University of Edinburgh

Educational Implications of Autism and Visual Impairment

Presented by Rita Jordan
Senior Lecturer Autism, School of Education, University of Birmingham

Autism and Visual Impairment Conference, Edinburgh, 27 March 1996

1 Introduction

There is nothing in either autism or visual impairment that protects the individual from the co-occurrence of the two disorders and, indeed, there are some conditions (such as Rubella in pregnancy) that predispose the infant to both. Thus, there are likely to be children who have both visual impairment and autism. This paper will consider the educational implications of the dual condition, but first it will establish the importance, for educational practice, of recognising both conditions. From this, it will show how similar behaviours may, nevertheless, have different causes, consequently different developmental paths, and require some differences in educational approach. The paper concludes with the argument that, in spite of these differences, there are also similarities and suggests some educational strategies that will be appropriate to the dual condition. I will assume, I am sure rightly, that my audience today are better acquainted with the educational implications of visual impairment than I, and so I will concentrate on autism, bringing in visual impairment only to make a point in relation to autism or to illustrate the dual disability.

2 Educational Implications of a Diagnostic Category

Visual impairment may be suspected from the behaviour of an infant, but it is diagnosed at the biological level; its presence, and even its extent, can be measured with some precision. Autism, on the other hand, is a psychiatric condition, diagnosed at the behavioural level according to certain criteria; our current understanding is that, the aetiology at the biological level is diverse and so the behavioural criteria are likely to remain as the source of diagnosis. Nevertheless, autism cannot be identified by any single behaviour or even a combination of them; there are no behaviours that are both specific to autism and characteristic of all those within the autistic spectrum of disorders. The behavioural criteria are descriptive of categories of behaviour, rather than specific behaviours and autism is only indicated when the full triad of impairments is present. In that sense, it makes no sense to talk of 'autistic features' or 'autistic tendencies' since that assumes that certain behaviours or features are of themselves 'autistic', whereas that is not the case. There are no specific behaviours in autism that cannot be found in the normally developing population at some time or other and in other clinical groups such as those with visual impairment. What makes it autism, is the combination of the three kinds of impairment and the underlying cause for this triad, found at the level of brain functioning, rather than behaviour: that is, the psychological level.

In this analysis, then, autism is a kind of theory, that there exists a common pathway to autistic spectrum disorders, at this psychological level, that accounts for the characteristic pattern of impairments. It is my contention, argued in more depth elsewhere (Powell and Jordan, I993a; Jordan and Powell, 1995), that it is important for educationists to understand autism at this level, if teaching is to be effective. Stuart Powell and I have argued that people with autism have a distinct way of thinking and learning (Powell and Jordan, I993b) and working on behaviour alone is unlikely to lead to any recognition, let alone understanding, of this and is, therefore, also likely to result in misinterpretation of that behaviour, according to our understandings based on our own way of thinking and what we have learnt of normal development. In our anxiety not to label and categorise, we may also deny difference and so further disadvantage those who do not fit our template of the world. Just as we recognise that the blind child is not the same as the sighted child minus sight (in other words, that blindness affects all levels of functioning), so we should recognise that autism also has profound effects on thinking and learning that both go beyond, and account for, the behavioural characteristics by which it is identified.

Of course, it is also important to realise that no child is defined by a disability, however profound, and special needs do not just arise from a disability either. Just as the visually impaired child will accommodate to that disability, so does the child with autism. What one is dealing with, then, at any one time, is a product of the original disability, the accommodation to it, the compensatory strategies the child has learnt or been taught, that particular child's strengths and weaknesses, the child's history, the learning environment, and the interaction of all these factors, including the personality of the child. Where the child has a dual disability, there will also be an interaction here so one cannot simply apply educational approaches appropriate to visual impairment, say, and add on those appropriate to autism. In some ways, as we shall see, the approaches may even be incompatible and so a new approach must be found that takes account of all the factors pertinent to the way that child learns. What I am arguing for is neither a completely individual nor a strictly categorical approach. We have to know the individual, perhaps even more so in autism than in other conditions, because autism has such idiosyncratic developmental pathways. But, without some understanding of the kinds of difficulties and differences caused by the autism, we will generally flounder. As Exley (1995) has said, autism is not a label; it is a signpost. Knowing a child has autism will not tell you exactly what to do or how to do it, but, without that knowledge, it is unlikely that any study of the individual's behaviour will lead to the understanding on which effective teaching must be based.

