Content uploaded by Simon Baron-Cohen
Author content
All content in this area was uploaded by Simon Baron-Cohen
Content may be subject to copyright.
0162-3257/04/0400-0163/0 © 2004 Plenum Publishing Corporation
163
Journal of Autism and Developmental Disorders, Vol. 34, No. 2, April 2004 (© 2004)
The Empathy Quotient: An Investigation of Adults
with Asperger Syndrome or High Functioning Autism,
and Normal Sex Differences
Simon Baron-Cohen
1
and Sally Wheelwright
1
Empathy is an essential part of normal social functioning, yet there are precious few instruments
for measuring individual differences in this domain. In this article we review psychological the-
ories of empathy and its measurement. Previous instruments that purport to measure this have
not always focused purely on empathy. We report a new self-report questionnaire, the Empathy
Quotient (EQ), for use with adults of normal intelligence. It contains 40 empathy items and
20 filler/control items. On each empathy item a person can score 2, 1, or 0, so the EQ has a max-
imum score of 80 and a minimum of zero. In Study 1 we employed the EQ with n = 90 adults
(65 males, 25 females) with Asperger Syndrome (AS) or high-functioning autism (HFA), who
are reported clinically to have difficulties in empathy. The adults with AS/HFA scored signifi-
cantly lower on the EQ than n = 90 (65 males, 25 females) age-matched controls. Of the adults
with AS/HFA, 81% scored equal to or fewer than 30 points out of 80, compared with only 12%
of controls. In Study 2 we carried out a study of n = 197 adults from a general population, to
test for previously reported sex differences (female superiority) in empathy. This confirmed that
women scored significantly higher than men. The EQ reveals both a sex difference in empathy
in the general population and an empathy deficit in AS/HFA.
KEY WORDS: Empathy; sex differences; Asperger syndrome; social difficulties.
1
Autism Research Centre, Departments of Experimental Psychology
and Psychiatry, University of Cambridge, Douglas House, 18b
Trumpington Rd., CB2 2AH, United Kingdom.
Empathy is without question an important ability.
It allows us to tune into how someone else is feeling,
or what they might be thinking. Empathy allows us to
understand the intentions of others, predict their be-
havior, and experience an emotion triggered by their
emotion. In short, empathy allows us to interact effec-
tively in the social world. It is also the “glue” of the
social world, drawing us to help others and stopping us
from hurting others.
Whereas the term “sympathy” has a long tradition,
the term “empathy” astonishingly only came into being
at the turn of the last century. Astonishingly, because
we believe that this ability is as old as Homo sapiens
itself. No doubt empathy itself has this long an evolu-
tionary history, but the word “empathy” was invented
by Titchener as a translation of the German word
“Einfuhlung,” itself a term from aesthetics meaning “to
project yourself into what you observe” (Titchener,
1909).
Despite the obvious importance of empathy, it is
a difficult concept to define. Researchers in this area
have traditionally fallen into one of two camps: the-
orists who have viewed empathy in terms of affect,
and those who have taken a more cognitive approach.
We argue that both approaches are essential to defin-
ing empathy, and that in most instances, the cognitive
and affective cannot be easily separated. Nevertheless,
for historical reasons, we begin by examining each in
turn.
483924.qxd 3/18/04 21:33 Page 163
164 Baron-Cohen and Wheelwright
THE AFFECTIVE APPROACH
The affective approach defines empathy as an ob-
server’s emotional response to the affective state of
another. This view of empathy arose from writings on
sympathy. Within the affective approach, different de-
finitions of empathy vary in how broad or narrow the
observer’s emotional response to another’s emotion has
to be.
As far as we can see, there are four varieties of
empathy: 1) the feeling in the observer must match that
of the person observed (e.g., you feel fright when you
see someone else’s fear; Eisenberg & Miller, 1987;
Hoffman, 1984); 2) The feeling in the observer is sim-
ply appropriate to the other person’s emotional state in
some other way, even though it doesn’t exactly match
it (e.g., you may feel pity at someone else’s sadness;
Stotland, 1969); 3) The feeling in the observer may be
any emotional response to another’s emotion (e.g., an
observer feeling pleasure at another’s pain). This is re-
ferred to as “contrast empathy” (Stotland, Sherman, &
Shaver, 1971). 4) The feeling in the observer must be
one of concern or compassion to another’s distress
(Batson, 1991).
Options 1, 2, and 4 all seem important to include,
but need not be mutually exclusive. One can include
all of these in a useful definition of empathy. However,
option 3 seems questionable. We argue that empathy
should exclude inappropriate emotions triggered by
someone else’s emotional state (e.g., feeling pleasure
at another’s pain). Rather, the affective definition of
empathy emphasises the appropriateness of the viewer’s
emotional response. Of course, defining what is an ap-
propriate emotional response is not straightforward. For
example, hearing of the death of a young friend who
had been suffering from a painful, terminal illness
might produce in you both relief (that their pain is over)
and sadness (that their life has been cut short). Both
emotions are appropriate emotional responses and can
therefore be classified as empathic. (Note that if you
feel sadness at the loss of this friend, this may be un-
related to empathy, as it may be purely self-centred,
albeit still appropriate. To count as empathy, your
emotion needs to be a consequence of their emotion.)
THE COGNITIVE APPROACH
Cognitive theories emphasize that empathy in-
volves understanding the other’s feelings (Kohler,
1929). These theories also refer to cognitive processes
such as role-taking, switching attention to take another’s
perspective (Mead, 1934), or “decentering”; that is, re-
sponding nonegocentrically (Piaget, 1932). During the
1940s and 1950s the term “social acuity” was also used
to refer to empathy (Chapin, 1942; Dymond, 1950; Kerr
& Speroff, 1954). In recent terminology, the cognitive
component is referred to as using a “theory of mind”
(Astington, Harris, & Olson, 1988; Wellman, 1990) or
“mindreading” (Baron-Cohen, 1995; Whiten, 1991). Es-
sentially, this involves setting aside one’s own current
perspective, attributing a mental state (or “attitude”) to
the other person (Leslie, 1987), and then inferring the
likely content of their mental state, given the experience
of that person.
2
In some accounts these processes ap-
pear purely cognitive in that there is no reference to any
affective state. For example, a person might infer that
because John was absent during a key event, he will not
know about it. In addition to this comprehension and
inferential process, the cognitive element also entails
the ability to predict another’s behavior or mental state
(Dennett, 1987). Thus, taking into account John’s
ignorance about a plan being changed can lead to the
prediction that he will go to the wrong place.
