J. Child Psychol. Psychiat. Vol. 39, No. 3, pp.

307–322, 1998
Cambridge University Press.
' 1998 Association for Child Psychology and Psychiatry
Printed in Great Britain. All rights reserved
0021–9630}98 $15.00­0.00

Practitioner Review : Psychological and Educational Treatments for

Autism
Patricia Howlin
St George’s Hospital Medical School, London, U.K.

The review discusses various interventions that have been used in the treatment of children
with autism. It concludes that no single mode of treatment is ever likely to be effective for all
children and all families. Instead, intervention will need to be adapted to individual needs
and the value of approaches that involve a functional analysis of problems is explored. It is
suggested that many so called ‘‘ challenging ’’ behaviours result from the child’s fundamental
difficulties in communication and social understanding, or from the ritualistic and
obsessional tendencies that are also characteristic of autism. Possible ways in which parents
and teachers might deal with problems in these areas are discussed. The importance of early
diagnosis, and with it early, practical advice for families is stressed.

Keywords : Autism, intervention, prevention of problems.

Background & Gould, 1979), recent studies have suggested much

higher prevalence rates. Fombonne (1997), in a recent
Autism is a pervasive developmental disorder that is overview of epidemiological research, found estimates to
usually apparent from early childhood (Volkmar, Stier, vary from 0±7 to 15±5 per 10,000. His analysis indicates
& Cohen, 1985). It is characterised by profound deficits in that although the average figure for children with
communication and social understanding and by ritu- ‘‘ classic ’’ autism is around 5 per 10,000, the rates are
alistic and obsessional behaviours. Although associated much higher if children within the wider ‘‘ autistic
with a range of possible causes, genetic factors are spectrum ’’ are included. This term includes individuals
probably implicated in the majority of cases (Bailey et al., who ‘‘ share a triad of impaired social interaction,
1995 ; Le Couteur et al., 1996). communication, and imagination, associated with a rigid,
The condition was first systematically described by Leo repetitive pattern of behaviour … The triad can be
Kanner in the U.S.A. in 1943, but a remarkably similar recognised at all levels of intelligence and can occur alone
account, written by Hans Asperger in Austria, appeared or together with any other physical or psychological
at much the same time (for an annotated translation of disorder ’’ (Wing, 1996). On the basis of this definition,
Asperger’s initial paper, see Frith, 1991). Although there Wing concludes that the prevalence of individuals within
continues to be some debate about whether the two the total autistic spectrum may be as high as 91 per 10,000.
conditions described are quantitatively or qualitatively Despite the wide variation in these figures, it is clear
different (Lord & Rutter, 1994), current research suggests that children with autistic spectrum disorders are to be
that there are few substantive differences, either in early found much more frequently than was once thought.
history or outcome, between high-functioning children Moreover, because of their high rates of behavioural
with autism and those with Asperger syndrome disturbance, such children may make considerable
(Szatmari, Bartolucci, & Bremner, 1989 ; Wing, 1981). demands on Child Psychiatric or Learning Disability
ICD-10 and DSM-IV criteria for Asperger syndrome services.
note the same obsessional and social-communication
deficits as in autism ; the distinguishing features are the
presence of relatively normal cognitive skills and the lack The Course of the Disorder
of early language delays in individuals with Asperger
It is now widely accepted that autism has an onset in
syndrome.
infancy or early childhood (Volkman et al., 1985), and
many parents have serious concerns about their child’s
Prevalence development in the first year of life (Frith, Soares, &
Wing, 1993 ; Gillberg et al., 1990 ; Howlin & Moore,
Once considered to be a very rare condition, affecting
1997 ; Johnson, Siddons, Frith, & Morton, 1992 ; Smith,
only 3–4 individuals in every 10,000 (Lotter, 1996 ; Wing
Chung, & Vostanis, 1994). These early anxieties tend to
focus around abnormalities in communication, play, or
Requests for reprints to : Dr Patricia Howlin, Department of social responsiveness, or on repetitive behaviours.
Psychology, St George’s Hospital Medical School, Cranmer As children grow older, the pattern of their devel-
Terrace, London SW17 0RE, U.K. opment is largely affected by their degree of cognitive

307
308 PATRICIA HOWLIN

impairment. Although the syndrome of autism can occur behaviours in order to foster social contacts, also claims
in individuals of all levels of ability, the majority (around ‘‘ miraculous ’’ results. Another approach, currently re-
70–75 %) have some associated learning disabilities and ceiving much publicity, is the intensive early intervention
around 50 % have an IQ below 50. In those with severe to programme of Lovaas and his colleagues (McEachin,
profound cognitive impairments useful speech is unlikely Smith, & Lovaas, 1993 ; Lovaas, 1993 ; Perry, Cohen, &
to develop ; this group also tends to develop more DeCarlo, 1995). It is said that 42 % of children with
disturbed behaviours, such as self-injury, and will almost autism ‘‘ maintain normal functioning at follow-up ’’
invariably require specialist education and life-long care. (average age of 11±5 years) if they are exposed to very
In the 20 % or so of individuals who have an IQ within the intensive (40 hours per week), home-based behavioural
normal range, outcome is much more variable. Most programmes from around the ages of 2 to 4 years (Lovaas,
individuals tend to improve with age (Eaves & Ho, 1996 ; 1996). Although the value of behavioural approaches has
Gillberg & Steffenberg, 1987 ; Lockyer & Rutter, 1969, clearly been demonstrated in many studies, including
1970 ; Piven, Harper, Palmer, & Arndt, 1995), but those of Lovaas himself, these recent claims have aroused
whereas for some, adolescence brings about an upsurge in criticism concerning possible biases in subject selection,
problems, for others, notably those who become more problems of research design, and, most importantly, in
aware of their difficulties, it can be a period of remarkable the definitions of ‘‘ normality ’’ used (Schopler, Short, &
improvement and change (Kanner, 1973). Within this Mesibov, 1989). Thus, children are claimed to have
group, some individuals remain highly dependent ‘‘ recovered ’’ if they are of normal IQ and can be
throughout their lives ; others make successful careers, assimilated into mainstream education. Since around
they may live more or less independently, develop 20 % of all autistic children are of normal intellectual
friendships or even (in a tiny minority of cases) get ability, and many attend mainstream schools despite
married (see Howlin, 1997a, in press for review). Never- showing the characteristic triad of impairments, such
theless, even amongst those who make greatest progress, criteria cannot be used as evidence of normal or even near
their communication and social problems continue to normal functioning (Mesibov, 1993).
affect many aspects of their lives, as first-hand accounts From a practical point of view, too, this approach
by people such as Donna Williams (1992, 1994), Temple poses many problems. The programme involves 40 hours
Grandin (1992), or Jim Sinclair (1992) clearly illustrate. of one-to one intervention a week for at least 2 years and,
whereas some families many be able to cope with such
Treatments for Autism demands, the cost for others in emotional, financial, or
practical terms is often far too high. Moreover, although
The fact that some children do show substantial programmes of this kind may enhance short-term prog-
improvements as they grow older has led to claims that ress there is little to show that the benefits last beyond
particular treatments can significantly affect outcome, or childhood. Thus Howlin, Goode, and Rutter (unpub-
even bring about a ‘‘ cure ’’. Among such interventions lished data) found that, in adulthood, there were few if
are ‘‘ Holding ’’ therapy (Richer & Zappella, 1989 ; Welch, any significant differences, either in cognitive attainments
1988) ; music therapy (Trevarthen, Aitken, Papoudi, & or in measures of independent living, between individuals
Roberts, 1996) ; scotopic sensitivity training, which in- who had received home-based behavioural programmes
volves the wearing of specialist spectacles (Irlen, 1995) ; when they were younger, and their matched controls who
auditory integration, which focuses on desensitisation to had not.
sounds of particular frequencies (Rimland & Edelson,
1994, 1995 ; Stehli, 1992) ; as well as various drug and Advising Parents about Different Treatments
vitamin treatments (see Howlin, 1997b, for a detailed
review of different therapies). To date, there is no good evidence that any cures for
Few if any of these claims are supported by adequate autism exist. Many able children do well despite totally
experimental data. One such example, prominent in the inadequate provision, and to a great extent eventual
early 1990s, was ‘‘ Facilitated Communication ’’. This outcome is dependant on innate cognitive, linguistic, and
involved physical support (to the hand, arm, or wrist) to social abilities. However, clinicians are frequently ap-
help children use communication boards of various kinds. proached by parents who have heard or read about
The technique was said to demonstrate that individuals particular ‘‘ miracle cures ’’ and who, understandably,
with autism were actually of superior intellect, and was want to know whether the treatment will work for their
claimed to lead to ‘‘ Communication Unbound ’’ (Biklen, own child. Although it is important not to appear unduly
1993). Subsequent controlled investigations have con- pessimistic or dismissive, it is equally important to try to
sistently indicated that responses are almost invariably help parents understand that there can be no universal
under the control of the facilitator, not the client (Bebko, panacea. After all, even within physical medicine, drugs
Perry, & Bryson, 1996 ; Green, 1994). that have brought widespread advantages to many, such
A variety of teaching approaches has also been as aspirin or antibiotics, may prove ineffective or even
promoted as having a significant impact on outcome. harmful for certain individuals. Before parents succumb
These include the Japanese-based ‘‘ Daily Life Therapy ’’, to the temptation of parting from large amounts of
with its focus on highly structured, physically oriented money, or in some cases (as in specialist boarding school
programmes, practised in the Higashi schools (see Gould, provision) from their own child, the clinician should help
Rigg, & Bignell, 1991, for details). The ‘‘ Options ’’ them to try to obtain as much information as possible,
method of Kaufman (1981), which relies on therapists not only about the children for whom the treatment has
participating in the child’s ritualistic and obsessional worked, but also about the characteristics of those for
 PSYCHOLOGICAL AND EDUCATIONAL TREATMENTS FOR AUTISM 309

Table 1
The Principal Diagnostic Features of Autism and Their Association with Behaviour Problems
Associated problems