3 Autism and 'Blindisms'

Gense and Gense (1994) have illustrated the many ways in which the behaviour of children with autism and those who have visual impairments is similar. They refer to many of these behaviours found in the visually impaired as 'blindisms', and the very label resonates with autism and the way that some have described the characteristic stereotypic behaviour of children with autism as 'isms' (Kaufman, 1994). Yet it is not the fact that visual impairment and autism may lead to the same kinds of behavioural response that is important of itself (although that clearly matters for differential diagnosis), but the reasons behind this sharing of symptoms'. I will take a few key areas where behaviours tend to coincide or overlap, to illustrate this point.

In writing about autism, it is interesting how often visual impairment is used as a kind of metaphor for the condition. As will be discussed below, Baron-Cohen (I 995) has recently used in the title of his book on autism the term 'Mindblindness' to refer to the autistic condition.

In the section below, I will explore that analysis, but here I just note that autism is often seen as a kind of blindness, albeit directed to mental rather than physical phenomena. Frith (1989) also, in her seminal account of autism, speculates on the advantages that would accrue, in terms of public understanding, if people with autism could carry the equivalent of their own white stick to indicate their disability. She also likens the disturbance of routine in the child with autism to the disorientation felt by someone who is blind when the familiar furniture arrangement is changed. The understanding and skills you bring as teachers of the visually impaired, then, are very useful in understanding the difficulties and the behaviours associated with autism; it just becomes a matter of recognising the kind of 'blindness' that occurs in autism and taking account of that in the same way as one would for a more direct sensory disability. But, one can close one's eyes, or stumble about in the dark and get at least an inkling of a child's visual impairment; one cannot switch off one's understanding of self and others, however, in the same way and so, in spite of our high degree of empathy for one another, we find it just as difficult to get inside and understand the person with autism as they do us.

One behaviour found in autism that is very reminiscent of visual impairment, even when the person has intact vision, is perimeter hugging (especially in large spaces) and needing to touch everything in a room before settling down. Often this is attributed to obsessional ritual behaviour in autism, as if this kind of label was an explanation. Of course, it is true that in autism, as in many other conditions and even in normal development, behaviour can become ritualised and habitual very quickly and the original reason may be lost in the fact that the person now does this in this setting because that is what they do in that setting; the setting itself triggers the behaviour. Yet there is usually a functional base to the behaviour; it is not chance that sets it off in the first place. We can gain an insight into the possible original cause by asking the more able who nevertheless still exhibit this behaviour. Donna Williams, for example, is a very able young woman with autism who has written several books and now lectures on autism (and offers support to families) as a living. Yet going into a new room, especially if it is empty, she will (when she can) go round the room touching everything before she sits down. When asked why she does this, she replies that she has difficulty perceiving herself in relation to her environment until she has done this; it gives her security, not just as a ritual but at a perceptual level, in interpreting her environment.

A characteristic behaviour of children with autism and those with visual impairment is to spin themselves, being able to do so, apparently, without becoming dizzy in the normal way. Where the visual impairment is not total, also, children often share the visual preferences of those with autism for spinning objects, for moving patterns and for peripheral focusing on objects, especially by rapidly moving them within the field of vision (eg, 'flicking' in the corner of the eyes). Here, the common pathway between autism and visual impairment may lie in Russell's (1994) recent suggestion of a basic deficit in autism. In normal development, the nervous system must not just control movements of the head and eyes to scan the environment or keep track of moving objects, but must record those movements, in order to distinguish the case where the movement is in the objects being viewed from the case where the movement is in the person's head or eyes, and the world is still. It is this 'efferent copying' that gives us a sense of our own agency, but which also makes us feel dizzy when we spin too rapidly or for too long, overloading the computational mechanism, which takes time to re- adjust. In visual impairment, this computation is presumably disturbed since the visual information part of the process is disturbed.