It is clear from the above discussion that empa-
thy consists of both the affective and cognitive com-
ponents (Davis, 1994). A pictorial representation of
the two-component model of empathy is presented in
Figure 1.
EMPATHY AND SYMPATHY: WHAT IS
THE RELATIONSHIP?
In moral philosophy, Adam Smith described sym-
pathy as the experience of “fellow-feeling” we have
when we observe someone else’s powerful emotional
state (Smith, 1759). Sympathy is therefore a clear in-
stance of the affective component of empathy. Sympa-
thy is said to occur when the observer’s emotional
response to the distress of another leads the observer
to feel a desire to take action to alleviate the other
person’s suffering (Davis, 1994). The observer may not
actually act on this desire, but at the very least the
observer has the emotion of wanting to take appropriate
action to reduce the other’s distress. Thus, in Figure 1,
sympathy is shown as a special subset of empathy. (We
assume sympathy can entail both the cognitive and
affective elements of empathy.)
2
A useful analogy here is to think of computers, where the user can
switch from the current “window” (their representation of the
world) to another window (someone else’s different representation
of the world). Taking this analogy further, some individuals will be
more empathic than others because they switch between windows
more frequently, or more easily, or because the content of the other
window (their representation of the other’s mind) is more detailed.
483924.qxd 3/18/04 21:33 Page 164
The Empathy Quotient: An Investigation of Adults with Asperger Syndrome 165
Fig. 1. A simple model showing the two overlapping components of empathy and how
sympathy is a special case of the affective component of empathy.
+ Feeling an appropriate emotion triggered by seeing/learning of another’s emotion.
# Understanding and/or predicting what someone else might think, feel, or do.
* Feeling an emotion triggered by seeing/learning of someone else’s distress which moves
you to want to alleviate their suffering.
To give an example, if you walk past a homeless
person in winter and you are “moved” or “touched”
(both interesting metaphors) to want to help them, this
would count as sympathy. You may do nothing more.
For example, you may feel that your action would be
futile given the many other homeless people in the same
neighborhood and the near impossibility of helping them
all. So you might walk past and do nothing. Your reac-
tion would still count as sympathy because you felt the
desire to alleviate another’s suffering. This same term
would also apply even if you did indeed take action, and
gave the homeless person your gloves. If, however, you
experienced an appropriate emotion (e.g., pity) to the
homeless person’s emotion (e.g., hopelessness), but you
did not experience any desire to take action to alleviate
his or her suffering, then this would count as empathy,
but not sympathy. As a final note of clarification, if you
felt an inappropriate emotion to the homeless person’s
emotional state (e.g., feeling glad that you had a warm
home with a well-stocked refrigerator), this would count
as neither empathy nor sympathy.
MEASURING EMPATHY IN ADULTS
There are several instruments that purport to mea-
sure empathy, but as we will argue, many of these may
not do so. For example, in the Chapin Social Insight
Test (Chapin, 1942), subjects are presented with
hypothetical scenarios (e.g., being disturbed by noisy
neighbors) and have to choose the most effective course
of action from four options. This is held to be a mea-
sure of empathy, though clearly it involves more than
this because choosing an effective course of action
might be based on knowledge of social rules, cultural
convention, and so forth.
A second measure of empathy used a rating scale
(Dymond, 1949, 1950). A group of subjects was left to
interact with each other. Each then estimated how each
of the others in the group rated the subject. This was
intended to measure how accurately one can predict
another’s view of oneself. However, it has been pointed
out that it is possible to achieve high levels of accu-
racy on this test without it reflecting empathy (Davis,
1994). For example, if all individuals tended to use the
midpoints on the scale, this would lead to apparent
accuracy (Cronbach, 1955).
A more widely used test is the Empathy (EM)
Scale (Hogan, 1969). The EM has 64 items and has
been found to have four relatively uncorrelated factors:
social self-confidence, even-temperedness, sensitivity,
and nonconformity (Johnson, Cheek, & Smither, 1983).
As can be seen from these factors alone, it is clear that
this scale is also not a pure measure of empathy. In fact,
only one of these factors is directly relevant to empa-
thy (sensitivity). The scale may be better thought of as
a measure of social skill (Davis, 1994).
483924.qxd 3/18/04 21:33 Page 165
166 Baron-Cohen and Wheelwright
The Questionnaire Measure of Emotional Empa-
thy (QMEE) was designed with the explicit aim of
assessing an individual’s tendency to react strongly to
another’s experience (Mehrabian & Epstein, 1972). It
contains seven subscales. The authors suggest that, as
the split-half reliability is high (0.84), the items are
likely to tap a single construct. The authors also sug-
gest that this single construct may be emotional arous-
ability to the environment in general, rather than to
people’s emotions in particular (Mehrabian, Young, &
Sato, 1988). Thus, although some items in the QMEE
may measure affective empathy, the scale as a whole
may be confounded.
A final questionnaire, the Interpersonal Reactivity
Index (IRI) (Davis, 1980) has four seven-item subscales,
tapping “perspective-taking,” “empathic concern,”
“personal distress,” and “fantasy.” In our view, the IRI
is the best measure of empathy developed to date, be-
cause three of the four factors are directly relevant to
empathy. However, we suspect the IRI may measure
processes broader than empathy, as included in the fan-
tasy subscale are items such as “I daydream and fanta-
size, with some regularity, about things that might
happen to me,” and included in the personal distress
subscale are items such as “In emergency situations, I
feel apprehensive and ill at ease.” These subscales may
assess imagination or emotional self-control, and
although these factors may be correlated with empathy,
it is clear that they are not empathy itself.
The above review highlights the need for a new
measure of empathy. We report one here, which we call
the Empathy Quotient (EQ).
3
THE EQ
The EQ was designed to be short, easy to use, and
easy to score. It is shown in the Appendix. The EQ com-
prises 60 questions, broken down into two types:
40 questions tapping empathy (items 1, 4, 6, 8, 10, 11,
12, 14, 15, 18, 19, 21, 22, 25, 26, 27, 28, 29, 32, 34,
35, 36, 37, 38, 39, 41, 42, 43, 44, 46, 48, 49, 50, 52,
54, 55, 57, 58, 59, and 60), and 20 filler items (items
2, 3, 5, 7, 9, 13, 16, 17, 20, 23, 24, 30, 31, 33, 40, 45,
47, 51, 53, and 56).