Area of deficit Less able children More able children

Impairments in communication
and understanding
Inadequate language Frustration, aggression ; unacceptable Inappropriate use of speech (echolalia, verbal
attempts to control environment routines, obsessional questions etc.)
Poor comprehension Anxiety, distress, and disruptive behaviours Apparent lack of co-operation
Lack of internal language No ability to play or occupy self Poor imaginative skills ; limited self control
Impairments in social
understanding
Lack of social awareness Withdrawal and isolation. Attempts to socialise often inappropriate ;
Disturbed}disruptive and inappropriate may offend or antagonise others.
behaviours in public. Inability to ‘‘ read ’’ others’ feelings makes
them appear insensitive, callous, even cruel.
Obsessions and rituals
Obsessional behaviour patterns Can severely limit the acquisition of other May involve other people in routines}rituals ;
more productive behaviours}skills can impose major limitations on other
people’s activities too
Disruption of routines Can result in serious distress, disruption and
aggression
Dislike of change Leads to very rigid and inflexible patterns of
behaviour and great distress and anxiety if
change is necessary
Obsessional interests May be pursued regardless of the
consequences. Constant talk about these can
antagonise others

whom it has been less successful. (Claims that the intervention, appropriately adapted to each individual
treatment works equally well for everyone need to be child’s pattern of strengths and weaknesses, can have a
treated with particular caution.) They should ask whether significant impact on minimising or avoiding behavioural
treatment seems to work better with older or younger problems and in helping to ensure that children are able
children ; for those with or without language ; or those to develop their existing skills to the full (Howlin &
who are more, or less, cognitively able. Parents should be Rutter, 1987).
encouraged to find out what sort of assessments are
carried out prior to treatment and what methods (other Understanding the Problems of Children with
than selective anecdotal reports) have been used to assess Autism
outcome. They also need accurate information about
what happens to other children with autism as they grow Articles on the association between autism and
older, so that reported outcomes following treatment can ‘‘ challenging ’’ behaviour appear throughout the child
be judged in the light of what might be expected in the psychiatric literature. Similarly, behavioural journals
absence of any special treatments. Before jumping head- abound with detailed accounts of the techniques that
long into therapy, families should be helped, too, to have been successful in modifying aggressive, destructive,
weigh up the overall cost of treatment ; the amount of and (although to a lesser extent) self-injurious behaviours.
time that will be involved ; any foreseeable pressures or However, whereas the seriousness of such problems
restrictions on other aspects of family life ; or the possible cannot be underrated, it is also important to recognise
impact on the child of having to cope with major change, that, given the severity of their social and communication
or even separation from the family. Finally, detailed deficits and their need for ritual and routine, many
information should also be provided on local facilities, children with autism perhaps show far fewer challenges
support groups, educational provision, and so forth. than might be expected. Imagine for a moment how any
These are unlikely to have been widely advertised, nor to ‘‘ normal ’’ person might react to a world in which they are
have featured in the national press, but may well be able able to understand almost nothing of what is happening
to provide families and children with much-needed, and around them ; in which they are thrown daily into an ever-
sometimes highly effective, help. If parents are better changing and unpredictable environment ; where they
informed of what is available within their own area, they lack even the rudimentary verbal skills necessary to make
may find less need to seek solutions further afield. their needs known ; where they have no access to the
The clinician needs to avoid encouraging unrealistic internalised, imaginative facilities that are so crucial for
expectations about future progress whilst not giving dealing effectively with anxiety, uncertainty, and distress.
parents too pessimistic a view of what the child is likely to Most such ‘‘ normal ’’ individuals would rapidly resort to
achieve. Thus, it is now well established that early a whole range of retaliatory behaviours, which if simply
310 PATRICIA HOWLIN

responded to by ‘‘ time-out ’’, or ‘‘ extinction ’’, or even strategies to those appropriate for a highly verbal 14-
rewards for ‘‘ incompatible behaviour ’’ would probably year-old with an IQ of 130. Programmes that are pitched
escalate dramatically. Of course, problem behaviours in inappropriately high or low will be unlikely to succeed.
autism, as in any other condition, will be maintained or Thus, standardised cognitive and linguistic assessments
increased by the attention or reinforcement they elicit, can be a crucial component of any intervention, since
and techniques such as time out, extinction, or differential informal observations alone may give a misleading
reinforcement can serve an important role (see Emerson, impression on intellectual ability. Children who appear
1995). However, they are not specific to the treatment of alert and interested in their environment, or who have
autism. The focus of this article, therefore, will be on the one or two isolated skills, may mistakenly be viewed as
underlying causes of behavioural disturbances and the very intelligent ; those who are able to follow simple
role that social, communication, and obsessional dif- instructions (often only with accompanying gesture or
ficulties play in causing or maintaining these (see Table other cues) may be described, quite erroneously, as
1). Intervention strategies, too, will concentrate on the ‘‘ understanding every word you say ’’ ; children who have
need to improve functioning in these areas, rather than on an extensive expressive vocabulary may actually have
the direct elimination of problems. very limited comprehension skills. Conversely, children
whose speech is slow and halting, and who appear to
show little interest in the activities around them, may be
Developing Intervention Programmes classified as having severe learning difficulties, when in
Pre-treatment Assessments fact many aspects of their cognitive development lie
within the normal range.
Before embarking on any treatment programme it is Unfortunately, because of the social and communi-
essential to obtain adequate information, not only on the cation difficulties associated with autism, it is often
behaviours to be modified, but on the child as well. This believed that traditional psychometric testing has little or
is because apparently similar problems can have very no role to play in assessment. Many psychologists,
different causes in different children, or within the same especially within the educational system, still tend to rely
child at different times. For example, ‘‘ aggressive be- on unstructured observations. Although these are un-
haviour ’’, may result from a child’s inability to com- doubtedly important, a relatively brief testing session can
municate ; because he or she lacks more effective strategies reveal unsuspected areas of skill or deficit, which may be
to control the environment ; because of the attention such very relevant for intervention. Moreover, as long as the
behaviours receive ; because of frustration, distress, appropriate tests are used, cognitive assessments on
anxiety, or because of disruption to rituals and routines. autistic children are just as reliable as testing in other
Moreover, in many instances, a combination of these groups (Rutter, 1985). The Appendix provides infor-
variables may be operating. Pain or physical illness are mation on standardised tests of language and cognition
other important factors to bear in mind. Certain stereo- that have proved useful when working with this group of
typed and self-injurious behaviours, for example, have children (see also Marans, 1997).
been found to follow minor illnesses including dermatitis
and otitis media (Hall, 1997 ; Oliver, 1995). As children
with autism may suffer more frequently from ear in-
fections (Konstantareas & Homatidis, 1987), it is clearly Behavioural Assessment
important for the clinician to rule these out as a possible Traditional behavioural therapy has long espoused the
cause of episodic behavioural changes, especially in ABC approach to the analysis of behavioural problems.
nonverbal children. Gunsett, Mulick, Fernald, and Thus, following precise delineation of the Behaviour
Martin (1989) also stress the importance of carrying out to be modified, attempts are made to identify the
medical screening before any psychological programmes Antecedents and the Consequences of that behaviour.
are implemented in patients with profound learning However, in the case of many children with autism, a
disabilities. In a series of 13 cases, referred for self- focus on the observed behaviour does not necessarily lead
injurious behaviours, 10 were found to have a physical to the most appropriate form of treatment (Emerson &
basis for the behaviour. These included limb fractures, Bromley, 1995). It may also prove extremely difficult to
hernias, urinary tract infections, ear infections, bowel establish, with any degree of certainty, either the ante-
problems, incorrect medication, toxic levels of anti- cedents or the consequences of the behaviour as perceived
convulsants, and progressive brain deterioration. by the child. For example, a child might begin to self-
injure in a particular setting, because he or she has been
Individual Assessment reminded of an earlier (but no longer existing) distressing
occurrence that had previously ceased when self-injury
As noted earlier, autism is frequently referred to as a commenced. In such a case, direct observations will be of
‘‘ spectrum ’’ disorder (Wing, 1996), in that it can range little use in identifying the relevant variables.
from relatively mild to profoundly handicapping. Simi- Because the observed form of a challenging behaviour
larly, it can affect children across the entire cognitive may give few clues as to its real role, recent intervention
continuum, and is found in those who are nonverbal as studies have focused instead on the function or ‘‘ message ’’
well as those whose spoken language is apparently of that behaviour. The aim is to establish what it achieves
unimpaired. for all the individuals concerned (carers as well as
Dealing with the temper tantrums of a nonspeaking 4- children) and to explore the alternative behaviours that
year-old child with an IQ of 30 will require very different might be encouraged to replace it (Sturmey, 1996).
 PSYCHOLOGICAL AND EDUCATIONAL TREATMENTS FOR AUTISM 311

