford University Press).
Hobson, R. P. (2005). Autism and emotion, In F. R. Volkmar, R. Paul, A. Klin, & D.
Cohen (Eds.) Handbook of autism and pervasive developmental disorders, 3rd
edition (pp. 406 – 424). New Jersey: John Wiley & Sons, Inc.
Hobson, R.P. (2007). Communicative depth: Soundings from developmental
psychopathology. Infant behavior and development, 30, 267-277.
Hobson, R.P., Chidambi, G., Lee, A., & Meyer, J. (2006). Foundations for self-
awareness: An exploration through autism. Monographs of the Society for
Research in Child Development,Serial no. 284, Vol 71.
Hobson, J.A., & Hobson, R.P. (2007). Identification: The missing link between imitation
and joint attention? Development and Psychopathology, 19, 411-431.
Anticipatory Concern, page 41
Hobson, R. P. & Lee A. (1998). "Hello and goodbye: A study of social engagement in
autism." Journal of Autism and Developmental Disorders, 28, 117 – 127.
Hobson, R.P., & Lee, A. (1999). Imitation and identification in autism. Journal of Child
Psychology and Psychiatry, 40, 649 – 659.
Hobson, R.P, Lee, A., & Hobson, J.A. (2007). Only connect? Communication,
identification, and autism. Social Neuroscience, 2, 320-335.
Hobson, R.P., & Meyer, J.A. (2005). Foundations for self and other: A study in autism.
Developmental Science, 8, 481 – 491.
Hoffman, M.L. (1982). Development of prosocial motivation: Empathy and guilt. In N.
Eisenberg (Ed.), Development of prosocial behavior, pp. 281 – 313. New York:
Hoffman, M.L. (2007). The origins of empathic morality in toddlerhood. In Brownell,
C.A, & Kopp, C.B. (eds.). Socioemotional development in the toddler years, pp
132-145. New York: Guilford.
Jarrold, C., Boucher, J., & Russell, J. (1977). Language profiles in children with autism:
Theoretical and methodologcal implications. Autism: The International Journal of
Research and Practice, 1, 57 – 76.
Jarrold, C., Boucher, J., & Smith, P. (1993). Symbolic play in autism: A review. Journal
of Autism and Developmental Disorders, 23, 281-307.
Kagan, J. (1982). The emergence of self. Journal of Child Psychology and Psychiatry, 23,
363 – 381.
Kanner, L. (1943). Autistic disturbances of affective contact. Nervous Child, 2, 217 – 250.
Anticipatory Concern, page 42
Kasari, C., Chamberlain, B., & Bauminger, N. (2001). Social emotions and social
relationships: Can children with autism compensate? In J.A. Burack, T. Charman,
N. Yirmiya, & P.R. Zelazo (Eds.), The development of autism (pp. 309 – 323).
Mahwah, NJ: Erlbaum.
Kasari, C., Sigman, M.D., Baumgartner, P. & Stipek, D.J. (1993). Pride and mastery in
children with autism. Journal of Child Psychology and Psychiatry, 34, 352 – 362.
Lazarus, R.S. (1984). Thoughts on the relations between emotion and cognition. In K.R.
Scherer & P.Ekman (eds.), Approaches to emotion, pp. 247-257. Hillsdale, N.J.:
Lee, A., & Hobson, R.P. (1998). On developing self-concepts: A controlled study of
children and adolescents with autism. Journal of Child Psychology and Psychiatry,
39, 1131 – 1141.
Lee, A., & Hobson, R.P. (2006). Drawing self and others: How do children with autism
differ from those with learning difficulties? British Journal of Developmental
Psychology, 24, 547-565.
Lewis, M. (2003). The development of self-consciousness. In J. Roessler and N. Eilan
(Eds.), Agency and self-awareness, pp. 275 – 295. Oxford: Clarendon.
Lewis, M. (2004). The emergence of human emotions. In M. L. Lewis and J. M. Haviland-
Jones (Eds.), Handbook of emotions: 2nd edition, pp. 265 – 292. New York:
Lockyer L. & Rutter, M. (1970). A five to fifteen year follow-up study of infantile
psychosis: IV Patterns of cognitive ability. British Journal of Social and Clinical
Psychiatry, 9, 152-163.
Anticipatory Concern, page 43
Lord, C. (1995). Follow-up of two-year-olds referred for possible autism. Journal of Child
Psychology and Psychiatry, 36, 1365 – 1382.
Lord, C., Storoschuk, S., Rutter, M., & Pickles, A. (1993). Using the ADI-R to diagnose
autism in preschool children. Infant Mental Health Journal, 14, 234 – 252.
Lord, C., Rutter, M., Dilavore, P., & Risi, S. (1999). Manual: Autism Diagnostic
Observation Schedule. Los Angeles, CA: Western Psychological Services.
MacDonald, H., Rutter, M., Howlin, P., Rios, P., LeCouteur, A., Evered, C. & Folstein, S.