The 20 filler items were included to distract the
participant from a relentless focus on empathy. An ini-
tial attempt to separate items into purely affective and
cognitive categories was abandoned because in most
instances of empathy, the affective and cognitive com-
ponents co-occur and cannot be easily disentangled.
Each of the items listed above scores 1 point if the
respondent records the empathic behavior mildly, or
2 points if the respondent records the behavior strongly
(see below for scoring of each item). Approximately
half the items were worded to produce a “disagree”
response and half to produce an “agree” response for
the empathic response. This was to avoid a response
bias either way. Following this, items were random-
ized. The EQ has a forced choice format, can be self-
administered, and is straightforward to score because
it does not depend on any interpretation.
AIMS
In the studies reported below, we had four aims:
to test whether adults with high-functioning autism
(HFA) or Asperger Syndrome (AS) score lower on the
EQ (study 1); to test whether the EQ is inversely
correlated with the AQ (Autism Spectrum Quotient)
(Baron-Cohen, Wheelwright, Skinner, Martin, &
Clubley, 2001b), as would be predicted if autism/AS is
an empathy disorder (study 1)—the AQ is a 50-point
self-report scale for use by adults with HFA or AS;
to test whether the EQ inversely correlates with the
FQ (Friendship Questionnaire; Baron-Cohen &
Wheelwright, 2003) as an index of the validity of the
EQ (study 1)—The FQ is a 25-point self-report scale
assessing reciprocity and intimacy in relationships, with
a scale range of 0–135; and to test for sex differences
in empathy, given earlier reports of a female superior-
ity (Davis, 1980; Davis & Franzoi, 1991; Hall, 1978;
Hoffman, 1977; study 2).
AUTISM
Autism is diagnosed when an individual shows ab-
normalities in social and communication development
in the presence of marked repetitive behavior and lim-
ited imagination (American Psychiatric Association,
1994). The term HFA is given when an individual meets
the criteria for autism in the presence of normal IQ. AS
is defined in terms of the individual meeting the same
criteria for autism but with no history of cognitive or
language delay (World Health Organization, 1994).
Language delay itself is defined as not using single
words by 2 years of age or not using phrase speech by
3 years of age.
3
The term “quotient” is used not in the statistical sense (the result
of dividing one quantity by another) but used on the basis of the
Latin root “quotiens” (meaning “how much” or “how many”).
483924.qxd 3/18/04 21:33 Page 166
The Empathy Quotient: An Investigation of Adults with Asperger Syndrome 167
There is growing evidence that autism and AS are
of genetic origin. The evidence is strongest for autism
and comes from twin and behavioral genetic family
studies (Bailey et al., 1995; Bolton & Rutter, 1990;
Folstein & Rutter, 1977, 1988). Family pedigrees of AS
also implicate heritability (Gillberg, 1991). There is
also an assumption that autism and AS lie on a
continuum (Baron-Cohen, 1995; Frith, 1991; Wing,
1981, 1988). One reason for testing adults with HFA
or AS on the EQ (study 1, below) was to explore the
notion that autism is an empathy disorder, given their
mindreading deficits (Baron-Cohen, 1995; Gillberg,
1992; Yirmiya, Sigman, Kasari, & Mundy, 1992).
STUDY 1
Piloting
Pilot testing of the EQ was conducted before the
cross-validation study, using a separate sample of con-
trols (n = 20). This was to check that questions were
comprehensible, that the instrument was producing a
good spread of scores, and that both floor and ceiling
effects were not evident. Data from this pilot study were
not analyzed statistically, because of the small sample
size, but all of these instrument properties were con-
firmed.
Subjects
Two groups of subjects were tested. Group 1 com-
prised 90 adults with AS/HFA (65 males, 25 females).
This sex ratio of 2.6:1 (m:f) is similar to that found in
other samples (Klin, Volkmar, Sparrow, Cicchetti, &
Rourke, 1995). All subjects in this group had been di-
agnosed by psychiatrists using established criteria for
autism or AS (American Psychiatric Association, 1994).
They were recruited via several sources, including the
National Autistic Society (UK), specialist clinics car-
rying out diagnostic assessments, and advertisements
in newsletters and Webpages for adults with AS/HFA.
Their mean age was 34.2 years (SD = 12.5, range
15.4–59.9). They had all attended mainstream school-
ing and were reported to have an IQ in the normal
range. (See below for a confirmation of this.) Their
occupations reflected their mixed socioeconomic sta-
tus (SES). Because we could not confirm age of onset
of language with any precision (because of the consid-
erable passage of time), these individuals are grouped
together, rather than attempting to separate them into
AS versus HFA.
Group 2 comprised 90 adults selected from a pool
of 197 volunteers on the basis of being age- and sex-
matched with group 1. The 197 volunteers are described
in study 2. The 90 comparison subjects, similar to
group 1, consisted of 65 males and 25 females. Their
mean age was 34.2 years (SD = 11.8, range 17.4–56.4).
Their SES profile was similar to that of group 1.
Method
Subjects were sent the EQ by post and were in-
structed to complete it on their own, as quickly as pos-
sible, and to avoid thinking about responses too long.
Subjects in group 2 had the option to remain anony-
mous. To confirm the diagnosis of adults in group 1
being high-functioning, 15 subjects in each group were
randomly selected and invited into the lab for intellec-
tual assessment using four subtests of the WAIS-R
(Wechsler, 1958). The four subtests of the WAIS-R
were Vocabulary, Similarities, Block Design, and Pic-
ture Completion. On this basis, all of the subjects had
a prorated IQ of at least 85, that is, in the normal range
(group 1, x = 106.5, SD = 8.0; group 2, x = 105.8,
SD = 6.3), and they did not differ from each other sta-
tistically (t-test, p > .05). These subjects were asked to
fill in the EQ for a second time 12 months later as a
test of its retest reliability.
Subjects in group 1 were also sent the AQ (Baron-
Cohen et al., 2001), and 45 of them were also sent the
FQ (Baron-Cohen & Wheelwright, 2003) by post. Their
mean AQ score was 35.7 (SD = 6.7). This is in the clin-
ical range on this measure, as our previous study using
the AQ shows that more than 80% of people with a
diagnosis of AS or HFA score equal to or above 32
(maximum = 50). Their mean FQ score was 54.8 (SD =
19.8), which is significantly lower than the controls
(x = 83.0, SD = 18.5, t =−7.9, p < .0001).