The Functional Analysis of Behaviour ascertaining how many unsuccessful studies may also
have been conducted.
Highly sophisticated, experimental studies of the Despite these difficulties, it is apparent that this
‘‘ challenging ’’ behaviours shown by children with autism approach to assessment and treatment can play a major
have consistently demonstrated that many such be- role in reducing challenging behaviours. A number of
haviours serve an important communicative function rating scales or questionnaires have now been produced
(Durand, 1990). Indeed, analyses of these behaviours to assist carers in identifying the possible functions of
suggest that they may sometimes be the only way in disruptive behaviours (see Sturmey, 1996). The most
which a child with limited linguistic abilities can widely used of these is the ‘‘ Motivation Assessment
rapidly, effectively, and predictably control his or her Scale ’’ (Durand & Crimmins, 1988), which attempts to
environment. classify behaviour into four main categories : attention
Five main functions of aggressive, self-injurious, seeking ; self stimulatory ; escape or avoidance ; or as
stereotyped, or other disruptive behaviours have been indicating the need for help or assistance. However, there
identified (Durand & Carr, 1991 ; Durand & Crimmins, are doubts about the reliability and validity of this scale
1988). These are : when used in naturalistic settings, and, more importantly,
(1) To indicate the need for help or attention. the four summary categories cannot encompass all the
(2) To escape from stressful situations or activities. possible reasons for disruptive behaviours. In particular,
(3) To obtain desired objects. they cannot identify idiosyncratic or multifunction causes
(4) To protest against unwanted events}activities. (Sturmey, 1995).
(5) To obtain stimulation. A somewhat less complex questionnaire has been
developed by Schuler, Peck, Willard, and Theimer (1989).
If the primary function of a behaviour can be identified it This can be used by parents or teachers and, by
is then possible to provide the child with alternative systematically exploring how the child expresses his or
means to obtain the same ends. The choice of strategy her need to do something (sit by someone, get attention,
taught will depend on the child’s cognitive and linguistic obtain food or other object, protest if something is taken
ability, but might range from teaching him to push a away, etc.), this process can again help to indicate how
button, lever, or switch, or to use signs, symbols, pictures, behaviours that are often viewed as ‘‘ inappropriate ’’
or words and simple phrases (such as ‘‘ Help me ’’). As (screaming ; self-injury ; tantrums ; aggression etc.) can
long as the newly acquired behaviour has a rapid and have important communicative functions. This infor-
predictable impact on the child’s environment this can mation can then be used to plan ways in which alternative
result in significant reductions in undesirable behaviours and more acceptable responses might be established.
(Durand & Carr, 1991). Moreover, by helping carers to appreciate that such
Despite the surge in publications emanating from this behaviours may be a function of poor communication
approach in recent years, it is important to be aware that skills, rather than being ‘‘ deliberate ’’ acts of aggression
the majority of studies has been conducted in highly or provocation, this approach can also have a very
staffed experimental settings. Detailed analyses of the positive effect on other people’s attitudes and responses
possible functions of undesirable behaviours may require towards the child.
considerable time, expertise, and technology and are In the following sections, strategies for dealing with the
often impracticable within mainstream settings. As problems that appear to underlie many challenging
Owens and MacKinnon (1993) note, ‘‘ Functional analy- behaviours are described. However, again, although there
sis isn’t as easy as ABC ’’. Emerson and Bromley (1995) are many single-case and small group studies, often using
also warn of the problems inherent in this approach. multiple baseline or ABA type designs, that testify to the
Human behaviour will always be determined by many success for these techniques, there are very few ran-
factors, and by different factors at different times, and domised treatment trials. Thus, information on overall
they found that often it was not possible to determine the success rates, or on the types of children or problems for
function underlying a particular challenging behaviour. which the treatments are most}least successful, remains
In around 25 % of cases, no specific function could be very limited.
identified, and a third of behaviours appeared to be
influenced by multiple factors. Hall (1997), in a very
sophisticated assessment of self-injurious and stereotyped Increasing Communicative Skills
behaviours in 16 children with severe to profound
learning disabilities, was able to identify a consistent Despite some intrinsic problems, the emphasis in
underlying function for self-injury in only 4 cases and for ‘‘ functional analysis ’’ studies on the need to develop
stereotyped behaviours in 6 cases. Moreover, knowledge effective communication skills has had a significant
of the function underlying one class of behaviours is not impact on therapy. It is estimated that around half of all
necessarily of any value in predicting the behavioural children with autism fail to develop functional speech
function underlying other forms of challenging behaviour (Lord & Rutter, 1994), and even amongst those with a
shown by the same individual. good expressive vocabulary there are persisting and
A further problem with studies employing this meth- pervasive impairments in the communicative use of
odology is that they do not tend to use random control language, and in understanding complex or abstract
trials. Most are multiple baseline, single-case, or small concepts.
group reports, and although those that have been For younger children, who are able to use some
published are certainly encouraging, there is no way of words or sounds spontaneously, individualised language
312 PATRICIA HOWLIN

Table 2
Assessment for Communication Strategies (from Layton & Watson, 1995)
Pictures}
Signing Pictographs Writing
Characteristics
Easily shaped Yes Yes Fairly
Portability High Moderate High
Permanence No Yes Yes
Speed Moderate}Low Low High}Moderate
Phrases possible Moderate Limited Yes
Iconicity Some Yes No
Reciprocity Limited Limited High
Skills required
Motor skills Yes No Yes
High co-operation Not initially Not initially Not initially
Demands on others’ understanding Yes No No

programmes are important for improving comprehen- acquisition of sign is, in fact, somewhat limited. Thus,
sion, increasing the complexity of speech, or correcting Kiernan (1983) found that although some children had
problems of intonation or articulation. However, it is acquired an extensive signing vocabulary (400 sign
essential that such programmes are aimed at a level combinations) or had begun to use speech after 2 to 3
that is appropriate to the child’s cognitive and linguistic years, others had managed to learn only 1 or 2 signs.
development and that the words or concepts involved are Other studies indicate that problems of generalisation
ones that are of direct relevance to the child. Language and maintenance are similar to those experienced
cannot be taught in brief, 1-hour speech therapy sessions, in verbal training programmes. Attwood, Frith, and
but language therapists can play a crucial role in ensuring Hermelin (1988) found that the signing of children with
that all those living and working with the child use autism was very similar to their use of language—i.e. it
effective and consistent strategies to encourage speech, was stereotyped, repetitive, and used mainly to achieve
develop imaginative skills, and improve understanding. the child’s immediate needs. Signing was rarely used to
Descriptive assessments of the way in which children use share experiences, to express feelings or emotions, or to
and understand language, such as the Derbyshire Lan- communicate in a reciprocal and spontaneous fashion.
guage Scheme, the Pragmatics Profile of Early Com- On the whole, a pictorially based system makes least
munication Skills (which is based on a structured parental demands on cognitive, linguistic, or memory skills,
interview, Dewart & Summers, 1988), or the Social Use of although again it is essential that the pictures or photo-
Language Programme (Rinaldi, 1992) can also be used to graphs used reflect the individual’s particular interests or
provide guidelines for improving receptive and conver- needs. It also seems to be important for the child to take
sational skills. As long as the words taught are of practical an active role in using and handling such materials (e.g.
value to the child, in that they allow him or her immediate using Velcro or stickers to indicate the activities com-
access to desired objects or activities, then artificial pleted, or to be done, rather than looking passively at
rewards should not be needed. Indeed, there is evidence charts constructed by others). The PECS (Picture Ex-
that the unnecessary introduction of extrinsic reinforcers change Communication System) of Bondy and Frost
may actually interfere with learning (Howlin & Rutter, (1996) is a good example of how this can be done.
1987). Once the association between the activity}object and
Follow-up studies indicate that most children with picture}symbol}chart is established, individualized sets
autism who have not developed useful speech by the age of photographs or pictures can be used very effectively to
of 6 or 7 years remain very impaired in their ability to increase both communication and understanding (Quill,
communicate verbally. For them, some form of alterna- 1995). Whatever system is used must be readily available
tive communication system will be required, and the to the child and his carers. Possible ways of ensuring easy
appropriate choice of system will depend on the child’s access are to have the picture}symbol}word sets on cards
particular pattern of skills and disabilities. Layton and in a simple ‘‘ Filofax ’’ system ; attached to the child’s belt ;
Watson (1995) provide a useful breakdown of the in a hip bag ; or worn, like an identity badge, on a chain.
different skills required for using signs, pictures or written Materials also need to be strongly bound, and quickly
words (see Table 2). replaceable. Equipment that is forever getting lost, or
Sign systems, especially those developed for use with dog-eared cards that no-one can read, are of little use.
children with learning disabilities, have been widely used Computerised communicative devices have become
to augment communication skills. The Makaton system, steadily more sophisticated in recent years. Some are now
for example (Walker, 1980), is extensively used in schools specially designed for children with autism, in that they
in the U.K. This has several different levels of complexity, have a specific focus on turn taking and reciprocal
and now incorporates symbols as well as signs. However, interaction. Interchangeable keyboards of increasing
in the case of children with autism, the evidence to show complexity make it possible for children to progress
that communication can be significantly enhanced by the gradually from single-symbol boards (with for example a
 PSYCHOLOGICAL AND EDUCATIONAL TREATMENTS FOR AUTISM 313

large red square or circle that will emit a sound to attract appropriate and noncommunicative, as well as sometimes
attention) to, eventually, the independent use of multi- being extremely irritating, careful analysis indicates that
symbol displays, which are personally tailored to the much echoing serves identifiable, and important, com-
individual’s own environment, needs or interests. Com- municative functions (Prizant & Schuler, 1987 ; Rydell &
puters can, of course, also be very valuable general Prizant, 1995). As with any other ‘‘ autistic ’’ behaviour it
teaching aids for more able children who may respond is crucial to assess the role that the echolalia serves before
better to visually, rather than verbally, presented ma- attempts are made to modify it.
terial. However, when using computers with this group, Echoing may be an indication of children’s lack of
care needs to be taken to ensure that some social understanding ; it may be important in helping them to
interaction is also required, otherwise an obsession with consolidate what others say, as well as providing them
the technology may take over. with the opportunity to practise new words or ex-
Some parents understandably express concerns that, pressions. Moreover, in that echolalia is likely to increase
by focusing on alternative forms of communication, this when children are distressed or anxious, it may signify
will minimise the chances of their child ever learning to that they are experiencing undue pressure (Rydell &
speak. However, developing competence in the use of Mirenda, 1994). Repetition can also play a role in
signs or symbols may actually encourage some previously rehearsing potentially worrying situations, in dealing
nonverbal children to use speech (Howlin, 1989). And, with feelings of anger, or in helping to allay anxiety.
for those who do not, it is crucial to establish an effective Greater understanding of why such behaviours may
and socially acceptable form of communication as early occur, and recognition of the potential importance of
as possible. these, should lead to more appropriate intervention
It is also important to be aware that children with no strategies, with a focus on altering the factors causing the
apparent language difficulties may also require some echolalia, rather than on the symptom itself.
augmentative communication systems from time to time. Ensuring that instructions or questions are fully under-
Almost all children with autism, no matter how able, stood, by simplifying the language used, or supple-
have difficulties with abstract language or in dealing with menting this with pictures, written instructions, or other
complex sequences of instructions. Thus, although they cues, may significantly reduce stereotyped and echolalic
may understand the individual words spoken, they may speech (Rydell & Prizant, 1995). Repetitive questioning
well misinterpret or fail to understand the underlying (which often tends to escalate the more adults respond to
meaning of what is said. If told, ‘‘ Go and ask your mum it), may also be reduced by directing the child to charts,
if she wants a cup of tea ’’ they may well do so, but are pictures, calendars, or lists, which provide him or her with
unlikely to bring back any reply ! In such circumstances it the required information in a more permanent form.
is essential that the speaker makes sure that the words Dealing with unnecessary stress, by supplying the child
used adequately convey what is wanted (i.e. ‘‘ Go and ask with adequate help in cognitively or socially demanding
your mum if she wants a cup of tea and then come situations, may also have a considerable impact. Mini-
straight back and tell me what she said ’’). In addition, mising disruption to daily routine, ensuring that daily life
especially when the task requirements are more complex, is as predictable and as consistent as possible, and that
checklists of simple instructions, picture or cartoon necessary changes are predicted well in advance, can all
sequences of the activities to be completed, or symbols help to decrease the frequency of repetitive speech.
designating the tasks to be done, can all help greatly to Stereotyped speech (bombarding visitors with ques-
improve co-operation. tions about the make of their car ; or lengthy monologues
One example of a very successful, visually based about the lighting systems on particular railway net-
instructional system is the TEACCH educational pro- works) may be an indication that, although the child
gramme, initially developed in North Carolina (Schopler, wishes to make social contact, he or she lacks the
Mesibov, & Hearsey, 1995). This relies heavily on visual necessary conversational skills. Again, help to initiate
cues or ‘‘ jigs ’’, so that throughout the child’s school day, and cope with basic conversational exchanges—perhaps
different coloured work areas or different coloured by utilising role-play or drama techniques—is often the
containers are used to indicate where the child should be ; most effective way of addressing problems of this kind.
what he or she should be doing ; where work should begin Of course, stereotyped speech may sometimes be used
or be placed when finished ; and even where to play. This deliberately because of the attention it generates. Re-
combination of a highly structured and visually based petitive phrases, swearing, or other provocative utter-
programme can be very effective in improving work- ances often provoke a rapid response from adults and
related skills and reducing inappropriate behaviours. other children, and are all too easily reinforced. In such
However, as with any treatment package, it is important cases, as well as increasing the child’s repertoire of
that the basic components are adapted to suit the needs of appropriate speech, strategies involving extinction (no re-
the individual and his or her environment. It is also inforcement) or ‘‘ time out ’’ (removal of ongoing rewards)
important to plan for the gradual reduction of such cues may be necessary. Such techniques, if used consistently,
if the child is eventually to be able to function in less can have a rapid and positive effect on behaviour—the
structured surroundings (Jordan & Powell, 1995). problem lies in being consistent. Thus, swearing or other
abusive language may be relatively easy to ignore by
Echolalic Speech parents who are relaxed and in control. After a sleepless
night ; when under stress ; or in conditions where they
Echolalia, both immediate and delayed, is a common have little control (e.g. in church, on the top of a bus, or
feature of autism. Although often considered as in- in a supermarket queue) it can be almost impossible not
314 PATRICIA HOWLIN