(1989). Recognition and expression of emotional cues by autistic and normal adults.
Journal of Child Psychology and Psychiatry, 30, 865 – 877.
Meltzoff, A.N,. & Brooks, R. (2001). ‘Like me’ as a building block for understanding other
minds: Bodily acts, attention, and intention. In Malle, B.F. (ed.), Intentions and
intentionality: Foundations for social cognition, pp 171-192. Cambridge, Mass:
Merleau-Ponty, M. (1964). The child's relations with others (W. Cobb, trans.) In M.
Merleau-Ponty, The primacy of perception, pp 96 – 155. Evanston, IL:
Northwestern University Press.
Meyer, J. A. & Hobson, R. P. (2004). Orientation in relation to self and other: The case of
autism. Interaction Studies, 5, 221 – 244.
Meyer, J., & Hobson, P. (2006). Identification: The missing link between joint attention
and imitation? Manuscript submitted for publication.
Minshew, N. J., Meyer, J., & Goldstein, G. (2002). Abstract reasoning in autism: A
disassociation between concept formation and concept identification.
Neuropsychology, 16, 327 – 334.
Anticipatory Concern, page 44
Moore, C. (2007). Understanding self and others in the second year. In C. A. Brownell &
C. B. Kopp (Eds.), Transitions in early socioemotional development: The toddler
years, pp. 43-65. New York: Guilford.
Perner, J. (1990). Understanding the representational mind. Cambride, MA:
Perner, J., & Lang, B. (1999). Development of theory of mind and executive control.
Trends in Cognitive Sciences, 3, 337 – 344.
Preston, S.D, & de Waal, F.B.M. (2002). Empathy: Its ultimate and proximate bases.
Behavioral and Brain Sciences, 25, 1-20.
Preston, S.D, Bechara, A., Damasio, H., Grabowski, T.J., Stansfield, R.B., Mehta, S., &
Damasio, A.R. (2007). The neural substrates of cognitive empathy. Social
Neuroscience, 2, 254-275.
Ricks, D.M. (1975). Vocal communication in pre-verbal normal and autistic children. In
N. O'Connor (Ed), Language, cognitive deficits, and retardation, pp. 75 – 80.
Scheler, M. (1954). The nature of sympathy (translated by P. Heath). London: Routledge &
Schopler, E., Reichler, R.J., & Renner, B.R. (1988). The Childhood Autism Rating Scale
(CARS). Los Angeles, CA: Western Psychological.
Sigman, M.D., Kasari, C., Kwon, J.H., & Yirmiya, N. (1992). Responses to the negative
emotions of others by autistic, mentally retarded, and normal children. Child
Development, 63, 796 – 807.
Anticipatory Concern, page 45
Snow, M.E., Hertzig, M.E., & Shapiro,T. (1987). Expression of emotion in autistic
children. Journal of the American Academy of Child and Adolescent Psychiatry,
26, 836 – 838.
Sroufe, A. (1995). Emotional development. Cambridge: Cambridge University Press.
Stein, E. (1989). On the problem of empathy (trans. W. Stein). Washington: ICS
Stone, W.L., & Lemanek, K.L. (1990). Parental report of social behaviors in autistic
preschoolers. Journal of Autism and Developmental Disorders, 20, 513 – 522.
Strayer, J. (1987). Affective and cognitive perspectives on empathy. In N. Eisenberg & J.
Strayer (eds.), Empathy and its development, pp. 218-244. Cambridge: Cambridge
Thompson, R.A. (1987). Empathy and emotional understanding: the early development of
empathy. In N. Eisenberg & J. Strayer (eds.), Empathy and its development, pp. 119
- 145. Cambridge: Cambridge University Press.
Tomasello, M., Carpenter, M., Call, J., Behne, T., & Moll, H. (2005). Understanding and
sharing intentions: The origins of cultural cognition. Behavioral and Brain Sciences,
28, 675 – 735.
Trevarthen, C. (1979). Communication and cooperation in early infancy: a description of
primary subjectivity. In M. Bullova (Ed.), Before Speech: The Beginning of Human
Communication, pp. 321 – 347. London: Cambridge University Press.
Trevarthen, C. & Aitken, K. J. (2001). Infant intersubjectivity: research, theory and clinical
applications. Journal of Child Psychology & Psychiatry & Allied Disciplines , 1, 3
Anticipatory Concern, page 46
Vaish, A., & Tomasello, M. (2007). The cognitive route to sympathy and its relation to
prosocial behavior (submitted).
Vostanis, P., Smith, B., Corbett, J., Sungum-Paliwal, R., Edwards, A., Gingell, K.,
Golding, R., Moore, A., & Williams, J. (1998). Parental concerns of early
development in children with autism and related disorders. Autism, 2, 229 – 242.
Wimpory, D.C., Hobson, R.P., Williams, J.M., & Nash, S. (2000). Are infants with
autism socially engaged? A study of recent retrospective parental reports.