Scoring the EQ
“Definitely agree” responses scored 2 points and
“slightly agree” responses scored 1 point on the fol-
lowing items: 1, 6, 19, 22, 25, 26, 35, 36, 37, 38, 41,
42, 43, 44, 52, 54, 55, 57, 58, 59, 60.
“Definitely disagree” responses scored 2 points
and “slightly disagree” responses scored 1 point on the
following items: 4, 8, 10, 11, 12, 14, 15, 18, 21, 27, 28,
29, 32, 34, 39, 46, 48, 49, 50.
For filler items, the total number of each possible
response was computed to check for systematic bias
in responding by each group. We predicted that the
HFA/AS group should not differ on how they responded
483924.qxd 3/18/04 21:33 Page 167
168 Baron-Cohen and Wheelwright
on these control items, compared with individuals in
the general population.
Definition of Empathy and Corroboration
with Other Experts
To go beyond the authors’ subjective assessment
of whether the chosen items in the EQ were good tests
of empathy, we provided a definition of empathy (see
below) to a panel of six judges (all experimental psy-
chologists working in this field) and asked them to rate
on a 2-point scale (yes or no) whether each of the key
items in the EQ related to the overarching definition of
empathy. The definition given was as follows: “Empa-
thy is the drive or ability to attribute mental states to
another person/animal, and entails an appropriate af-
fective response in the observer to the other person’s
mental state.” Results showed that all 40 empathy items
were rated as being related to empathy and all 20 filler
items were correctly identified as being unrelated to
empathy by at least five out of six judges. The proba-
bility of obtaining such agreement on each item by
chance is p < .003.
Results
Mean EQ scores are shown in Table I, and Fig-
ure 2 displays the distribution of EQ scores in the two
groups. A t-test was used to compare groups 1 and 2 on
total EQ score. As predicted, the AS/HFA group scored
significantly lower than the controls (t =−13.07, df =
178, p < .0001). As predicted, in group 1, EQ scores
were inversely correlated with the AQ (r =−0.56,
p < .0001) and directly correlated with the FQ (r =
0.59, p < .001).
When the percentage of subjects in each group
scoring at or above each EQ score was calculated out,
this revealed that a useful cut-off with which to sepa-
rate the groups is equal to or fewer than 30 points. Of
adults with AS/HFA, 81.1% score at or below this cut-
off, versus only 12.2% of the comparison group. This
cut-off is the most useful one because it generates the
largest difference between the two groups. An analysis
of the percentage of subjects attaining each possible
score on each empathy item within the EQ showed that
Table I. Means and SDs of Total Empathy Quotient Score
in Study 1
Group Total (max = 80)
AS/HFA (n = 90)
Mean 20.4
SD 11.6
Controls (n = 90)
Mean 42.1
SD 10.6
Note: AS, Asperger syndrome; HFA, high functioning autism.
Fig. 2. Empathy Quotient scores in Asperger syndrome/high-functioning autism group and controls.
483924.qxd 3/18/04 21:33 Page 168
The Empathy Quotient: An Investigation of Adults with Asperger Syndrome 169
the AS/HFA group scored more frequently than the
comparison group on only two items, numbers 11 and
39. An analysis was carried out of the mean percent-
age of subjects in each group giving each of the four
possible responses for the filler items. This showed that
the AS/HFA group and the comparison groups did not
differ in the use of the response scale by the two groups.
Finally test-retest reliability for the EQ was r = 0.97,
which is also highly significant (p < .001).
Discussion
The results of study 1 show that as a group, the pa-
tients with HFA/AS scored significantly lower on the EQ
than controls matched for age and gender who were
drawn from a general population. This provides some
support for the view of HFA/AS as an empathy disorder
(Gillberg, 1992). Many of the EQ items tap what could
also be described as a need for a “theory of mind,” which
previous studies have found to be impaired in autism
(Baron-Cohen, 1995; Baron-Cohen, Leslie, & Frith,
1985; Perner, Frith, Leslie, & Leekam, 1989). The EQ
thus confirms an empathizing deficit in HFA/AS, as mea-
sured by self-report and using items referring to every-
day understanding of minds (Baron-Cohen, 2002).
Our clinical interviews with a series of adults with
AS (n = 50) provided an opportunity to probe the rea-
sons for their lower score on the EQ. They reported that
even though they have difficulty judging/explaining/
anticipating or interpreting another’s behavior, it is not
the case that they want to hurt another person. When it
is pointed out to them that their behavior was hurtful
(e.g., because they failed to pick up when someone
around them needed comforting, or because they had
said something that caused offense), they typically feel
bad about the hurt they caused. Often they cannot rec-
ognize that the cause of the other person’s hurt was their
responsibility, or they cannot see how they could have
acted differently to avoid such hurt, but nevertheless
wish that such hurt could have been avoided. From this
we can conclude that people with AS/HFA are not like
unfeeling psychopaths. Rather, psychopaths might be
expected to show the opposite profile—being able to
judge and predict how another person might feel, even
if they have little concern about that person’s emotion
(Blair, 1995; Blair, Jones, Clark, & Smith, 1997). We
have not tested psychopaths directly on the EQ, as a self-
report scale might not be the best way to assess indi-
viduals who have an acknowledged tendency to deceive.
A concern might be raised as to whether condi-
tions like AS or HFA might have impaired the subject’s
ability to judge their own social or communicative
behaviour, because of subtle mind-reading problems
that are found even in adult patients (Baron-Cohen,
Jolliffe, Mortimore, & Robertson, 1997; Baron-Cohen,
Wheelwright, & Jolliffe, 1997; Happe, 1994). However,
if this impairment had occurred, it would have led the
person to score higher on the EQ, rating their own be-
havior as more empathic than it might really be. There-
fore, any inaccuracies of this kind would cause an
elevated estimate of the person’s true EQ score. This
concern is thus not relevant, as if anything, any such
overrating would have reduced the likelihood of find-
ing a group difference.
The finding that the EQ is inversely correlated with
the AQ is one indicator of the validity of the EQ, in that
two of the domains of the AQ measure social sensitivity
and sensitive communication, both of which require em-
pathy. The fact that more than 80% of adults who have
a diagnosis of AS or HFA score above 32 on the AQ and
below 30 on the EQ may indicate their potential for use
as screening instruments within clinic settings. We would
not recommend these instruments for use in screening
the general population, as the rate of individuals in the
general population scoring in these ranges is at least 2%.
(What proportion of these high scorers are true or false
positives is not yet known.) Finally, the strong positive
correlation between the EQ and the FQ provides further
validation that the EQ is measuring empathy, as the FQ
assesses empathy in the context of close relationships.