In autism, Russell suggests, it may be disturbed at the stage of 'efferent copying' so the person does not have a sense of themselves as an agent (just as C. Frith [1992] has suggested that in schizophrenia this efferent copying breaks down and the person loses their sense of having intended their own actions, and so feels as if they are being controlled by others). A by-product of this is that the person would not get dizzy from spinning and might crave the kinds of sensations that would provide movement stimulation free from having to compute head and eye movements in relation to the world. Such situations are provided by spinning objects, by movement in the visual field (as in fires or running water) and by rapid small movements in peripheral vision.

Other movement disorders may result from neurological abnormalities that often accompany autism and some kinds of visual impairment, or they may be reactions to sensory deprivation. Self stimulatory behaviours such as rocking and rhythmic head banging (and even some forms of self mutilation) are readily induced in experimentally produced sensory deprivation conditions and so can be seen as reactions to the visual loss of stimulation in children with visual impairments. Characteristically, they disappear when the child is stimulated or has developed more productive ways of self stimulation. Yet in autism, they are often more persistent and will be seen even where the child has no apparent sensory loss. Nevertheless, it is helpful to make sense of them in relation to sensory deprivation, even in these cases.

In autism there appears to be a difficulty in abstracting relevant information from background stimulation (Frith, 1989; Grandin, 1995; Jordan and Riding, 1995) and so the result is often the experience of overwhelming 'noise' (in a technical sense that refers to visual as well as auditory sensations) rather than stimulation. Providing 'white noise' (meaningless stimulation) is one way of inducing sensory deprivation either because the direct effect is that the person 'switches off' all sensation (as sometimes happens in autism) or because undifferentiated sensation is not really stimulation at all.

There are also common features in the language development of children with autism and visual impairment (Andersen et al, 1983; Fraiberg, 1979; Hobson, 1993). In particular, two distinctive features of 'autistic language' echolalia and 'reversal' of personal pronouns - are also seen in the language of visually impaired children, although, once again, they may be more readily overcome in such cases than in autism. Echolalia can be seen to arise from a problem in establishing reference to the real world and in treating language as unanalysed chunks. As Fay and Schuler (1980) have shown, in autism, in visual impairment and in normal development, immediate echolalia occurs in situations where the language has not been fully understood or processed. It is not difficult to see how this may arise with the visually impaired, but in autism it may occur because of difficulties with certain key words that have a communicative function and so are incomprehensible (Jordan, 1996). Such words may be j 'show' (where re-phrasing of the instruction replacing 'show' with the more easily understood 'point to' or 'put your hand on' results in compliance instead of the echoing) or the use of personal pronouns and may develop into a strategy for having a conversational turn when the child does not understand the language used or the topic-comment structure of a conversation (Bruner and Feldman, 1993). Delayed echolalia is more problematic, but again may be a general style of learning where actions are not spontaneous but need to be cued by particular aspects of the environment or by the child's own sensations. Thus, the child who shouts out "I hate jacket potatoes" half way up a rock face is responding to his feelings of fear and giving the automatic phrase that has frequently accompanied that sensation, evoked by the appearance of the jacket potato.

The teacher needs to have an understanding of the individual's life history if the phrase's meaning is to be deciphered (and the child helped to express his fear more conventionally and appropriately) and not merely dismissed as 'delayed echolalia' as if this were an explanation rather than a description. Although the phenomenon expressing a difficulty with personal pronouns is commonly referred to as 'pronoun reversal' there is seldom any actual reversal in autism or in visual impairment. A more common phenomenon is that 'you' is understood as referring only to the child and the child may then use 'you' to refer to herself. 'I' is seldom used, although 'my' or 'me' may be used for the other person and the child will also sometimes refer to him or herself by the third person pronoun (ie, as 'she' or 'he').

Quite common is the lack of any personal pronouns and the use of proper names in self or other reference (Jordan, 1989). The evidence is not that the child with autism has any difficulty with understanding the self and other from an objective stance (from the 'outside' as it were) but they have specific difficulty in developing a sense of themselves as social agents (Russell, 1994) and of understanding social roles in conversations.