Journal of Autism and Developmental Disorders, 30, 525 – 536.
Wing, L. (1969). The handicaps of autistic children – A comparative study. Journal of
Child Psychology and Psychiatry, 10, 1 – 40.
Wispé, L. (1987). History of the concept of empathy. In N. Eisenberg & J. Strayer (eds.),
Empathy and its development, pp. 17-37. Cambridge: Cambridge University Press.
Woodruff Smith, D. (1989). The circle of acquaintance. Dordrecht: Kluwer Academic.
Yirmiya, N., Kasari, C., Sigman, M. & Mundy, P. (1989). Facial expressions of affect in
autistic, mentally retarded and normal children. Journal of Child Pschology and
Psychiatry, 30, 725 – 735.
Anticipatory Concern, page 47
Yirmiya, N., Sigman, M.D., Kasari, C. & Mundy, P. (1992). Empathy and cognition in
high-functioning children with autism. Child Development, 63, 150 – 160.
Zajonc, R.B. (1984). On primacy of affect. In K.R. Scherer & P.Ekman (eds.),
Approaches to emotion, pp. 259-270. Hillsdale, N.J.: Erlbaum.
Zahn-Waxler, C., & Robinson, J.A. (1995). Empathy and guilt: Early origins of feelings of
responsibility. In Tangney, J.P. & Fischer, K. W. (eds.), Self-conscious emotions,
pp. 143 – 173.
Zimmerman, I. L., Steiner, V. G., Evatt Pond, R., Boucher, J. & Lewis, V. (1997)
Preschool Language Scale (PLS-3) (UK Adaptation). The Psychological
Anticipatory Concern, page 48
Chronological Age Verbal Mental Age
M SD range M SD range
Autism 11;2 2;2 8;3 - 15;4 6;6 2;11 2;9 - 14;3
10;8 2;4 6;3 - 14;3 6;5 2;8 3;3 - 12;4
5;9 0;4 5;2 - 6;2 6;0 0;12 3;11 - 7;0
Note: Ages presented in years;months.
Anticipatory Concern, page 49
Degree of Empathy Examples Written by Judge Blind to Diagnosis – Score 0
Degree of Empathy: Score 0 – No Empathy
Participant is quite interested in his own drawing and reaches for it in the box. He is focused on
the task and smiles at E2 while saying “turtle,” when he is shown her drawing. His expression
changes slightly when her drawing is being torn. His eyes and mouth narrow and he stares at
the drawing that is being torn. After the drawing is torn up (and E1 says “okay”), he smiles
again slightly and nods. This suggests that he was unsure of how to react when the drawing
was torn and that he is guided by the fact that neither adult has reacted.
Participant stares at the drawings as they are shown to him, yet doesn’t respond. He reaches for
one drawing, but doesn’t follow through with his reach. He knows that E2 drew the turtle,
saying “E2” when asked who drew it, but then stares at his lap. When E1 secures his attention,
he continues to stare as the drawing is being torn, but doesn’t react to the event and maintains a
Participant stares intently and smiles at the drawings as they are shown to him and appears to
be genuinely interested. He continues to smile (seen throughout) as E1 tears E2’s drawing and
proclaims, “oh, oh, why rip it up?” He doesn’t look at E2 throughout the task and reaches for
his own drawing to indicate that he would like to keep it. He appears to be more concerned
about the drawing than about E2.
Anticipatory Concern, page 50 Download full-text
Degree of Empathy Examples Written by Judge Blind to Diagnosis – Score 1
Degree of Empathy: Score 1 – Limited Empathy
Participant is focused on the drawings with a slight smile on his face, looking occasionally at
E2. He gives E2 a big smile after being shown the turtle that she drew. He is then briefly
distracted and stares into space, but is looking at the drawing when E1 is tearing it up. His
expression remains neutral while E1 is tearing the drawing and he has a delayed reaction to the
act. He gasps, but responds with “I don’t know,” when E1 asks him what happened. He smiles
in response to the reassurance of the two testers (which indicates that E2 didn’t want her
drawing), but he doesn’t react to the act itself. He decides that he would also like to mimic the
act, ripping his own picture in four pieces.
Participant focuses intently on the pictures when they are shown to her, with a slight smile on
her face. When shown the drawing that E2 drew, she smiles and looks at E2. When the drawing
is ripped, she doesn’t really react to the act. She looks at E2, but her expression here is
comparable to what it was before. While she is studying what E1 is doing, she does not display
any outward expression of empathy.
Participant looks at the turtle that E2 drew when it is shown to him. His expression is pretty
much the same throughout the activity (he is focusing on the pictures yet al.so looks a bit
unsure about everything). When E2's drawing is torn, his eye movements suggest that he
registers this act, yet he still keeps the same facial expression. He smiles slightly and then
empathetically looks at E2 a few times after the drawing has been completely torn.