Firstly, and at a relatively academic/theoretical level, there is reference to the nature of attentional deficits in children with autistic spectrum disorders, and the areas of neurological dysfunction that may underlie the deficits.

Secondly, there is reference to the matter of how to determine school placement, and to the most appropriate management styles within the school setting by which to intervene over particular (behavioural) symptoms.

M.J.Connor June 1999

A number of brain regions including the cerebellum and medial temporal lobes have been implicated in previous research as possible regions of abnormality linked to autism. It has also been noted that a dysfunction in one area of the brain will have a negative impact upon the development and functioning of related brain areas.

A review of evidence completed by these present authors has demonstrated a link between the medial temporal lobe and related structures of the limbic system. For example a study by Barth et al (1995) demonstrated that lower-functioning children with autism showed impairment on a visual recognition and memory task which depended upon medial temporal lobe functioning. Meanwhile, it is recognised that children with autism have particular difficulty with the processing of social and emotional stimuli, such as recognising emotion or the significance of facial expressions in other people, or the formation of a theory of mind, and problems of this social cognition type indicate that part of a temporal lobe (amygdala, hippocampus and entorhinal cortex ) and the orbital frontal cortex, which make up the limbic system, are significant for the typical impairments observed in autism.

A further source of evidence has involved lesion studies where, for example, Bachevalier (1994) demonstrated that monkeys with lesions of the hippocampus or amygdala, made early in life, showed long-term and severe cognitive and social impairments as well as stereotyped and self-stimulatory behaviours.

Also, there is evidence from post mortem studies in which individuals diagnosed as autistic have been found to have abnormalities, such as reduced neuronal cell size or increased density of cells, in brain areas comprising the limbic system.

Other studies have argued that autism is a matter of disorder of higher cortical functions, located within the prefrontal cortex. For example, core characteristics of autism are linked to impairments in executive functioning, such as working memory; and supportive evidence emerges from studies of higher-functioning individuals with autism who are impaired on memory tasks, such as imitating a motor activity, which, in turn, is found to co-exist with specific autistic features such as an inability to recognise the mental states of other people. This has led to a developmental view of autism in which a primary impairment in motor imitation disrupts social-emotional development.

In this study, Dawson et al (opp.cit) examined the performance of autistic children on two neuro- psychological tasks.... the first known to be linked to limbic functioning and the other to the prefrontal cortex and their relationship to the degree of impairment in those areas typical of early symptoms of autism, including orienting to social stimuli, motor imitation, response to emotional stimuli, and shared attention.

The participating subjects included a group of autistic children, and control groups of children with Down’s Syndrome, and a normally developing group. The groups were matched in terms of receptive language age, and the autistic and Down's groups were matched for actual age (around 64 months [with typically developing children having a mean age around 30 months]) and verbal IQ.

The tasks involved:

i. Delayed non-matching to sample ( where the children were to learn to select one of two stimulus items that did not match a given model item).

ii. Delayed response (where the children were to look for a toy placed in a box which was then screened and paired with a similar box placed either to its left or right. The location of the box containing the toy was reversed after 2 correct searches, and trials continued until the child correctly switched the search or until 14 searches had been administered).

Meanwhile, the children were assessed in terms of core autistic symptoms such as social orienting, motor imitation, shared attention, response to distress in others, and symbolic play.

In respect of "typical" autistic behaviours, the authors adopted the subgrouping proposed by Wing and Gould (1979) viz

"Aloof' - a tendency to ignore or withdraw from other children.

"Passive" - responsive if approached and willing to remain socially engaged, but only while the other child maintains the interaction.

"Active but odd" - may seek interaction with other children, but in an awkward or overly-persistent manner.

In the neuropsychological tasks, rule learning (the non-matching) was achieved by all the typically developing children, by 73% of the Down's group, but only 61% of the autistic group.

On the delayed response task, the autistic group was again significantly impaired compared to both control groups.

In respect of the social behaviours, the autistic children performed more poorly than members of the other groups in all areas... social orienting, symbolic play, etc.. However, strong correlations were found only between the rule learning performance and severity of autistic symptoms (or the most socially impaired classification on the Wing/Gould scale).