Without an understanding of such roles, personal pronouns come to be used as labels for people. 'You' is most salient in its role of referring to the child him or herself, as may be 'he' or 'she'. Nor does the child notice (through spontaneous eye direction detection and understanding the communicative intent of others) that others are addressed as 'you' or referred to as 'she' or 'he so this limited understanding of pronouns as labels is not challenged. Normally developing children who go through a stage of 'pronoun reversal' are found to resolve it through noticing these conversational usages and through the self assertiveness that comes from an establishing sense of ego, whereby the child is vigorous in asserting his or her agent status ("I do it!) or possessiveness (mine!). None of this is a feature of autistic development and may take longer to establish in children with visual impairment who must rely on other senses to give the kind of feedback on which such understanding is based.

4 Blindness versus 'Mindblindness'

From the examples of common 'difficulties' or differences in development between autism and visual impairment, given above, it can be seen that the social impairment in autism leads to a kind of blindness to socially meaningful stimulation. This view has been formalised by Baron-Cohen (1995), using an information processing model of the normal development of mind-reading' ability: the ability to understand mental states in oneself and others.

Baron-Cohen proposes four underlying mechanisms by which this comes about. The first of these is called an 'intentionality detector', although this seems to confuse intention with desire, and what is really meant is a 'desire' detector whereby non-random movement is automatically interpreted as goal directed behaviour and desires are imputed to the person being thus goal-directed. Since this is assumed to be functioning in autism and visual impairment, it will not be further discussed here. The next mechanism is referred to as an 'eye direction detector' which computes whether the other is looking at the self or at something else and interprets looking as seeing (presumably in analogy with one's own experience, although Baron-Cohen does not explain how this notion of the self and this self awareness originates). This also is assumed to be intact in autism, although Baron-Cohen does not appear to take account of some of his own research showing that it does not operate spontaneously in autism, but needs to be cued (Leekam et al, 1994). Now this mechanism cannot be operating in visual impairment and so one would expect disturbance in understanding that depended on noting where others were looking, that is, in social interaction and in the language problems to do with pronominal reference and echolalia noted above.

Yet the difficulties experienced by children with autism in these self-same areas, appear to rest with the next mechanism proposed by Baron-Cohen, which is intact in visual impairment and dysfunctional in autism; this is the 'shared attention mechanism'. This allows the individual not just to compute where someone's attention is directed (as in the case of the EDD with visual attention), and where one's own attention is directed but to go further and compute the fact that 'we are both attending to the same thing'. Baron-Cohen sees this as a cognitive process, although 1 would postulate (along with Hobson, 1993) that emotions have a role here in developing this sense of a shared experience. Baron-Cohen proposes that EDD has preferential access to SAM, so that the fact that there is no functioning EDD in visual impairment will mean a delay in the functioning of SAM while information on attention from other senses is computed. This would lead to some of the problems mentioned above in terms of resolving pronominal reference and understanding language as referring to the world and not being an 'object' of imitation in its own right; but it would also account for the fact that these problems are usually overcome more easily in visual impairment than in autism, through the child's understanding of social roles and shared attention gained through auditory experience or tactile stimulation. What is even more remarkable, and illustrates the inborn nature of the mechanism involved, is that congenitally blind children can still understand someone else's focus of visual attention and can perform communicative acts designed to establish joint visual attention, such as holding up objects for inspection.

In autism, on the other hand, the child can tell if someone is looking at them or where they are looking, and can follow the direction of their gaze when instructed to do so. Yet, not only does this not trigger them to look back or to follow eye gaze spontaneously, they do not have any awareness of a sharing of attention and the problems leading from this are thus more profound. Nor is it any easier for them to compute or experience shared attention through auditory channels (knowing that they are hearing the same thing as someone else) or through any other route; their problem is at a conceptual level, and thus applies to all sensory channels.