In study 2, reported next, we tested the last of our aims;
namely, testing whether there is a sex difference on the
EQ in the general population.
STUDY 2
Subjects
Group 1 comprised 71 males with a mean age of
38.8 years (SD = 13.7, range 17.4–69.6). Group 2 com-
prised 126 females with a mean age of 39.5 (SD = 12.8,
range 17.7–73.0). Both groups were recruited from two
main sources: volunteers from staff at two large su-
permarkets in Cambridge (Sainsbury’s and Marks and
Spencer’s) and volunteers from a village in Glouces-
tershire. These people were recruited in an effort to test
a sample that was mixed in terms of social class. There
were no differences in the SES of groups 1 and 2. The
method was the same as that used in Study 1.
Results
Mean EQ scores are shown in Table II. A t-test was
used to compare groups 1 and 2 on total EQ score.
483924.qxd 3/18/04 21:33 Page 169
170 Baron-Cohen and Wheelwright
Table II. Means and SDs of Total Empathy Quotient
Score in Study 2
Group Total (max = 80)
Males (n = 71)
Mean 41.8
SD 11.2
Females (n = 126)
Mean 47.2
SD 10.2
As predicted, males scored significantly lower than
females (t = 3.4, df = 196, p < .0001). Using the cut-
off established from study 1, only 4% of females scored
equal to or fewer than 30 points, versus 14% of males.
Twice as many women as men scored equal to or more
than 54 points, and more than three times as many
women as men scored equal to or more than 62 points.
An item analysis showed that males scored more fre-
quently than females on only three items, numbers 4,
14, and 41. An analysis was carried out of the mean per-
centage of subjects giving each of the four possible
responses for the filler items. Males and females did
not differ in the use of the response scale. Finally,
Cronbach’s alpha was calculated for the EQ as a whole,
using all subjects from studies 1 and 2. Alpha was 0.92,
which is high.
Discussion
As predicted, study 2 showed that women score
slightly but significantly higher on the EQ than men.
This replicates a series of earlier studies reporting sex
differences (female superiority) on questionnaire mea-
sures of empathy (Davis, 1980; Davis & Franzoi, 1991;
Hall, 1978; Hoffman, 1977). Specifically, more than
three times as many men (14%) as women (4%) scored
in the “AS/HFA range” (i.e., equal to or fewer than
30 points), whereas more than three times as many
women (9.5%) as men (2.8%) scored in the “super-
empathic range” (i.e., equal to or more than 62 points).
Whether this reflects women’s greater willingness to
report empathic behavior or their higher levels of un-
derlying empathy cannot be determined from this study.
As in study 1, there do not appear to be any response
biases in one sex (or in one clinical group), as mea-
sured by response patterns across the control items.
SUMMARY
In this article we have reviewed the definition of
empathy at a psychological level and described a new
self-assessment instrument, the EQ, for measuring
empathy in adults of normal intelligence. As predicted,
adults with AS/HFA scored significantly lower on the
EQ than matched controls (study 1). Again, as predicted,
the EQ was inversely correlated with the AQ (Baron-
Cohen et al., 2001), and positively correlated with the
FQ (Baron-Cohen & Wheelwright, 2003). Finally, in
the normal control group, as predicted, women scored
slightly but significantly higher than men (study 2).
The results from both studies are consistent with
the extreme male brain (EMB) theory of autism
(Asperger, 1944; Baron-Cohen, 2002, 2003; Baron-
Cohen & Hammer, 1997). The EMB theory recognises
two psychological dimensions: “empathizing” (E) and
“systemizing” (S). Empathizing is the drive to identify
another’s mental state and to respond with an appro-
priate emotion to this. Systemizing is the drive to
analyze a system in terms of its underlying lawful reg-
ularities and to construct systems using such lawful
regularities. The male brain is defined as individuals in
whom S > E, and the female brain is defined by the
converse psychometric profile (E > S). The EMB the-
ory predicts that individuals on the autistic spectrum
will show an exaggerated male profile (S E). The
results of the EQ study above are consistent with this
theory, as are a series of other studies (Baron-Cohen
et al., 1997; Baron-Cohen, O’Riordan, Jones, Stone, &
Plaisted, 1999; Baron-Cohen, Wheelwright, Scahill,
Lawson, & Spong, 2001a). This theory may have
implications for the marked sex ratio in AS (8m:1f)
(Wing, 1981).
The approach to studying empathy using self-
report methods has a number of inevitable limitations.
First, empathy may comprise both state and trait
components. Some individuals will be higher in em-
pathy than others for trait reasons, which could reflect
both genetic or early experiential factors (Fonagy,
Steele, Steele, & Holder, 1997). We expect that em-
pathy traits are being assessed by the EQ, albeit in
terms of the individual’s belief about their own em-
pathic traits. However, empathy can also vary as a
function of a person’s current state. Thus, if you are
drunk, you might continue to drive your point home in
a discussion for far longer than is sensitive to your lis-
tener, and in this sense act unempathically. If you are
angry or depressed, your own current emotional state
might cloud your ability to see the other person’s per-
spective in an argument, so that you are temporarily
only able to see your own. That is, your ability to
switch perspectives may be reduced by your current
state. It is unlikely that self-report instruments assess-
ing empathy are sensitive to such changes in state. Sec-
ond, we recognize that the EQ only assesses the
483924.qxd 3/18/04 21:33 Page 170
The Empathy Quotient: An Investigation of Adults with Asperger Syndrome 171
individual’s beliefs about their own empathy, or how
they might like to be seen or think about themselves,
and that this may be different to how empathic they
are in reality. Future work might usefully compare an
individual’s own self-assessed EQ score with that
based on the ratings by a partner or parent of that same
individual.
Despite these limitations, the EQ appears to be
picking up considerable individual, sex, and group
differences, in both a general population sample and
a clinical sample. Thinking about autism spectrum
conditions as empathy disorders may be a useful
framework and may teach us something about the
neuro-developmental and genetic basis of empathy.
Future work using the EQ needs to include psychi-
atric samples other than autism so that we can learn
about the sensitivity and specificity of this instrument.