to respond. Intermittent reinforcement of this kind can doing things that in themselves are not unacceptable, but
actually result in an increase in unwanted behaviour, and at the wrong time, in the wrong place, or with the wrong
hence extinction programmes, although highly effective person).
in principle, can prove much more difficult to implement In many ways, the more obvious social problems are
in practice. Parents also need to be given support during often easier to deal with. Firm and consistent guidance is
the early stages of such programmes when the ‘‘ extinction needed from the outset about what behaviours are, or are
burst ’’ (an initial increase in the behaviour when re- not, acceptable. If, as a small child, there are clear and
inforcement is first withdrawn) is likely to occur. invariable rules—such as never undressing or mastur-
bating in public, not touching strangers or their
Improving Others’ Communication skills belongings—and if the child also learns that disruptive
behaviour in response to such prohibitions results in the
Although, with appropriate help and encouragement, cessation of more pleasurable activities, then such be-
children with autism may show improvements in both haviours are much less likely to give rise to problems in
their use and understanding of language, the communi- later childhood. Children with autism are, by definition,
cation deficit is central to the disorder and no amount of somewhat rigid in their behaviour patterns, and if
therapy will overcome this entirely. Much can be acceptable behaviours are established when they are very
achieved, however, by making the adults in the child’s young these will tend to persist. The converse, of course,
environment more attentive to the language that they is also true, so that once unacceptable behaviours take
themselves use. Instructions should be simple and concise, hold, they will be very difficult to shift in later years,
and every attempt made to ensure that the words used especially as the child grows bigger. Parents, however,
actually mean what they say. Metaphor, slang, and may need a great deal of help and support during these
colloquialisms are all best avoided, whilst irony or threats early years if they are to develop effective management
such as ‘‘ I’ll kill you if you do that again ’’ can cause strategies. Young children with autism are clearly often
untold distress. Even vague concepts such as ‘‘ Perhaps ’’ ; deeply disturbed and confused, and most parents, un-
‘‘ I’ll think about it ’’ ; ‘‘ We’ll see ’’ are liable to produce willing to increase their distress, will tend to give in to
confusion and anxiety. If the child is required to do many of their demands. Helping parents to understand
something, an unambiguous request such as ‘‘ Please give when it is acceptable or necessary to say ‘‘ No ’’, and to
me the bread ’’ will be more productive than a phrase such recognise when consistency is crucial, can help to avoid
as ‘‘ Can you pass the bread ? ’’ (which might well be future problems. Removing a screaming 3-year-old from
answered in the affirmative but without resulting in any a shop because he cannot have what he wants may be
action). Other apparently minor changes in wording can embarrassing enough ; attempts to remove a screaming
have surprising effects. One girl’s severe distress at being 13-year-old will prove far more difficult.
told, prior to a trip to France, that she would be ‘‘ going It is also important that parents are made aware of
to sleep on the train ’’ changed to pleasure and relief when behaviours that, although not necessarily inappropriate
this was altered to ‘‘ going to bed in the train ’’. A teenage in a young child, may become progressively more
boy who became rude and abusive if ever he was asked by unacceptable as he or she grows older. A young girl who
teachers or doctors ‘‘ What year is your birthday ? ’’ (the warmly hugs and kisses everyone she meets, or a little boy
obvious answer, to him, was ‘‘ Every year ’’) answered who loves the feel of women’s tights, may be treated with
politely if they asked instead, ‘‘ What year were you fond indulgence. The same behaviours in older teenagers
born ? ’’ or adults will provoke a very different response ! Dif-
Unfortunately, predicting in advance what particular ficulties in social understanding and awareness mean
turn of phrase is likely to give rise to problems is very that the child with autism will either be impervious to
much a matter of trial and error. However, whenever a other peoples’ changed reactions or that they will be
request is not complied with, or if a statement meets with totally confused by the fact that behaviours that were
an upsurge in echolalia, irritation, or anxiety, the speaker once tolerated, even encouraged, are suddenly deemed
should first assume that what he or she has said has been to be ‘‘ wrong ’’. On the whole, it is far preferable to
misunderstood or misinterpreted. Simplifying or chang- introduce simple and invariable rules (you only kiss
ing the words that are used may have a much greater people in the family ; you can only touch mummy’s tights)
impact than attempts to modify the child’s response. And that may be relaxed in later years if necessary, than to
again, the value of written or pictorial cues to augment have initially very loose guidelines that suddenly have to
the words used cannot be over-emphasised. be made more restrictive. A toddler who has been allowed
to take off all his clothes whenever he wanted will find it
Ameliorating Social Difficulties very difficult to change this behaviour when he begins
attending school. On the other hand, a child who has only
The social impairment in autism affects almost every ever been allowed to take her clothes off in the home can
aspect of the child’s functioning, whatever his or her be taught, as she grows older, that it may be acceptable to
intellectual ability. In children who are more severely remove her clothes in other specified situations, such as
handicapped, highly inappropriate behaviours such as the family doctor’s surgery.
screaming, undressing, or masturbating in public may be The major problem, here, is that social behaviour is not
a major cause of disturbance. In the case of those who are governed by simple rules ; if such rules do exist they are
more able, the problems tend to be much more subtle, highly complex and constantly changing according to the
and include impairments in empathy, social understand- social context. Nevertheless, as far as the child with
ing, or reciprocity and synchronisation (i.e. saying or autism is concerned it is preferable to have consistent (if
 PSYCHOLOGICAL AND EDUCATIONAL TREATMENTS FOR AUTISM 315

sometimes inadequate) rules than no rules at all. It is autism (Howlin, Baron-Cohen, Hadwin, & Swettenham,
dealing with more complex and subtle social deficits that in press).
presents much more of a challenge. Knowing how to
make friends, recognising what other people are feeling Coping with Obsessional and Ritualistic
or thinking, and reacting appropriately, are fundamental Behaviours
human aptitudes ; they are not rule-based skills that are
acquired through teaching. Thus, interventions designed Ritualistic and obsessional behaviours are a further
to overcome such basic deficits are almost certain to be major cause of problems in autism. Many different ways
limited in their effectiveness. There is some evidence that of dealing with these problems have been reported in the
social skills groups, specifically designed for children or literature, but generally it seems that a ‘‘ graded change ’’
adults with autism, can improve certain aspects of social approach to intervention is the most effective. For a child
functioning (Mesibov, 1984 ; T. I. Williams, 1989), but with autism, obsessional and ritualistic activities often
on the whole generalisation to untrained settings tends play a crucial role in reducing anxiety or in providing
to be limited. Social skills training is best conducted them with some control over what is otherwise a
in each and every situation to which the child is exposed, very confusing and unpredictable world (see Jolliffe,
so that he or she learns how to respond appropriately at Lansdown, & Robinson, 1992). If an attempt is suddenly
home, with relatives, in shops, at school, or with the peer made to restrict or prohibit such behaviours, this can lead
group. Each of these situations will require different to unacceptably high levels of anxiety and distress, and
social strategies and teaching in situ is far more likely because of the resulting disturbance most parents quickly
to be effective than teaching in the relative isolation give in. Moreover, without careful planning, children
of a ‘‘ social skills group ’’. may well develop replacement rituals or obsessions that
Learning how to interact appropriately with children prove even more disruptive.
of their own age is often one of the most difficult and In order to maintain parental co-operation and con-
demanding tasks for a child with autism. The ‘‘ rules ’’ of sistency, it is generally more effective to modify the
engagement ; of knowing how to enter a group of behaviour gradually, until it no longer interferes with the
children ; how to join in with their activities ; and how to child’s, or the family’s, other activities. Howlin and
talk to them, are all highly complex, unwritten, generally Rutter (1987) and Schopler (1995) describe a variety of
poorly understood (Dodge, Schlundt, Schocken, & strategies that can be used to reduce obsessional and
Delugach, 1983), and hence almost impossible to teach. ritualistic behaviours. However, the crucial goals are to
Because of this, some researchers have shifted the focus of minimise anxiety and distress (for both parents and
attention onto nonautistic peers, systematically teaching child) ; to aim for gradual but achievable behavioural
them to play and interact more effectively with the child change, rather than dramatic improvements ; to weigh
with autism (Lord, 1995a ; Wolfberg & Schuler, 1993). up the potential advantages, as well as obvious dis-
Roeyers (1996) has also shown that simply providing advantages, of the obsession and wherever possible to
nonhandicapped peers with information about children capitalise on these.
with autism and general instructions about ways to The following guidelines seem to be particularly
encourage them to play can improve the frequency and helpful.
style of joint interactions. Programmes of this kind can (1) Establish clear and consistent rules for : Where and
result in important short-term gains, and are clearly When the activity is permitted ; Who it can be
important for improving opportunities for integration. carried out with ; or How long it can go on for. This
Nevertheless, they do require skilled input from teachers ensures that the child knows not only when the
if the interactions are to be effective, and it can prove behaviour is not permitted, but also when it is
difficult to maintain peers’ co-operation over the longer allowed.
term (Lord, 1984). (2) Ensure that change is introduced one step at a time,
Other recent approaches to the treatment of social so that any distress to the child is kept to a
problems have focused on more fundamental deficits in minimum. Setting very small goals optimises the
‘‘ theory of mind ’’. The inability of children with autism chance of long-term success.
to ‘‘ mind-read ’’, i.e. to understand other people’s beliefs, (3) Explore possible underlying factors. High levels of
ideas, thoughts or feelings, has received much attention obsessional behaviour are often an indication of
over recent years (see Baron-Cohen, 1995, for an excellent uncertainty, anxiety, or distress. Such problems
summary). Several studies have found that even after can be reduced significantly by ensuring that the
relatively brief intervention programmes involving com- child’s daily programme is predictable, and
puters, pictures, photographs, toys, or actors, children appropriately stimulating and structured.
with autism do show improvements in their ability to (4) Consider possible environmental modifications. Re-
understand beliefs and emotions (Swettenham, 1995 ; ducing unnecessary demands on the child, en-
Ozonoff & Miller, 1995 ; Hadwin, Baron-Cohen, Howlin, couraging more flexible attitudes in adults, or
& Hill, 1996). Not unexpectedly, given the brevity and making relatively simple modifications to the daily
limitations of such programmes, generalisation to other, routine or environment can also help to reduce
untrained aspects of ‘‘ theory of mind ’’ is poor. Never- obsessional behaviours. In mainstream schools, for
theless, even this limited success suggests that training example, many children with autism become very
packages specifically designed to increase the ability to distressed (and hence more ritualistic) if they are
‘‘ mind-read ’’ could be an important and valuable ad- forced to take part in group games or ‘‘ join in ’’ at
dition to the educational curricula for many children with play times, or when they have to scramble to find
316 PATRICIA HOWLIN