The final stage in Baron-Cohen's (op cit.) analysis on the 'mind-reading' system is borrowed from Leslie's (1991) 'Theory of Mind' theory and is a 'Theory of Mind Module' through which the person can come to understand mental states such as believing, thinking, and knowing. Baron-Cohen proposes that the TOMM relies on information from SAM for its operation so that it should be intact (if a little delayed) in visual impairment but impaired in autism. There is currently some dispute about how far children with autism are specifically impaired in developing a theory of mind outside of their difficulties with non-literal language and their lack of social experience of the contexts in which mental state terms are used and made explicit (Clements and Perner, 1993; Happe, 1994). Indeed, Stuart Powell and I have suggested that far from lacking a 'theory of mind', people with autism are the only group who need to develop such a theory since they have no natural intuitive route through to personal and social understanding and have quite literally to 'work it out' for themselves (Grandin, 1995).

Nevertheless, regardless of whether difficulties in this area are a prime disability in autism or are a secondary response to another difficulty with shared emotional experience (Hobson, 1993), there are clear differences in the ways in which children with autism understand (or, rather, fail to understand) themselves and others compared to the understanding of visually impaired children who do not have autism.

5 Particular Problems of the Dual Disability

That brings us to the question of the effects of a dual disability. What is it like to have a visual impairment and autism and how can the teacher take account of both factors in devising educational strategies? From the point of view of teaching a child with a visual disability, the knowledge that he or she also has autism means that autism-specific approaches will need to be adopted. From the alternative perspective then, of teaching a child with autism, what difference will the additional visual disability have? In one sense, the difficulties associated with autism remain the same, in that they are conceptual and not related to any sense modality. However, the compensatory approaches that are used will be affected. Because of difficulties in perceiving social and communicative intent, the most effective teaching strategies are those where meanings are made explicit and where the task is structured so that the child knows what to do, where, how, and for how long. Difficulties in attending to auditory information and for keeping intentions 'on-line' so that they direct and monitor actions also means that the child needs a permanent and accessible form of this information. Most systems developed for teaching pupils with autism (such as TEACCH: Rutter and Schopler, 1983) have relied on visual information for this since that fulfils those requirements and it has been found that most people with autism are visual rather than verbal thinkers (Jordan and Riding, 1995).

Clearly, where there is a visual impairment, the teacher will need to take the same principles (of clarity, explicitness and structure) and adapt them to the provision of support through the other senses. However, verbal instructions will not serve as a substitute and it may be better to think of musical support for actions and sequences of actions and tactile information providing support during problem solving tasks. Just as the child with autism may find it easier to take in and use written language rather than spoken language, so tactile language systems such as Braille may be easier for them. It is interesting to speculate, however, (and you may be aware of the answer to this) whether the visually impaired child with autism might not find spoken language easier to interpret (as long as there are not too many implications to be drawn, and it remains explicit) than do children with only autism. This is suggested because one source of difficulty with spoken language for the child with autism is that it comes accompanied by many confusing non-verbal social signals such as facial expressions and gestures, which cannot be interpreted. A visual disability would rob speech of these sources of confusion and, if it was also robbed of the confusing intonational characteristics, as happens with computerised speech, it may become accessible for them.

6 Effective Educational Strategies

When it comes to making effective provision for pupils with a dual disability, 1 feel that the approaches used for the visually disabled have a lot to offer, as long as the specific difficulties associated with autism are also accounted for. Just as the child with a visual disorder needs to have the environment clearly mapped for them and for that environment to remain constant, if the child is to feel secure and able to move around, so the child with autism needs these same structures, but this time to also include the social environment. One would not think of moving the 'furniture' without warning for the child with a visual impairment and so one would not change a routine or social ritual for the child with autism without appropriate warning also. 'Blindisms' are understood in terms of the child's attempts to manage his or her world and so too autistic stereotypies must be seen as serving a function for the child and only to be replaced with more appropriate behaviour if that behaviour can be demonstrated (to the child) to serve the same function. In the end one cannot teach effectively unless one also understands and that must apply, whatever the disability, or combination of disabilities.

7 References

Anderson, ES., Dunlea, A and Kekelis, LS (1984), Blind children's language: resolving some differences. Journal of Child Language, 11, 645-664.