In their discussion, the authors highlight the impairment among children with autism on these tasks which are linked to the medial temporal lobe and to limbic structures, and which tap the prefrontal cortex; and they stressed the finding that degree of impairment in social-emotional domains was strongly linked to performance on rule-learning, but not consistently so with a delayed response task. Meanwhile, they also noted that a significant relationship between delayed response and severity of autistic symptoms was found only in respect of one type of task, namely immediate imitation.

They suggest that the severity of early behavioural impairment in autism is linked to functioning of the medial temporal lobe and the limbic structures, particularly the prefrontal cortex. They go on to hypothesise that the correct functioning of these systems is vital for certain developmental processes such as face recognition, emotional recognition, etc.; and a failure in respect of these skills will be linked, in turn, with a failure to develop joint attention, imitation skills, and so on. If there is a failure to participate in social imitative interactions there is likely to be a problem in the development of other social skills. In other words, the conclusion is that the function and integrity of these neurological structures are critical for the early development of social perception and cognition, and observed deficits will help to explain the core symptoms of autism that are apparent by early childhood.

Such characteristics underlie theories that suggest disordered attention to be the core deficit in autism. For example, one theory has it that changed arousal mechanisms and an unusual response to novelty are at the heart of the behavioural characteristics of autism.

The study by Pascualvaca et al (1998) set out to assess the attentional skills of children on a range of measures originally used by Mirsky (1987). It had been found that different tests of attention focused upon different and unique aspects of attention such as the ability to focus upon a particular stimulus, or to sustain this focus over an extended period, and to shift attention flexibly. It has also been found that these separate processes are linked to distinct neurological locations.

The present study, therefore, set out to examine this issue by using two computer tasks which involved two different aspects of shifting attention. In the first case, the children must decide whether some aspect of three different stimuli objects was the same or different so that they have to switch their attention constantly among the features of the different items. In the second case, they must shift their attention after focusing on a particular target for some while with the prediction that this would be particularly difficult for an autistic sample. Reference was also to be made to the possibility that any difficulty in attention might be secondary to lowered motivation since tangible rewards have been found to facilitate learning and attention in autistic children.... but the suggestion was that autistic children with a significant arid core deficit in attention would not be influenced by the use of any kind of reinforcement by which to increase motivation.

The sample of boys and girls with autism ranged in age from six to twelve years and their level of functioning was towards the upper end of the spectrum as measured by a cognitive scales. Two control groups were used, matched by skill level. In the one case, the match was in verbal mental age, and the other in performance mental age.

The various tasks included the following:

A continuous performance test where children see individual letters on the computer screen and must respond as quickly as possible whenever they see a particular letter. In the auditory version, the children listen through earphones and have to respond to the letter 0 when preceded by an L. The digit cancellation task involves looking at an array of numbers and crossing out the numbers 3 and 7 whenever they appear, as quickly as possible. In a card sorting test, the children have to match series of cards according to various criteria, such as colour, form, and number.

The matching tasks included one test where the children see four pictures on a screen and have to match the one which appears in the top half to one of the three appearing in the bottom half. Matching principles could include colour, size, and shape. The other task presented the child with three stimuli pictures and the child must press a key if the three items are the same and another key if they are different. The level of difficulty could be altered in terms of how many features were the same or different.

In respect of focused attention, the children with autism did not differ from either a control group on any of the measures derived in that they took the same time to look at the items and they made similar numbers of errors.

In respect of sustained attention, it was also found that the autistic sample did not differ from controls in the number of correct responses, or mistakes, or reaction time in the continuous performance task. All children performed better in the visual version compared to the auditory version. However, an examination of errors of commission showed that the autistic children made more errors in terms of responding to a non-target letter which suggested a heightened degree of impulsivity.

In the tests of shifting attention, no differences were found in the same-different computer task. (This appears to be in contrast to other research findings, involving card sorting activity, where the autistic children completed fewer categories and made more errors than children in either of the control groups. The children had particular difficulty in a shifting strategies.)

The summary and conclusion indicated that the autistic children can be differentiated from normally-developing children in terms of unusual attentional style (the disparity, for example, between focusing upon minute details or repetitive actions and a failure to respond to approaches by other people or to cues around them) and that differences in performance on particular tests of attention can be identified.

The significant finding in this study was that children with autism were able to shift their attention in a setting where continuous shifts were necessary, but they had great difficulty in shifting attention once engaged in a particular activity. In other words, the problems that influence the children with autism reflect a difficulty in disengaging or in re-establishing a focus of attention.