Table 3
Stages in Setting Limits on a 3-year-old Boy’s Obsession with Thomas the Tank
Engine
1 Picture calendar, indicating when access to ‘‘ Thomas ’’ videos and train sets is allowed,
produced by parents.
2 Videos made unavailable before school ; ‘‘ Thomas ’’ book read over breakfast instead.
3 Videos restricted to one per evening after school ; weekend access unrestricted ; unlimited
access to train sets.
4 Limits on ‘‘ Thomas ’’ clothing imposed ; school agree that no ‘‘ Thomas ’’ clothes can be
worn there. No restrictions at home.
5 Access to ‘‘ Thomas ’’ train sets gradually restricted by increasing alternative activities
(including ‘‘ Thomas ’’ books, board games, etc.)
6 New electric train set provided at home ; allowed in conjunction with some ‘‘ Thomas ’’ toys ;
but not train sets
7 ‘‘ Thomas ’’ trains moved to grandmother’s house ; access only available at weekends.
Interests in trains, generally, encouraged.

somewhere to sit at the start of each lesson. Pharmacological Treatments

Allowing children to avoid such socially de-
manding situations by letting them spend play} Although behavioural approaches to intervention are
game times in the library or carrying out other usually the most desirable form of treatment for younger
tasks, or providing them with a set place in which autistic children, there are times when, because of severe
to sit, may again have much greater impact than a behavioural disturbance (especially self-injury or ag-
complex behavioural programme. gression), sleeping problems, overactivity, anxiety or
(5) Help children to cope with change. Although a depression, or marked obsessional and compulsive be-
structured and settled daily programme is essential haviours, medication may be considered. In the U.S.A.
for progress, it is neither possible, nor productive, pharmacological interventions are very common, and in
to avoid change completely. Fortunately, in many a recent survey of 838 carers of children with autism,
cases it is unpredictable change that causes most conducted by Aman, van Bourgondien, Wolford, and
difficulties. Thus, the solution is to ensure that the Sarphare (1995) it was found that over 50 % were taking
child is fully aware of what is going to happen at some form of drug or vitamin treatment. Although most
any time. Since verbal explanations are rarely medication was prescribed for specific behavioural rea-
adequate, visual representations (in the form of sons, or for physical problems such as epilepsy, some
picture calendars, symbols, or written lists) of drugs, notably fenfluramine and more recently certain
forthcoming activities, or of alterations to the selective serotonin uptake inhibitors (fluoxetine and
regular routine, are most likely to be effective. fluvoxamine ; Lewis, 1996 ; McDougle et al., 1996), have
(6) Make use of obsessions. Although it may sometimes been recommended as a means of reducing autistic
be necessary to eliminate certain ritualistic ac- symptomatology more generally.
tivities entirely, on the whole, once an acceptable However, evaluations of even the most commonly used
level of control is reached, obsessional behaviours drugs are frequently inadequate (Lewis, 1996). Campbell,
and interests can have many positive features. They Schopler, Cueva, and Hallin (1996) conclude that only
may serve as extremely powerful reinforcers for haloperidol, fenfluramine, naltrexone, clomipramine,
developing more productive activities (Howlin & and clonidine have been appropriately investigated, and
Rutter, 1987) ; they may also be an essential source all of these have their drawbacks and unwanted side
of comfort or self-occupation for a child with few effects. As for the myriad of other pharmacological
other interests or abilities. Follow-up studies sug- treatments that have been tried over the years, Campbell
gest, too, that if obsessional skills and interests are et al. (1996) warn ‘‘ No conclusions can be made con-
appropriately encouraged and developed they can cerning the efficacy and safety of these agents because the
play a crucial role in later social and educational findings are based on small sample sizes and open studies
integration (Kanner, 1973). without placebo control ’’. (For updates on the effects of
other drug treatments see recent reviews by Lewis, 1996 ;
Table 3 presents an example of how one child’s Campbell & Cueva, 1995.)
overwhelming obsession with Thomas the Tank Engine
trains was gradually overcome by his parents’ using Educational Placement
several of the strategies described above. By the time he
was 3 years old, they realised that the obsession was Although there may be no miracle cures for autism, it
becoming so pervasive that they had to intervene in some has long been recognised that the provision of appro-
way. At that age their son would wear only Thomas the priately structured educational programmes is one of the
Tank Engine clothes, spent all his time watching, or re- most important aspects of successful treatment (Rutter &
enacting Thomas videos, and talked of almost nothing Bartak, 1973 ; Schopler, Brehm, Kinsbourne, & Reichler
else. 1971).
 PSYCHOLOGICAL AND EDUCATIONAL TREATMENTS FOR AUTISM 317

Effective teaching programmes (such as TEACCH ; with severe emotional problems, psychoanalytically
Schopler et al., 1995) stress the importance of appropriate based interventions may be considered (Maratos, 1996),
environmental organisation and the use of clear visual although there is little good evidence that such ap-
cues to circumvent communication difficulties, as well as proaches are helpful (Campbell et al., 1996). Never-
the need to develop individually based learning pro- theless, for older, more able children with autism,
grammes. Within this framework, however, there are individual psychotherapy or counselling may be useful in
many different approaches to teaching, some of which helping them to deal with anxiety or depression, and the
involve integrated provision, some segregation, and some pain that comes from recognising their difficulties and
a mixture of the two. The essential component is that differences. However, experience suggests that this must
educational strategies and curricula should be adapted to be combined with direct practical advice on how to deal
the specific patterns of skills and disabilities shown by the with problems, otherwise children tend to become ob-
child with autism (see Jordan & Powell, 1995 ; Powell & sessed with the past, or with other possible explanations
Jordan, 1997 for descriptions of a variety of imaginative for their difficulties, making it almost impossible for them
and innovative techniques that can be used to enhance to ‘‘ move on ’’ in a positive way.
learning). It is also generally more productive—and
certainly more rewarding for all concerned—to focus on Help for Children with Asperger Syndrome
developing the child’s existing skills, rather than at- (or with High-functioning Autism)
tempting to overcome fundamental deficits. Equal em-
phasis needs to be placed, too, on meeting children’s Although often described as a ‘‘ mild ’’ variant of
social and emotional needs. For more able children in autism, the symptoms of Asperger syndrome are, in many
particular, this latter goal can prove difficult to achieve. cases, just as pervasive and as devastating as those of less
In specialised schooling it is often not possible to provide able children. However, because of their relatively high
them with sufficient intellectual stimulation ; on the other cognitive ability, and their apparently competent use of
hand, unless extra support is provided within mainstream language, this group of children is often least well served
school, teasing and bullying by other pupils, and lack of or understood. In fact, many have extensive linguistic and
understanding from teachers, may often result in severe comprehension difficulties (especially involving abstract
emotional stress. or complex concepts) ; their understanding of the more
There is good evidence (Rogers, 1996) to show that the subtle aspects of social interaction is often profoundly
most effective educational programmes are those that limited, and their obsessional interests and behaviours
begin early (between the ages of 2 to 4), and once the right also prove a barrier to social integration. Many, too, are
placement is found this can help greatly to reduce the painfully aware of their deficits and differences. However,
pressure on families. And if parents, teachers, and other only a minority receive any specialist provision. Most
professionals work together, this will markedly improve have to cope in mainstream school with little or no help.
the consistency of management techniques and help to Their parents may be dismissed as over-protective, or too
ensure the generalisation and maintenance of newly lax, and can find it very difficult to get the support or
acquired behaviours. Thus, for any clinician involved in advice that they need. Moreover, the children’s good
the treatment of autism, good liaison with Educational vocabulary, and even their well-developed obsessional
services, to ensure that the child receives early and interests, frequently give the impression that they are
appropriate provision, is a vital component of any capable of far higher levels of achievement than is actually
package of care. the case. Others’ expectations of their social and academic
potential tend to be unrealistically high, and when these
Other Approaches to Intervention expectations are not met the children are viewed as
negative, unco-operative, unmotivated, or rude and
The clinician may also need to ensure that the family’s manipulative. Seemingly so close to ‘‘ normality ’’, there is
needs in other areas are appropriately addressed. Respite constant pressure for them to ‘‘ fit in ’’ in ways that would
care, on a planned and regular basis, can offer parents the never be demanded of a less able autistic child. This can
rest they often so badly need, provide them with the lead to enormous pressure, resulting in extreme levels of
opportunity to spend time with their other children, and anxiety and stress, which in turn further impede social
give the child with autism the chance to spend time away and educational progress.
from home. Families may also need guidance to ensure Because of their very uneven profile of skills and
that they receive all the benefits to which they are entitled. deficits, these children may require even more highly
Money may not improve the child’s behaviour, but specialised help than those with global learning dif-
worrying about the lack of it can certainly interfere with ficulties. Unfortunately, such help is rarely available, and
parents’ ability to cope. Even apparently minor benefits, although provision for autistic children generally has
such as a Disabled Parking Badge, can make the improved markedly over the past few decades, our
difference between being able to take the child shopping knowledge of how to help this particular group effectively
or not. lags far behind.
Finally, of course, the child with autism may need help
in his or her own right. Cognitive-behavioural strategies The Need for Early Diagnosis
to help cope with anxiety, fears or anger may prove
effective with older, more able children, although these Despite growing awareness of the need for early
rarely work in isolation and generally require the co- intervention and support for children with autism and
operation of both school and family. For some children their families, diagnosis before the age of 3 years is still
318 PATRICIA HOWLIN