It is clear that autism is not the only psychiatric con-
dition in which empathy is compromised, and for this
reason the EQ is unlikely to be useful as a diagnos-
tic. For this reason, too, even in clinical screening,
we advise that the EQ be accompanied by other in-
struments, such as the AQ, high scores on which may
be specific to autism. Such specificity tests need to
be carried out in the future. Nevertheless, the EQ can
be said to have reasonable construct and external va-
lidity in having a high alpha coefficient and in being
correlated with independent measures. Future work
needs to further test the validity of the EQ, perhaps
using “live” measures of empathy. Current evidence
for the convergent and divergent validity of the EQ
is limited to the AQ and FQ. More recently, an in-
verse correlation between the Systemizing Quotient
(SQ) and the EQ (Baron-Cohen, Richler, Bisarya,
Gurunathan, & Wheelwright, 2003) has also been
found, and the sex difference on the EQ has been
replicated. An important next step will be to test the
validity of the EQ against existing empathy ques-
tionnaires such as the IRI (Davis, 1994). Even on the
strength of the present data, however, we have
suggestive evidence for autism spectrum conditions
entailing an impairment in empathy.
ACKNOWLEDGMENTS
Simon Baron-Cohen and Sally Wheelwright were
supported by the Medical Research Council (UK) and
the James S McDonnell Foundation during the devel-
opment of this work. We are grateful to the Three
Guineas Trust for the funding of the clinical aspects of
this work and to Peter Lipton, Peter Carruthers, and
Gabriel Segal for discussion of the concepts in the
article.
APPENDIX: THE EQ
How to Fill Out the Questionnaire
Below is a list of statements. Please read each statement carefully and rate how strongly you agree or disagree
with it by circling your answer. There are no right or wrong answers, or trick questions.
IN ORDER FOR THE SCALE TO BE VALID, YOU MUST ANSWER EVERY QUESTION.
Examples
E1. I would be very upset if I couldn’t listen to music strongly slightly slightly strongly
every day. agree agree disagree disagree
E2. I prefer to speak to my friends on the phone rather strongly slightly slightly strongly
than write letters to them. agree agree disagree disagree
E3. I have no desire to travel to different parts of the strongly slightly slightly strongly
world. agree agree disagree disagree
E4. I prefer to read than to dance. strongly slightly slightly strongly
agree agree disagree disagree
1. I can easily tell if someone else wants to enter a strongly slightly slightly strongly
conversation. agree agree disagree disagree
2. I prefer animals to humans. strongly slightly slightly strongly
agree agree disagree disagree
3. I try to keep up with the current trends and strongly slightly slightly strongly
fashions. agree agree disagree disagree
(continued)
483924.qxd 3/18/04 21:33 Page 171
172 Baron-Cohen and Wheelwright
4. I find it difficult to explain to others things that I strongly slightly slightly strongly
understand easily, when they don’t understand it agree agree disagree disagree
first time.
5. I dream most nights. strongly slightly slightly strongly
agree agree disagree disagree
6. I really enjoy caring for other people. strongly slightly slightly strongly
agree agree disagree disagree
7. I try to solve my own problems rather than strongly slightly slightly strongly
discussing them with others. agree agree disagree disagree
8. I find it hard to know what to do in a social strongly slightly slightly strongly
situation. agree agree disagree disagree
9. I am at my best first thing in the morning. strongly slightly slightly strongly
agree agree disagree disagree
10. People often tell me that I went too far in driving strongly slightly slightly strongly
my point home in a discussion. agree agree disagree disagree
11. It doesn’t bother me too much if I am late meeting strongly slightly slightly strongly
a friend. agree agree disagree disagree
12. Friendships and relationships are just too difficult, strongly slightly slightly strongly
so I tend not to bother with them. agree agree disagree disagree
13. I would never break a law, no matter how minor. strongly slightly slightly strongly
agree agree disagree disagree
14. I often find it difficult to judge if something is strongly slightly slightly strongly
rude or polite. agree agree disagree disagree
15. In a conversation, I tend to focus on my own strongly slightly slightly strongly
thoughts rather than on what my listener might be agree agree disagree disagree
thinking.
16. I prefer practical jokes to verbal humor. strongly slightly slightly strongly
agree agree disagree disagree
17. I live life for today rather than the future. strongly slightly slightly strongly
agree agree disagree disagree
18. When I was a child, I enjoyed cutting up worms to strongly slightly slightly strongly
see what would happen. agree agree disagree disagree
19. I can pick up quickly if someone says one thing strongly slightly slightly strongly
but means another. agree agree disagree disagree
20. I tend to have very strong opinions about morality. strongly slightly slightly strongly
agree agree disagree disagree
21. It is hard for me to see why some things upset strongly slightly slightly strongly
people so much. agree agree disagree disagree
22. I find it easy to put myself in somebody else’s strongly slightly slightly strongly
shoes. agree agree disagree disagree
23. I think that good manners are the most important strongly slightly slightly strongly
thing a parent can teach their child. agree agree disagree disagree
24. I like to do things on the spur of the moment. strongly slightly slightly strongly
agree agree disagree disagree
25. I am good at predicting how someone will feel. strongly slightly slightly strongly
agree agree disagree disagree
26. I am quick to spot when someone in a group is strongly slightly slightly strongly
feeling awkward or uncomfortable. agree agree disagree disagree
27. If I say something that someone else is offended strongly slightly slightly strongly
by, I think that that’s their problem, not mine. agree agree disagree disagree
28. If anyone asked me if I liked their haircut, I would strongly slightly slightly strongly
reply truthfully, even if I didn’t like it. agree agree disagree disagree
29. I can’t always see why someone should have felt strongly slightly slightly strongly
offended by a remark. agree agree disagree disagree
30. People often tell me that I am very unpredictable. strongly slightly slightly strongly
agree agree disagree disagree
31. I enjoy being the center of attention at any social strongly slightly slightly strongly
gathering. agree agree disagree disagree
32. Seeing people cry doesn’t really upset me. strongly slightly slightly strongly
agree agree disagree disagree
APPENDIX (Continued)
483924.qxd 3/18/04 21:33 Page 172
The Empathy Quotient: An Investigation of Adults with Asperger Syndrome 173
33. I enjoy having discussions about politics. strongly slightly slightly strongly