Some of the inconsistencies in performance in different types of task (computerised tasks in this study - face to face settings in other studies, such as those of Courchesne [e.g. 1995]) might be attributable to the issue of receptivity to feedback from the examiners who administered the tasks... i.e. a failure with the social aspects of the setting. It is possible that the human examiner, unlike a computer, provides additional albeit subtle cues, such as facial expression or gestures, which the autistic child does not perceive or does not interpret appropriately. Therefore, differences in performance between autistic and non-autistic children may be the greater in tasks organised by another person ; but differences are less evident when the tasks are presented via computer. Alternatively, the autistic child may simply be less susceptible to, or influenced by, social reinforcement and feedback.

In any event, the final point was a restatement of the lack of support for a generalised or core deficit in shifting attention among children with autism; rather, it is a matter of specific elements of the task or setting which will give rise to impairments in attentional shifts. If the children can shift their focus of attention when given additional time, then the observed deficit is not in shifting per se, but is secondary to difficulties in co-ordinating attentional resources as well as in the lesser activating effects of motivation.

A number of explanations have been put forward for the way in which these attentional problems can contribute to the social impairments that are at the heart of autism. For example, the ability to share attention with others, which is impaired in autism, requires the child rapidly to shift his attention between different stimuli. Further, it has been proposed that the impairments are more evident for social stimuli, and that, because social stimuli, such as facial expressions, are complex and unpredictable, children with autism will have difficulty processing and understanding them so that their attention is not naturally drawn. This lack of attention to social cues will limit the child's opportunity to take part in early social experiences, and social development will remain limited.

However, this present study recognised that little evidence has been gained from observing young children with autism in natural settings, and set out to examine the autistic child’s ability to orient to social and non- social stimuli delivered in a naturalistic manner. The major hypothesis was that, compared to control children with Down's Syndrome and typically developing children, the autistic children would selectively fail to respond to social stimuli as compared to non-social stimuli.

The various samples of children used in this study were the same as those described in the first section of these notes, but the procedure was such that each child was individually tested while seated at a table with a familiar adult. The orienting task involved a second adult who delivered four types of stimuli. Social stimuli consisted of either clapping hands three times or calling the child's name three times. The non- social stimuli consisted of either playing a musical box for six seconds or shaking a rattle for six seconds.

There were also two types of shared attention tasks in which the adult either gazed at an object or pointed to an object, waited until the child seemed to be losing interest, and then sought to gain the child's attention by taking the object and holding it near the child’s face. Once the child's attention had been gained, the original object was removed from sight and one of four shared-attention cues was delivered. These consisted of pointing to yellow crosses attached to the wall either in front of the child or behind the child, or conspicuously looking at the cross in front of the child or behind the child.

Research assistants coded the child's performance in terms of errors (defined as failure to turn the eyes towards the stimulus within a I5 second period).

The results indicated that autistic children more frequently failed to orient to any stimuli, but the failure was more marked for social stimuli. For the children in the control groups, the number of errors for both social and non-social stimuli was virtually zero. For the children with autism, too, the mean number of errors made to non- social stimuli was approaching zero. However, the autistics failed to orient to the social stimuli on around fifty per cent of the trials. Even in the case of subjects who did orient to both social and non-social cues, it was found that the autistic group showed greater delay in responses in general and that the delays were greater to the social stimuli.

In respect of a shared attention, the children with autism made significantly more errors than children in either of the two control groups.

The summary and conclusion highlighted the finding that children with autism show a general impairment in orienting ability, and that the impairment is more severe for social stimuli.

The authors, therefore, claimed support for their hypothesis that children with autism are particularly impaired in their ability to orient to social stimuli.

A clear correlation was found to exist between shared attention performance and the ability to respond to social stimuli, but there was found to be no correlation between shared attention and response to non-social stimuli.

This pattern of results was taken as support for the hypothesis that impairment in shared attention among children with autism may be the result, at least partially, of a more basic failure selectively to attend to social stimuli such as another person's eyes or facial expressions. Shared attention skills, such as shifting the gaze between a toy and another person's face, will require that the child is interested in attending to another person.