rare (Baron-Cohen et al., 1996). In a recent large-scale interfere with many other activities. Understandably, the
survey in the U.K. (Howlin & Moore, 1997), the average bewildered parents of a young infant tend to give in to the
age of diagnosis for children with autism was 5±5 years ; screams or tantrums that can occur because they have
for children with Asperger syndrome the age of diagnosis taken the ‘‘ wrong ’’ route, or have tried to change the way
was considerably later, at 11±3 years. It was also clear in which a task is completed. Over the years, however, the
from this survey that parents continue to face many child’s demands may become increasingly Draconian,
delays and frustrations in their attempts to obtain a until parents find themselves trapped in a web of complex
diagnosis. and elaborate routines.
It is important, of course, to be aware that early Without appropriate help, parents are unlikely to be
diagnosis has problems as well as advantages. Accurate able to identify potential problems, or will lack the
diagnosis in children aged 2 years or under is known to courage and consistency to respond firmly to these,
present difficulties. Judgements based on clinical diag- because of fears that resistance will further distress or
nosis may not always agree with diagnosis based on damage their child. It is the role of professionals to
formal diagnostic criteria, such as the ADI or CARS provide families with the information and support they
(Autism Diagnostic Interview-Revised ; Lord, Rutter, & need, in order to enable them to identify problems at an
Le Couteur, 1994 ; Childhood Autism Rating Scale ; early stage, and to help them evolve management
Schopler, Reichler, & Renner, 1986), and there can be strategies that will minimise the impact of the child’s
particular problems in distinguishing between children social, communicative, and obsessional problems in later
with autism and nonautistic, nonverbal children with life.
severe cognitive impairments (Lord, 1995b ; Lord,
Storuschuk, Rutter, & Pickles, 1993). Nevertheless, even
when children are as young as 2 years, diagnosis based on References
skilled clinical judgement is relatively stable, and by the Aman, M. G., van Bourgondien, M. E., Wolford, P. L., &
age of 3 years children tend to meet criteria on a variety Sarphare, G. (1995). Psychotropic and anticonvulsant drugs
of diagnostic measures (Lord, 1995). in subjects with autism : Prevalence and patterns of use.
Unnecessary delays in diagnosis have important prac- Journal of the American Academy of Child and Adolescent
Psychiatry, 34, 1672–1681.
tical implications. There is evidence (Rogers, 1996) to
Attwood, T., Frith, U., & Hermelin, B. (1988). The under-
show that the most effective intervention programmes are standing and use of gestures by autistic and Down’s syndrome
those that begin early—between the age of 2 to 4 years. It children. Journal of Autism and Developmental Disorders, 18,
also seems that the establishment of appropriate man- 241–258.
agement strategies in the early years can help to minimise, Bailey, A., Le Couteur, A., Gottesman, I., Bolton, P., Simenoff,
or even avoid, many subsequent behavioural problems E., Yuzda, E., & Rutter, M. (1995). Autism as a strongly
(Howlin & Rutter, 1987). genetic disorder ; evidence from a British twin study. Psycho-
First, it is apparent that the development of effective, if logical Medicine, 25, 63–77.
simple communication strategies from early childhood Baron-Cohen, S. (1995). Mindblindness : An essay on autism and
will almost certainly help to reduce or avoid disruptive theory of mind. Cambridge, MA : MIT Press.
behaviours, which may otherwise become the child’s Baron-Cohen, S., Cox, A., Baird, G., Swettenham, J.,
Nightingale, N., Morgan, K., Drew, A., & Charman, T.
principal means of controlling his or her environment (1996). Psychological markers in the detection of autism in
(Durand & Carr, 1991). infancy in a large population. British Journal of Psychiatry,
Second, as already noted, it is essential not to allow or 168, 158–163.
encourage behaviours in young children that will be Bebko, J. M., Perry, A., & Bryson, S. (1996). Multiple method
viewed as ‘‘ challenging ’’ or unacceptable as they grow validation study of facilitated communication : II Individual
older. All too frequently, when the behaviour of an older differences and subgroup results. Journal of Autism and
child or adolescent gives cause for concern, it becomes Developmental Disorders, 26, 19–42.
apparent that the behaviour itself has not changed, and Biklen, D. (1993). Communication unbound : How facilitated
may well have been in existence for many years pre- communication is challenging traditional views of autism and
viously. Problems arise not because the nature of the ability}disability. New York : Teachers College Press.
Bondy, A., & Frost, L. (1996). Educational approaches in pre-
behaviour has changed, but because other people’s school : Behavior techniques in a public school setting. In E.
attitudes to it have altered. Schopler & G. B. Mesibov (Eds.), Learning and cognition in
Third, behaviours may become unacceptable as in- autism (pp. 311–334). New York : Plenum Press.
dividuals become more skilled or determined at carrying Campbell, M., & Cueva, J. E. (1995). Psychopharmacology in
them out. For example, one young child had a relatively child and adolescent psychiatry : A review of the past seven
innocuous obsession with watching people’s washing years. Part II. Journal of the American Academy of Child and
machines. Family friends and neighbours were happy to Adolescent Psychiatry, 34, 1262–1272.
encourage this, but by the time he was 17 he was in Campbell, M., Schopler, E., Cueva, J. E., & Hallin, A. (1996).
constant trouble with the police for breaking into Treatment of autistic disorder. Journal of the American
property in order to indulge his obsession. Academy of Child and Adolescent Psychiatry, 35, 134–143.
Dewart, H., & Summers, S. (1988). Pragmatics profile of early
Fourth, particularly in the case of rituals and ob-
communication skills. Windsor, U.K. : NFER.
sessions, there is an inverse relationship with many Dodge, K., Schlundt, D., Schocken, I., & Delugach, J. (1983).
appropriate behaviours (Koegel, Valdez-Menchaca, & Competence and children’s sociometric status : The role of
Koegel, 1994). Thus there is real risk that if such peer group entries. Merrill-Palmer Quarterly, 29, 306–309.
behaviours are not brought under effective control when Durand, B. M. (1990). Severe behavior problems : A functional
the child is young they may steadily escalate until they communication approach. New York : Guilford Press.
 PSYCHOLOGICAL AND EDUCATIONAL TREATMENTS FOR AUTISM 319