agree agree disagree disagree
34. I am very blunt, which some people take to be strongly slightly slightly strongly
rudeness, even though this is unintentional. agree agree disagree disagree
35. I don’t tend to find social situations confusing. strongly slightly slightly strongly
agree agree disagree disagree
36. Other people tell me I am good at understanding strongly slightly slightly strongly
how they are feeling and what they are thinking. agree agree disagree disagree
37. When I talk to people, I tend to talk about their strongly slightly slightly strongly
experiences rather than my own. agree agree disagree disagree
38. It upsets me to see an animal in pain. strongly slightly slightly strongly
agree agree disagree disagree
39. I am able to make decisions without being strongly slightly slightly strongly
influenced by people’s feelings. agree agree disagree disagree
40. I can’t relax until I have done everything I had strongly slightly slightly strongly
planned to do that day. agree agree disagree disagree
41. I can easily tell if someone else is interested or strongly slightly slightly strongly
bored with what I am saying. agree agree disagree disagree
42. I get upset if I see people suffering on news strongly slightly slightly strongly
programmes. agree agree disagree disagree
43. Friends usually talk to me about their problems as strongly slightly slightly strongly
they say that I am very understanding. agree agree disagree disagree
44. I can sense if I am intruding, even if the other strongly slightly slightly strongly
person doesn’t tell me. agree agree disagree disagree
45. I often start new hobbies but quickly become strongly slightly slightly strongly
bored with them and move on to something else. agree agree disagree disagree
46. People sometimes tell me that I have gone too far strongly slightly slightly strongly
with teasing. agree agree disagree disagree
47. I would be too nervous to go on a big rollercoaster. strongly slightly slightly strongly
agree agree disagree disagree
48. Other people, often say that I am insensitive, strongly slightly slightly strongly
though I don’t always see why. agree agree disagree disagree
49. If I see a stranger in a group, I think that it is up to strongly slightly slightly strongly
them to make an effort to join in. agree agree disagree disagree
50. I usually stay emotionally detached when watching strongly slightly slightly strongly
a film. agree agree disagree disagree
51. I like to be very organized in day-to-day life and strongly slightly slightly strongly
often make lists of the chores I have to do. agree agree disagree disagree
52. I can tune into how someone else feels rapidly and strongly slightly slightly strongly
intuitively. agree agree disagree disagree
53. I don’t like to take risks. strongly slightly slightly strongly
agree agree disagree disagree
54. I can easily work out what another person might strongly slightly slightly strongly
want to talk about. agree agree disagree disagree
55. I can tell if someone is masking their true emotion. strongly slightly slightly strongly
agree agree disagree disagree
56. Before making a decision I always weigh up the strongly slightly slightly strongly
pros and cons. agree agree disagree disagree
57. I don’t consciously work out the rules of social strongly slightly slightly strongly
situations. agree agree disagree disagree
58. I am good at predicting what someone will do. strongly slightly slightly strongly
agree agree disagree disagree
59. I tend to get emotionally involved with a friend’s strongly slightly slightly strongly
problems. agree agree disagree disagree
60. I can usually appreciate the other person’s strongly slightly slightly strongly
viewpoint, even if I don’t agree with it. agree agree disagree disagree
Thank you for filling this questionnaire in. © SBC/SJW
483924.qxd 3/18/04 21:33 Page 173
174 Baron-Cohen and Wheelwright
REFERENCES
American Psychiatric Association (1994). Diagnostic and Statisti-
cal Manual of Mental Disorders (4th ed.). Washington, DC:
Author.
Asperger, H. (1944). Die “Autistischen Psychopathen” im
Kindesalter. Archiv fur Psychiatrie und Nervenkrankheiten,
117, 76–136.
Astington, J., Harris, P., & Olson, D. (1988). Developing theories of
mind. New York: Cambridge University Press.
Bailey, T., Le Couteur, A., Gottesman, I., Bolton, P., Simonoff, E.,
Yuzda, E., & Rutter, M. (1995). Autism as a strongly genetic
disorder: Evidence from a British twin study. Psychological
Medicine, 25, 63–77.
Baron-Cohen, S. (1995). Mindblindness: An essay on autism and
theory of mind. Boston: MIT Press/Bradford Books.
Baron-Cohen, S. (2002). The extreme male brain theory of autism.
Trends in Cognitive Sciences, 6, 248–254.
Baron-Cohen, S. (2003). The essential difference: Men, women and
the extreme male brain. London: Penguin.
Baron-Cohen, S., & Hammer, J. (1997). Is autism an extreme
form of the male brain? Advances in Infancy Research, 11,
193–217.
Baron-Cohen, S., Jolliffe, T., Mortimore, C., & Robertson, M. (1997).
Another advanced test of theory of mind: Evidence from very
high functioning adults with autism or Asperger Syndrome.
Journal of Child Psychology and Psychiatry, 38, 813–822.
Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autis-
tic child have a “theory of mind”? Cognition, 21, 37–46.
Baron-Cohen, S., O’Riordan, M., Jones, R., Stone, V., & Plaisted, K.
(1999). A new test of social sensitivity: Detection of faux pas
in normal children and children with Asperger syndrome. Jour-
nal of Autism and Developmental Disorders, 29, 407–418.
Baron-Cohen, S., Richler, J., Bisarya, D., Gurunathan, N., &
Wheelwright, S. (2003). The Systemising Quotient (SQ): An
investigation of adults with Asperger syndrome or high func-
tioning autism and normal sex differences. Philosophical Trans-
actions of the Royal Society, Series B, Special issue on “Autism:
Mind and Brain,” 358, 361–374.
Baron-Cohen, S., & Wheelwright, S. (2003). The Friendship Ques-
tionnaire (FQ): An investigation of adults with Asperger syn-
drome or high functioning autism, and normal sex differences.
Journal of Autism and Developmental Disorders, 33, 509–517.
Baron-Cohen, S., Wheelwright, S., & Jolliffe, T. (1997). Is there
a “language of the eyes”? Evidence from normal adults and
adults with autism or Asperger syndrome. Visual Cognition, 4,
311–331.
Baron-Cohen, S., Wheelwright, S., Scahill, V., Lawson, J., & Spong,
A. (2001a). Are intuitive physics and intuitive psychology in-
dependent? Journal of Developmental and Learning Disorders,
5, 47–78.
Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., & Clubley,
E. (2001b). The Autism Spectrum Quotient (AQ): Evidence from
Asperger syndrome/high functioning autism, males and females,
scientists and mathematicians. Journal of Autism and Develop-
mental Disorders, 31, 5–17.
Batson, C. D. (1991). The altruism question: Toward a social-
psychological answer. Hillsdale, NJ: Lawrence Erlbaum
Associates.
Blair, R. J. (1995). A cognitive developmental approach to mortal-
ity: Investigating the psychopath. Cognition, 57, 1–29.