One implication of this impairment of social orienting among autistic children is the need to target basic social attention skills early in a programme of intervention. If a child with autism fails to recognise that social stimuli can be interesting and rewarding, that child may fail to acquire more complex communicative or social skills that depend upon paying attention to other people.

An impairment in social orienting may represent one of the earliest symptoms of autism, which predates shared attention skills (and one notes the reference to an earlier study involving first birthday parties in which failure to orient to names was a clearly observable behaviour that distinguished young children with autism from normally-developing peers)

The work of Simpson and Myles (1998) highlights this point in referring to the misunderstanding surrounding Asperger Syndrome and to the tendency for interventions to focus upon seeking to change the individual rather than acknowledging the nature of the disabilities linked to the syndrome, and modifying demands or circumstances accordingly.

For example, they argue that children and young people diagnosed with this syndrome are not inherently prone to aggression or violence. However, the social difficulties may often lead to anxiety or frustration or stress which, in turn, can lead to behaviour which appears aggressive. Converging evidence has shown that what might seem to be aggressive behaviour is actually motivated by a need to escape from a stressful situation.

Further, the current authors remind us that the social interactions of young people who have Asperger Syndrome are noted for their lack of reciprocity, and play may be individual and unusual. As a result, the person may appear odd or even provocative, and other young people may be irritated by some of these behaviours, such as an unwillingness to take turns, a misunderstanding of social rules, agitation if his/her space is encroached upon, and the apparent self-centredness.

As a result, a kind of vicious circle is set up whereby the young person becomes increasingly set apart, continues to have difficulty in expressing how (s)he feels, and develops greater feelings of stress linked to what appears to be a lack of control over uncomfortable situations, until a crisis point is reached.

The feelings of stress and loss of control may result in behaviours which appear aggressive but which, in fact, reflect the difficulty in functioning in a world seen to be unpredictable. This being so, the strategies to deal with episodes of aggressive behaviour among children or young people with Asperger Syndrome are not likely to be the same as those appropriate for other normally-developing children and young people. In particular, it is recommended that one should not rely upon the use of punishment or sanctions imposed upon the individuals concerned; rather, it should be a matter of involving the individuals in monitoring their own behaviour and selecting appropriate options.

Prevention of problem behaviour is the primary goal and the most appropriate school structure will involve clear instructions for what constitutes appropriate social behaviour, models of acceptable behaviour, opportunities to plan and rehearse acceptable behaviour, and feedback describing both good and poor performance.

Given the preference for predictability and consistency, one would usefully provide the young people concerned with clear routines together with advance warning of any changes to routine. Meanwhile in order to minimise the probability of teasing or bullying, it would be helpful to ensure that peers are given some insight into the nature of the syndrome so that the particular disabilities are recognised and understood.

Simpson and Myles (opp.cit) go on to describe a number of behavioural options which can be supportive. For example, they refer to cognitive behaviour modification which involves teaching the individual to monitor behaviour and to deliver self reinforcement. The goal is the shifting of the focus of control for managing behaviour from teachers or other adults to the young person him or herself, and they cite the case of a teenager who found the noise and "atmosphere" of the changing room before and after PE to be threatening. He was trained to go to a quiet area of the changing room or gymnasium and to do breathing exercises for a couple of minutes and then to select a shower and changing area which were relatively quiet.

Behaviour contracts involving the young people and teachers and parents may be helpful in specifying which behaviours are inappropriate and specifying, also, the consequences if those behaviours continue.

Nevertheless, the emphasis is put upon positive outcomes so that, for example, a contract for an Asperger individual might state that he may choose daily from a list of reinforces when there is no threatening behaviour towards peers.

Mention has been made elsewhere of the usefulness of social stories which prescribe what to do in various social situations; and social scripts may also be used to highlight desired behaviour so that a student may remind himself about day-to-day events in school thus to reduce unpredictability, and about appropriate behaviours in given circumstances.

A further technique described is the social autopsy which seeks to analyse aggressive incidents and which may help young people to develop an understanding of their social mistakes with a view to avoiding their repetition. To some extent, it is a matter of teaching theory of mind in gaining for the individual some insight into how certain behaviours may be perceived and interpreted by peers.

There is reported to be a risk of a kind of power struggle developing between adults who need to manage individuals with Asperger Syndrome and the individuals themselves. Again, the implication is to intervene at an early stage before problems escalate, and to focus upon desirable behaviour without too much initial discussion of the social consequences of the undesirable behaviours.