Durand, B. M., & Carr, E. G. (1991). Functional communi- Howlin, P., & Moore, A. (1977). Diagnosis in autism : A
cation training to reduce challenging behaviour : Mainten- survey of over 1200 parents. Autism : The International
ance and application in new settings. Journal of Applied Journal of Research and Practice, 1, 135–162.
Behavior Analysis, 24, 251–254. Howlin, P., & Rutter, M. (1987). Treatment of autistic children.
Durand, B. M., & Crimmins, D. B. (1988). Identifying the Chichester, U.K. : Wiley.
variables maintaining self-injurious behavior. Journal of Irlen, H. (1995). Viewing the world through rose tinted glasses.
Autism and Developmental Disorders, 18, 99–117. Communication, 29, 8–9.
Eaves, L. C., & Ho, H. H. (1996). Brief report : Stability and Johnson, M. H., Siddons, F., Frith, U., & Morton, J. (1992).
change in cognitive and behavioral characteristics of autism Can autism be predicted on the basis of infant screening
through childhood. Journal of Autism and Developmental tests ? Developmental Medicine and Child Neurology, 34,
Disorders, 26, 557–570.
316–320.
Emerson, E. (1995). Challenging behaviour : Analysis and
Jolliffe, T., Lansdown, R., & Robinson, T. (1992). Autism : A
intervention. Cambridge : Cambridge University Press.
Emerson, E., & Bromley, J. (1995). The form and function of personal account. London : The National Autistic Society.
challenging behaviours. Journal of Intellectual Disability Jordan, R., & Powell, S. (1995). Understanding and teaching
Research, 39, 388–398. children with autism. Chichester, U.K. : Wiley.
Fombonne, E. (1997). Epidemiological studies of autism. In F. Kanner, L. (1943). Autistic disturbances of affective contact.
Volkmar (Ed.), Autism and developmental disorders. New Nervous Child, 2, 217–250.
York : Cambridge University Press. Kanner, L. (1973). Childhood psychosis : Initial studies and new
Frith, U. (1991). Autism and Asperger syndrome. Cambridge : insights. New York : Winston}Wiley.
Cambridge University Press. Kaufman, B. (1981). A miracle to believe in. New York :
Frith, U., Soares, I., & Wing, L. (1993). Research into the Doubleday.
earliest detectable signs of autism : What parents say. Com- Kiernan, C. (1983). The use of non-vocal communication
munication, 27, 17–18. systems with autistic individuals. Journal of Child Psychology
Gillberg, C., Ehlers, S., Schaumann, H., Jakobsson, G., and Psychiatry, 24, 339–376.
Dahlgren, S. O., Lindblom, R., Bagenholm, A., Tjus, T., & Koegel, L. K., Valdez-Menchaca, M. C., & Koegel, R. L.
Blidner, E. (1990). Autism under age 3 years : A clinical study (1994). Autism : Social communication difficulties and related
of 28 cases referred for autistic symptoms in infancy. Journal behaviors. In V. B. Van Hasselt & M. Hersen (Eds.), Ad-
of Child Psychology and Psychiatry, 31, 921–934. vanced abnormal psychology (pp. 165–187). New York :
Gillberg, C., & Steffenberg, S. (1987). Outcome and prognostic Plenum.
factors in infantile autism and similar conditions : A Konstantareas, M. M., & Homatidis, S. (1987). Ear infections
population-based study of 46 cases followed through puberty. in autistic and normal children. Journal of Autism and
Journal of Autism and Developmental Disorders, 17, 272–288. Developmental Disorders, 20, 591–593.
Gould, G. A., Rigg, M., & Bignell, L. (1991). The Higashi
Layton, T. L., & Watson, L. R. (1995). Enhancing communi-
experience : The report of a visit to the Boston Higashi School.
cation in non-verbal children with autism. In K. Quill (Ed.),
London : National Autistic Society Publications.
Grandin, T. (1992). An inside view of autism. In E. Schopler & Teaching children with autism : Strategies to enhance
G. B. Mesibov (Eds.), High functioning individuals with communication and socialization (pp. 73–104). New York :
autism (pp. 105–125). New York : Plenum Press. Delmar.
Green, G. (1994). The quality of the evidence. In H. C. Shane Le Couteur, A., Bailey, A., Goode, S., Pickles, A., Robertson,
(Ed.), Facilitated communication : The clinical and social S., Gottesman, I., & Rutter, M. (1996). A broader phenotype
phenomenon (pp. 156–226). San Diego, CA : Singular Press. of autism : The clinical picture in twins. Journal of Child
Gunsett, R. P., Mulick, J. A., Fernald, W. B., & Martin, J. L. Psychology and Psychiatry, 37, 785–802.
(1989). Brief report : Indications for medical screening prior Lewis, M. H. (1996). Brief report : Psychopharmacology of
to behavioral programming for severely and profoundly autism spectrum disorders. Journal of Autism and Develop-
mentally retarded clients. Journal of Autism and Develop- mental Disorders, 26, 231–236.
mental Disorders, 19, 167–172. Lockyer, L., & Rutter, M. (1969). A five to fifteen year follow-
Hadwin, J., Baron-Cohen, S., Howlin, P., & Hill, K. (1996). up study of infantile psychosis : III Psychological aspects.
Can we teach children with autism to understand emotions, British Journal of Psychiatry, 115, 865–882.
belief or pretence ? Development and Psychopathology, 8, Lockyer, L., & Rutter, M. (1970). A five to fifteen year follow-
345–365. up study of infantile psychosis : IV Patterns of cognitive
Hall, S. (1997). The early development of self-injurious behaviour abilities. British Journal of Social and Clinical Psychology, 9,
in children with developmental disabilities. Unpublished PhD 152–163.
thesis, University of London. Lord, C. (1984). The development of peer relations in children
Howlin, P. (1989). Changing approaches to communication with autism. In F. J. Morrison, C. Lord, & D. P. Keating
training with autistic children. British Journal of Disorders of (Eds.), Applied developmental psychology (pp. 166–230). New
Communication, 24, 151–168.
York : Academic Press.
Howlin, P. (1997a). Autism : Preparing for adulthood. London :
Lord, C. (1995a). Facilitating social inclusion : Examples from
Routledge.
Howlin, P. (1997b). Prognosis in autism : Do specialist peer intervention programs. In E. Schopler & G. Mesibov
treatments affect outcome ? European Child and Adolescent (Eds.), Learning and cognition in autism (pp. 221–239). New
Psychiatry, 6, 55–72. York : Plenum Press.
Howlin, P. (in press). Outcome in adult life for individuals Lord, C. (1995b). Follow-up of two-year-olds referred for
with autism. In F. Volkmar (Ed.), Autism and developmental possible autism. Journal of Child Psychology and Psychiatry,
disorders. New York : Cambridge University Press. 36, 1365–1382.
Howlin, P., Baron-Cohen, S., Hadwin, J., & Swettenham, J. Lord, C., & Rutter, M. (1994). Autism and pervasive de-
(in press). Teaching children with autism to mindread. A prac- velopmental disorders. In M. Rutter, E. Taylor, & L. Hersov
tical manual for parents and teachers. Chichester, U.K. : (Eds.), Child and adolescent psychiatry : Modern approaches
Wiley. (3rd Ed.) (pp. 569–593). Oxford : Blackwell.
320 PATRICIA HOWLIN

Lord, C., Rutter, M., & Le Couteur, A. (1994). Autism Quill, K. A. (1995). Teaching children with autism : Strategies to
Diagnostic Interview-Revised : A revised version of a di- enhance communication and socialization. New York : Delmar.
agnostic interview for caregivers of individuals with possible Richer, J., & Zappella, M. (1989). Changing social behaviour :
pervasive developmental disorders. Journal of Autism and The place of Holding. Communication, 23, 35–39.
Developmental Disorders, 24, 659–685. Rimland, B., & Edelson, S. M. (1994). The effects of Auditory
Lord, C., Storuschuk, S., Rutter, M., & Pickles, A. (1993). Integration Training on autism. American Journal of Speech-
Using the ADI-R to diagnose autism in preschool children. Language Pathology, 5, 16–24.
Infant Mental Health Journal, 14, 234–252. Rimland, B., & Edelson, S. M. (1995). Brief report : A pilot
Lotter, V. (1966). Epidemiology of autistic conditions in young study of Auditory Integration Training in autism. Journal of
children. I : Prevalence. Social Psychiatry, 1, 163–173. Autism and Developmental Disorders, 25, 61–70.
Lovaas, O. I. (1993). The development of a treatment-research Rinaldi, W. (1992). Social Use of Language Programme.
project for developmentally disabled and autistic children. Windsor, U.K. : NFER.
Journal of Applied Behavior Analysis, 26, 617–630. Roeyers, H. (1996). The influence of nonhandicapped peers on
Lovaas, O. I. (1996). The UCLA young autism model of service the social interaction of children with a pervasive devel-
delivery. In C. Maurice (Ed.), Behavioral intervention for opmental disorder. Journal of Autism and Developmental
young children with autism (pp. 241–250). Austin, TX : Disorders, 26, 303–320.
Pro-Ed. Rogers, S. J. (1996). Brief report : Early intervention in autism.
Marans, W. D. (1997). Communication assessment. In Journal of Autism and Developmental Disorders, 26, 243–246.
D. J. Cohen & F. R. Volkmar (Eds.), Handbook of autism Rutter, M. (1985). Infantile autism and other pervasive de-
and pervasive developmental disorders (pp. 427–447). New velopmental disorders. In M. Rutter & L. Hersov (Eds.),
York : John Wiley. Child and adolescent psychiatry : Modern approaches (2nd
Maratos, O. (1996). Psychoanalysis and the management of Edn.) (pp. 545–566). Oxford : Blackwell.
pervasive developmental disorders, including autism. In Rutter, M., & Bartak, L. (1973). Special educational treatment
C. Trevarthen, K. Aitken, D. Papoudi, & J. Robarts (Eds.), of autistic children : A comparative study. II. Follow-up
Children with autism : Diagnosis and interventions to meet their findings and implications for services. Journal of Child
needs (pp. 161–171). London : Jessica Kingsley. Psychology and Psychiatry, 14, 241–270.
McDougle, C. J., Naylor, S. T., Cohen, D. J., Volkmar, F. R., Rydell, P. J., & Mirenda, P. (1994). The effects of high and low
Heninger, G. R., & Price, L. H. (1996). A double blind, constraint utterances on the production of immediate and
placebo-controlled study of fluvoxamine in adults with delayed echolalia in young children with autism. Journal of
autistic disorder. Archives of General Psychiatry, 53, Autism and Developmental Disorders, 24, 719–730.
1001–1008. Rydell, P. J., & Prizant, B. (1995). Assessment and intervention
McEachin, J. J., Smith, T., & Lovaas, O. I. (1993). Long-term strategies for children who use echolalia. In K. A. Quill (Ed.),
outcome for children with autism who received early in- Teaching children with autism : Strategies to enhance com-
tensive behavioral treatment. American Journal of Mental munication and socialization (pp. 105–132). New York :
Retardation, 97, 359–372. Delmar.
Mesibov, G. B. (1984). Social skills training with verbal autistic Schopler, E., Brehm, S. S., Kinsbourne, M., & Reichler, R. J.
adolescents and adults : A program model. Journal of Autism (1971). Effects of treatment structure on development in
and Developmental Disorders, 14, 395–404. autistic children. Archives of General Psychiatry, 20, 174–181.
Mesibov, G. B. (1993). Treatment outcome is encouraging : Schopler, E., Mesibov, G. B., & Hearsey, K. (1995). Structured
Comments on McEachin et al. American Journal of Mental teaching in the TEACCH system. In E. Schopler & G. B.
Retardation, 97, 379–380. Mesibov (Eds.), Learning and cognition in autism (pp.
Oliver, C. (1995). Self-injurious behaviour in children with 243–267). New York : Plenum Press.
learning disabilities : Recent advances in assessment and Schopler, E., Reichler, R. J., & Renner, B. R. (1986). The
intervention. Journal of Child Psychology and Psychiatry, 36, Childhood Autism Rating Scale (CARS ) for diagnostic screen-
909–928. ing and classification of autism. New York : Irvington.
Owens, R. G., & MacKinnon, S. (1993). The functional analysis Schopler, E., Short, A., & Mesibov, G. (1989). Relation of
of challenging behaviours : Some conceptual and theoretical behavioral treatment to ‘‘ normal functioning ’’ : Comment on
problems. In R. S. P. Jones & C. B. Eayrs (Eds.), Challenging Lovaas. Journal of Consulting and Clinical Psychology, 57,
behaviour and intellectual disability : A psychological per- 162–164.
spective (pp. 224–239). Avon, U.K. : BILD Publications. Schuler, A. L., Peck, C. A., Willard, C., & Theimer, K. (1989).
Ozonoff, S., & Miller, J. (1995). Teaching theory of mind : A Assessment of communicative means and functions through
new approach to social skills training for individuals with interview : Assessing the communicative capabilities of indi-
autism. Journal of Autism and Developmental Disorders, 25, viduals with limited language. Seminars in Speech and
415–434. Language, 10, 51–61.
Perry, R., Cohen, I., & DeCarlo, R. (1995). Case study : Sinclair, J. (1992). Bridging the gap : An inside out view of
Deterioration, autism and recovery in two siblings. Journal of autism (Or, do you know what I don’t know ?). In E. Schopler
the American Academy of Child and Adolescent Psychiatry, & G. B. Mesibov (Eds.), High functioning individuals with
34, 233–237. autism (pp. 294–302). New York : Plenum Press.
Piven, J., Harper, J., Palmer, P., & Arndt, S. (1995). Course of Smith, B., Chung, M. C., & Vostanis, P. (1994). The path to
behavioral change in autism : A retrospective study of high- care in autism : Is it better now ? Journal of Autism and
IQ adolescents and adults. Journal of the American Academy Developmental Disorders, 24, 551–564.
of Child and Adolescent Psychiatry, 35, 523–529. Stehli, A. (1992). The sound of a miracle ; a child’s triumph over
Powell, S., & Jordan, R. (Eds.) (1997). Autism and learning : autism. London : Fourth Estate Publications.
A guide to good practice. London : David Fulton. Sturmey, P. (1995). Analog baselines : A critical review of the
Prizant, B., & Schuler, A. (1987). Facilitating communication : methodology. Research in Developmental Disabilities, 16,
Language approaches. In D. Cohen & A. Donnellan (Eds.), 269–284.
Handbook of Autism and Pervasive Developmental Disorders Sturmey, P. (1996). Functional analysis in clinical psychology.
(pp. 316–332). New York : Wiley. Chichester, U.K. : Wiley.
 PSYCHOLOGICAL AND EDUCATIONAL TREATMENTS FOR AUTISM 321