Blair, R. J., Jones, L., Clark, F., & Smith, M. (1997). The psycho-
pathic individual: A lack of responsiveness to distress cues?
Psychophysiology, 34, 192–198.
Bolton, P., & Rutter, M. (1990). Genetic influences in autism. Inter-
national Review of Psychiatry, 2, 67–80.
Chapin, F. S. (1942). Preliminary standardization of a social insight
scale. American Sociological Review, 7, 214–225.
Cronbach, L. J. (1955). Processes affecting scores on understanding
of others and assuming “similarity.” Psychological Bulletin, 52,
177–193.
Davis, M. H. (1980). A multidimensional approach to individual dif-
ferences in empathy. JSAS Catalog of Selected Documents in
Psychology, 10, 85.
Davis, M. H. (1994). Empathy: A social psychological approach. CO:
Westview Press.
Davis, M. H., & Franzoi, S. L. (1991). Stability and change in ado-
lescent self-consciousness and empathy. Journal of Research in
Personality, 25, 70–87.
Dennett, D. (1987). The intentional stance. Cambridge, MA: MIT
Press/Bradford Books.
Dymond, R. F. (1949). A scale for the measurement of empathic abil-
ity. Journal of Consulting Psychology, 13, 127–133.
Dymond, R. F. (1950). Personality and empathy. Journal of Con-
sulting Psychology, 14, 343–350.
Eisenberg, N., & Miller, P. A. (1987). Empathy and prosocial
behavior. Psychological Bulletin, 101, 91–119.
Folstein, S., & Rutter, M. (1977). Infantile autism: A genetic study
of 21 twin pairs. Journal of Child Psychology and Psychiatry,
18, 297–321.
Folstein, S., & Rutter, M. (1988). Autism: Familial aggregation
and genetic implications. Journal of Autism and Developmen-
tal Disorders, 18, 3–30.
Fonagy, P., Steele, H., Steele, M., & Holder, J. (1997). Attachment
and theory of mind: Overlapping constructs? ACPP Occasional
Papers, 14, 31–40.
Frith, U. (1991). Autism and Asperger’s syndrome. Cambridge:
Cambridge University Press.
Gillberg, C. (1991). Clinical and neurobiological aspects of
Asperger syndrome in six family studies. In U. Frith (Ed.),
Autism and Asperger syndrome. Cambridge: Cambridge
University Press.
Gillberg, C. (1992). The Emanuel Miller Lecture, 1991: Autism and
autistic-like conditions: Subclasses among disorders of empa-
thy. Journal of Child Psychology and Psychiatry, 33, 813–842.
Hall, J. A. (1978). Gender effects in decoding nonverbal cues. Psy-
chological Bulletin, 85, 845–858.
Happe, F. (1994). An advanced test of theory of mind: Understand-
ing of story characters’ thoughts and feelings by able autistic,
mentally handicapped, and normal children and adults. Journal
of Autism and Development Disorders, 24, 129–154.
Hoffman, M. L. (1977). Sex differences in empathy and related
behaviors. Psychological Bulletin, 84, 712–722.
Hoffman, M. L. (1984). Interaction of affect and cognition in empa-
thy. In C. E. Izard & R. B. Kagan (Eds.), Emotions, cognition,
and behavior (pp. 103–131). Cambridge: Cambridge University
Press.
Hogan, R. (1969). Development of an empathy scale. Journal of
Consulting and Clinical Psychology, 33, 307–316.
Johnson, J. A., Cheek, J. M., & Smither, R. (1983). The structure of
empathy. Journal of Personality and Social Psychology, 45,
1299–1312.
Kerr, W. A., & Speroff, B. G. (1954). Validation and evaluation
of the empathy test. Journal of General Psychology, 50,
369–376.
Klin, A., Volkmar, F., Sparrow, S., Cicchetti, D., & Rourke, B. (1995).
Validity and neuropsychological characterization of Asperger
syndrome: Convergence with nonverbal learning disabilities
syndrome. Journal of Child Psychology and Psychiatry, 36,
1127–1140.
Kohler, W. (1929). Gestalt psychology. New York: Liveright.
Leslie, A. M. (1987). Pretence and representation: The origins of
“theory of mind.” Psychological Review, 94, 412–426.
Mead, G. H. (1934). Mind, self, and society. Chicago: University of
Chicago Press.
Mehrabian, A., & Epstein, N. (1972). A measure of emotional
empathy. Journal of Personality, 40, 525–543.
483924.qxd 3/18/04 21:33 Page 174
The Empathy Quotient: An Investigation of Adults with Asperger Syndrome 175
Mehrabian, A., Young, A. L., & Sato, S. (1988). Emotional empathy
and associated individual differences. Current Psychology:
Research & Reviews, 7, 221–240.
Perner, J., Frith, U., Leslie, A. M., & Leekam, S. (1989). Exploration
of the autistic child’s theory of mind: Knowledge, belief, and
communication. Child Development, 60, 689–700.
Piaget, J. (1932). The moral judgment of the child. London: Kegan
Paul, Trench, Trubner.
Smith, A. (1759). The theory of moral sentiments (Republished
1976 ed.). Oxford: Clarendon Press.
Stotland, E. (1969). Exploratory investigations of empathy.
In L. Berkowitz (Ed.), Advances in experimental social
psychology (Vol. 4, pp. 271–314). New York: Academic
Press.
Stotland, E., Sherman, S., & Shaver, K. (1971). Empathy and birth
order: Some experimental explorations. Lincoln: University of
Nebraska Press.
Titchener, E. (1909). Elementary psychology of the thought processes.
New York: Macmillan.
Wechsler, D. (1958). Sex differences in intelligence: The Measure-
ment and Appraisal of Adult Intelligence. Baltimore: Williams &
Wilking.
Wellman, H. (1990). Children’s theories of mind. Bradford: MIT Press.
Whiten, A. (1991). Natural theories of mind. Oxford: Basil Blackwell.
Wing, L. (1981). Asperger syndrome: A clinical account. Psycho-
logical Medicine, 11, 115–130.
Wing, L. (1988). The autistic continuum. In L. Wing (Ed.), Aspects
of autism: Biological research. London: Gaskell/Royal College
of Psychiatrists.
World Health Organization (1994). International classification of
diseases (10th ed.). Geneva, Switzerland: Author.
Yirmiya, N., Sigman, M., Kasari, C., & Mundy, P. (1992). Empathy
and cognition in high functioning children with autism. Child
Development, 63, 150–160.
483924.qxd 3/18/04 21:33 Page 175