There is a reminder, too, of the need to use clear and unambiguous language in giving directions or advice.

Finally there is an underlining of the fact that children and young people with this syndrome may be unable to understand or to demonstrate their own emotions so that an appearance of particular detachment or calm may actually reflect stress; and a major goal for teachers or parents is to recognise when the stress is being generated thus to alter the situation before the stress becomes critical.

In respect of "connections", there are problems with attention in being over-focused upon detail thus failing to have an overview; with sensory integration in being able to use only one channel at a time such as listening but not seeing; and with information processing in being unaware of the context of any given data, or being unable to recall information if the circumstances are not identical to those at the time of the original learning.

Problems with "tolerance" include a sensory and emotional over-sensitivity, such that feelings are sometimes overwhelming even if they cannot be expressed or even understood, and a fear of other people or of being found unable to know what to do.

In respect of "control", there may be a focus upon a particular interest at the exclusion of all others, an acute anxiety (albeit irrational) about a range of stimuli, and a possibility that other people's words or actions can spark off some unintended action.

Nevertheless, there are very marked differences between individuals diagnosed as autistic and the triad of impediments can be demonstrated in a range of ways. For example, there may be considerable variation in verbal skill or in the ability to learn visually.

On the other hand, it is likely that all people with an autistic spectrum disorder will have difficulty in using pragmatic language or with subtle or idiomatic language (and the general emphasis on verbal explanations in senior schooling may not suit many autistic students); but they may all have relative strengths in acquiring and using information that is not context bound, such as rote learning of facts, reading accuracy, spelling, and computation.

The net effect of all this may be a genuine misunderstanding of whole-class instructions and a tendency to work in an individual way, which might be misinterpreted as a lack of co-operation. Annoyance or anger on the part of the teacher(s) may only serve to heighten the stress in the student who may be very sensitive to signs of emotion in others even if (s)he cannot readily recognise, understand, or name the emotions.

It may even be that students with autism do not see themselves as a part of the class and not, therefore, as being included in class instructions; and it is noted by Jordan and Powell (1995) that they may not remember their experiences in completing earlier assignments or activities such that there is a dependence upon external prompts when a similar task is presented. Some students might be helped by the use of cards which contain statements describing feelings or needs (such as "I don't understand" or "This is difficult") and which can be selected when the student is struggling but cannot put the feelings into words.

In any event, confrontations will be pointless given the probability that the autistic student has no automatic wish to please, and may not recognise or understand authority as an ordinary member of the class would.

Rules and routines need to be made very clear and specific, and the individual may need help in gaining some awareness of the effect of actions upon others, such as repetitive questioning.

Alternatively, one might deal with such issues simply by setting up a rule involving some time limit after which the subject is to be closed. In respect of positive reactions to good behaviour and progress, it needs to be recognised that tangible reinforcers may be more effective than social reinforcers such as a smile or a positive comment.

There will be problems in incidental learning or in recognising the many different cues and sources of information in social interactions in a classroom. Failure to understand another person's point of view or intention may lead to difficulties in recognising social rules, or being unaware of the need to adjust behaviour according to the particular circumstances.

Problems may be very marked in the unstructured periods during break and lunchtimes, and it may be appropriate to provide access for the autistic individual to specific activities or clubs especially if this provides the opportunity to interact with non-autistic peers. Alternatively, such periods could be used for social skill work involving small groups especially if participating non-autistic peers are given some information about autism and autistic behaviours.

At the very least, discrete supervision will be necessary during these times, perhaps linked to some specific teaching of the individual about how to gain access to playground games or groups, and how to maintain satisfactory peer relationships (and the encouraging of self monitoring to ensure that tensions do not arise if (s)he is not always or quickly successful). The difficulties of meaningful integration and progress in the secondary school are not to be underestimated, but there is converging evidence that autistic individuals are more likely to make satisfactory "connections" in the mainstream setting where there is access to positive peer modelling, as long as teacher and peer understanding and expectations of the autistic individuals are informed and realistic.

In particular, the significance of what might appear to be aloofness or aggressiveness (but actually indicates a continuing vulnerability) should be understood, alongside an awareness that the autistic student may experience considerable anxiety in a setting in which there is a lack of control over, or predictability of, events; or a feeling of stress whose source (s)he may not easily locate or even recognise, and almost certainly fail to express in words.