Swettenham, J. (1995). Can children with autism be taught to Williams, D. (1994). Somebody somewhere. London : Corgi
understand false beliefs using computers ? Journal of Child Books.
Psychology and Psychiatry, 37, 157–166. Williams, T. I. (1989). A social skills group for autistic children.
Szatmari, P., Bartolucci, G., & Bremner, R. S. (1989). As- Journal of Autism and Developmental Disorders, 19, 143–156.
perger’s syndrome and autism : A comparison of early history Wing, L. (1981). Asperger’s syndrome : A clinical account.
and outcome. Developmental Medicine and Child Neurology, Psychological Medicine, 11, 115–129.
31, 709–720. Wing, L. (1996). Autistic spectrum disorders. British Medical
Trevarthen, C., Aitken, K., Papoudi, D., & Roberts, J. M. Journal, 312, 327–328.
(1996). Children with autism. Diagnosis and interventions to Wing, L., & Gould, J. (1979). Severe impairments of social
meet their needs. London : Jessica Kingsley. interaction and associated abnormalities in children :
Volkmar, F., Stier, D., & Cohen, D. (1985). Age of recognition epidemiology and classification. Journal of Autism and
of pervasive developmental disorders. American Journal of Developmental Disorders, 9, 11–29.
Psychiatry, 142, 1450–1452. Wolfberg, P. J., & Schuler, A. L. (1993). Integrated play groups :
Walker, M. (1980). The Makaton vocabulary (Revised Edn.). A model for promoting the social and cognitive dimensions
Camberley, U.K. : The Makaton Vocabulary Development of play. Journal of Autism and Developmental Disorders, 23,
Project. 1–23.
Welch, M. (1988). Holding time. London : Century Hutchinson.
Williams, D. (1992). Nobody nowhere. London : Corgi Books. Accepted manuscript received 12 May 1997

Appendix
Cognitive and Language Tests for Use with Children with Autism

This is not meant as a comprehensive list, but includes to have good predictive validity (as long as child is able to
tests that have been found useful in follow-up and research complete several different tasks). Requires little or no verbal
studies. The author and publisher of the tests are given in understanding ; materials tend to maintain children’s interest
parentheses. and hence useful for nonverbal children. Provides IQ, Mental
The tests listed (with the exception of the Play Test) provide Age, and Percentile scores.
standardised scores as well as age equivalents since these are
necessary for any comparative studies. However, there are also Ravens Progressive Matrices (Raven ; NFER-Nelson : Windsor)
many other scales available that can be used to inform clinicians Age range : Child to older adults
about the child’s functioning in different areas. Normative samples : American ; British and many others
Type : Test of perceptual ability. Requires little or no verbal
ability}explanation, but tasks themselves difficult for most very
Tests of General Cognitive Ability young and}or delayed children. Colour form usually best for
young children and Board form (in which pieces attached with
Bayley Scales of Infant Development (Bayley, 1993 ; The
velcro) can be particularly useful in this group. Provides Age
Psychological Corporation : Sidcup, U.K.)
Equivalent scores.
Age range : 1–42 months
Normative sample : American Vineland Adaptive Behavior Scales (Sparrow, Balla, &
Type : Mainly nonverbal tests ; useful for very young or very Cicchetti, 1984 ; Circle Pines, MN : American Guidance
delayed children. Provides IQ and Mental Age scores. Service).
Kaufman Assessment Battery for Children (Kaufman, 1983 ; Age range : 0–19 years
NFER-Nelson : Windsor, U.K.) Normative sample : American
Age range : 30 months–12±5 years Type : Informant based scale ; assesses motor, social, com-
Normative sample : American munication, and daily living skills. Also contains items on
Type : Mixture of verbal and nonverbal tests. Assesses sequen- maladaptive behaviour. Useful if co-operation on other tests
tial processing, simultaneous processing, and achievement. cannot be obtained. However, cultural differences can lead to
Supplies helpful profile of skills and difficulties, which can then some problems when used with non-American subjects. Cor-
be used for designing educational programmes. Provides relations with other measures of IQ are only modest in Learning
Percentile Ranks, Age Equivalent, and Standard scores. Disabled samples. Provides Standard Scores and Age Equiv-
alents for functioning in the different domains.
Leiter International Performance Scale Battery (Leiter, 1948
& 1979 ; NFER-Nelson : Windsor, U.K.) Wechsler Intelligence Scale for Children–Third UK Edition
Age range : 2–18 years (WISC-III UK) (Wechsler, 1992 ; The Psychological Corpor-
Normative samples : European and American ation : Sidcup, U.K.)
Type : Nonverbal tests ; largely involving matching and se- Age range : 6–16 years
quencing. Developed specifically for children with hearing} Normative samples : U.K. (but many other editions available)
linguistic problems ; however level of conceptual understanding Type : Verbal and nonverbal tests, almost all requiring some
required can present difficulties for young autistic children. verbal comprehension. Supplies helpful profile of skills and
Works best with older}more able group. Provides IQ and difficulties. Assesses Performance, Verbal, and Full Scale IQ.
Mental Age scores. Younger children tend to cope better with the Performance
Scale, but older, more able subjects may do better on the Verbal
Merrill-Palmer Pre-School Performance Scale (Stutsman, Scale. Provides IQ, Age Equivalent, and Standard Scores.
1931 ; NFER-Nelson : Windsor, U.K.)
Age range : 18 months–6 years Wechsler Pre-school and Primary Scale of Intelligence : Revised
Normative sample : American UK Edition (WPPSI-R UK) (Wechsler, 1990 ; The Psycho-
Type : Verbal and nonverbal tests ; credit can be given for logical Corporation : Sidcup, U.K.)
omitted items. Despite having very old norms has been shown Age range : 3–7 years
322 PATRICIA HOWLIN

Normative samples : U.K. (but many other editions available) very young or very delayed children. Wide age range is useful,
Type : Verbal and nonverbal tests mostly requiring some verbal although some items very ‘‘ American ’’. Like BPVS, nature of
comprehension. Assesses Performance, Verbal, and Full Scale task very circumscribed. Provides Age Equivalent, Standard
IQ. Experience suggests that only the most able autistic children Scores and Percentiles.
in this age group can cope well with the materials. (Merrill
Palmer is often more viable option). Provides IQ, Age Reynell Developmental Language Scales : Second Revision
Equivalent, and Standard scores. (Reynell & Huntley, 1985 ; NFER-Nelson : Windsor, U.K.)
Age range : 1–7 years
Normative sample : British
Tests of Language and Play Type : Scored from observations of child and structured tasks
with toys and pictures. Even quite young children seem to enjoy
British Picture Vocabulary Scale (Dunn, Dunn, Whetton, & the materials. Low age ceiling the main drawback. Provides Age
Pintilie, 1982 ; NFER-Nelson : Windsor, U.K.) Equivalent and Standard Scores for Comprehension and
Age range : 2±5–18 years Expression.
Normative sample : British (but based on American Peabody
Picture Vocabulary Test) Symbolic Play Test : Second Edition (Lowe & Costello, 1988 ;
Type : Assesses verbal comprehension. Child is required to NFER-Nelson : Windsor, U.K.).
identify one picture from a set of four. Not very useful} Age range : 1–3 years
interesting for very young or very linguistically delayed children. Normative sample : British
In older, more verbal group, the wide age range is valuable and Type : Scored from observations of child and structured tasks
a short version is also very quick and easy to administer. with toys. Enjoyed by quite young children but low age ceiling
However, because of very circumscribed nature of task, may not the main drawback. Provides only Age Equivalent scores.
provide very accurate assessment of linguistic functioning
within the wider social context. Provides Age Equivalent, Test of Reception of Grammar (TROG ; Bishop, 1989 ; MRC
Standard Scores, and Percentiles. Applied Psychology Unit : Cambridge)
Age range : 3–11 years
Expressive One Word Vocabulary Test (Gardner, 1982 ; Western Normative sample : British
Psychological Services : Los Angeles, CA) Type : Assesses grammatical understanding. Child selects one
Age range : 1±1–12 years. Upper extension : to 18 years picture from set of three. Useful for more verbal children ;
Normative sample : American one of few well standardised tests of grammatical ability, al-
Type : Assesses expressive vocabulary. Child is required to though again skills tested rather circumscribed. Provides Age
name individual pictures. Again, not very useful}interesting for Equivalent, Standard Scores, and Percentiles.
This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy.
Users should refer to the original published version of the material.