The reflections on how best to arrange schooling for children with autistic spectrum disorders, set out by Alderson and Goodey (1999), begin with a reference to the questionable coherence of definitions of Autism. Reference is made to the triad of impairments but Frith (1989) quoted as highlighting the ease with which autism may be confused with other conditions. Accordingly, she emphasises only two core features:

1. Autistic Aloneness ..... the intangible difference of autistic children, pervading all sorts of behaviour.

2. Obsessive Insistence on Sameness including repetitiveness, single-mindedness, rigidity, and the inability to perceive subtle differences.

Asperger Syndrome is included within autistic spectrum disorders, and this is characterised by borderline or normal IQ, social isolation or inappropriate social style, a focused interest in only one or two areas, a narrow lifestyle, limited or odd verbal and non-verbal communication, and poor motor co-ordination. It needs to be recognised that so many characteristics have been linked to the autistic spectrum that no one person is likely to have all of them, hence some difficulty in precisely defining autism or in identifying children so-affected.

If it is the case that there is a clear correlation between the level of stimulation in the environment and the level of difficulty faced by the autistic child in coping, then the question is raised about when adults should allow for a child's difficulties, or try to compensate for them or overcome them. This has implications for school placement, and Alderson and Goodey (opp. cit) pose the question whether a more controlled or narrow diet will reinforce or even establish a narrow and rigid style in the children.

A further implied question concerns the actual strengths and weaknesses in a given child compared to the observable strengths and weaknesses in a school setting where the teaching and learning styles and goals are themselves relatively narrow and inflexible.

In respect of specialist schooling, the authors argue that, if the children have particular difficulty with communication and social interaction, then it is of questionable logic to put them all together in one setting and away from articulate and socially skilled children at school, especially if this undermines their opportunities to make friends in their home area and if expectations upon them are lowered. They quote a headteacher of a specialist unit as commenting that "coping socially is the hardest part while lessons are the easy part", and suggest that, if indeed lessons are the easy part, there seems even less purpose in specialist and separate schooling.

The conclusion, based upon surveying the operation of a range of school and unit settings, has it that autistic tendencies of isolation and self-absorption are more evident in the specialist settings where these characteristics appear to be projected upon the children. Even if one might have liked to see some discussion about the range of levels of severity of observed symptoms, and the co-existence of challenging behaviour and/or language impairment; it is significant to note the Alderson and Goodey view that "comparative evidence of inclusive and special ASD schooling seriously challenges assumptions about the advantages of special schooling".

M. J Connor June 1999

Bachevalier J. 1994 Medial temporal lobe structures and autism. Neuropsychologia 32 627-648

Barth C., Fein D., and Waterhouse L. 1995 Delayed matching to sample performance in autistic children. Developmental Neuropsychology 11 53-69

Courchesne E. 1995 Infantile Autism : part 1. International Pediatrics 10 141-154

Dawson G., Meltzoff A., Osterling J., and Rinaldi J. 1998 Neuropsychological correlates of early symptoms of autism. Child Development 69(5) 1276-1285

Dawson G., Meltzoff A., Osterling J., Rinaldi J., and Brown E. 1998ii Children with autism fail to orient to naturally occurring social stimuli. Journal of Autism and developmental Disorders 28(6) 479-485

Frith U. 1989 Autism : Explaining the Enigma. Oxford: Blackwell

Harrison J. 1998 Improving learning opportunities in mainstream secondary schools and colleges for students on the autistic spectrum. British Journal of Special Education 25(4) 179-183

Jordan R. and Powell S. 1995 Understanding and Teaching Children with Autism. Chichester: Wiley

Mirsky A. 1987 Behavioural and psychophysiological markers of disordered attention. Environmental Health Perspective 74 191-199

Pascualvaca D., Fantie B., Papageorgiou M., and Mirsky A. 1998 Attentional capacities in children with autism. Journal of Autism and Developmental Disorders 28(6) 467-478

Simpson R. and Myles B. 1998 Aggression among children and youth who have Asperger Syndrome. Preventing School Failure 42(4) 149-153

Wing L. and Gould J. 1979 Severe impairment of social interaction and associated abnormalities in children: epidemiology and classification. Journal of Autism and Developmental Disorders 9 11-29