Baron-Cohen, S (1995), Mindblindness: an Essay on Autism and Theory of Mind. London, MIT Press.

Bruner, J and Feldman, C (1993), Theories of mind and the problem of autism in Baron-Cohen, S, Tager-Flusberg, H and Cohen, DJ (Eds). Understanding Other Minds: Perspectives from Autism. Oxford, Oxford University Press.

Clements, WA and Perner, J (1994), Implicit understanding of belief Cognitive Development, 9, 377-395.

Exeley, R (1995), My History: Personal recollections of educational history'. Autism Research Conference, Durham, 1995.

Fay, W and Schuler, A (1980), Emerging Language in Autistic Children. New York, University Park Press.

Fraiberg, S (1979), Blind infants and their mothers: an examination of the system. In Bullowa, M (Ed) Before Speech. Cambridge, Cambridge University Press.

Frith, C (1992), Schizophrenia: a Cognitive View. Cambridge, Cambridge University Press.

Frith, U (1989), Autism: Explaining the Enigma. Oxford, Blackwells Press.

Gense, MH and Gense, DJ (I 994), Identifying autism in children with blindness and visual impairment. Review, 26, 56-62. Grandin, T (1995), How people with autism think in Schopler, E and Mesibov, GB (Eds) Learning and Cognition in Autism. New York, Plenum Press.

Happe, FGE (1994), Wechsier IQ profile and theory of mind in autism: a research note. Journal of Child Psychology and Psychiatry, 35, 1461- I 471.

Hobson, RP (1993), Autism and the Development of Mind. Hove, Erlbaum.

Jordan, RR (1989), Understanding and Use of Speaker Addressee Pronouns by Autistic Children. British Journal of Disorders of Communication, Vol. 24 No. 3 169-179.

Jordan, RR (1996), Teaching Communication to Individuals within the Autistic Spectrum REACH- Journal of Special Needs Education in Ireland, 9, 95-102.

Jordan, RR and Powell, SD (1995), Understanding and Teaching Children with Autism. Chichester, Wileys.

Jordan, RR and Riding, RJ (1995), Cognitive style in autism in Lindfoot, G, Shattock, P, Finnigan, R and Savery, D (Eds) Psychological Perspectives in Autism Proceedings of Durham Conference, 5-7 April 1995.

Kaufman, B (I 994), Son Rise: The Miracle Continues. New York, Better Books.

Leekham, S, Baron-Cohen, S, Perrett, D, Milders, M and Brown, S (1994), Eye-direction detection: a dissociation between geometric and joint attention skills in autism. Unpublished manuscript, Institute of Social Psychology, University of Kent.

Leslie, AM (1991), The theory of mind impairments in autism: evidence for a modular mechanism of development? in Whitten, A (Ed) The Emergence of Mind-reading. Oxford, Blackwells.

Powell, SD and Jordan, RR (1993a), Diagnosis, intuition and autism. British Journal of Special Education, 20, 26-29.

Powell, SD and Jordan, RR (1993b), Being subjective about autistic thinking and learning to learn. Educational Psychology, 13, 359-370.

Russell, J (1994), Agency and early mental development in Bermudez, J, Marcel, AJ and Eilan, N (Eds) The Body and the Self. Cambridge Ma MIT Press.

Rutter, M and Schopler, E (1978), Autism: a Reappraisal of Concepts and Treatments. New York, Plenum Press.

8 Biographical Detail

Rita Jordan took her first degree in Psychology and, while at home with her young children, she worked as a primary and nursery teacher part-time and started a playgroup for pre-school children with special educational needs and then a Toy Library. Three of the original 20 children who attend that group had autism, one of them with Asperger syndrome. After taking an MSc in Child Development, Rita worked as a research officer for the London Institute of Education, developing a curriculum for individuals with profound and complex learning difficulties and then as a teacher in a school for children with severe learning difficulties. She was Deputy Principal for 81/2 years at a school for pupils with autism. During this time she took an MA in applied linguistics. Throughout her teaching experience Rita has had pupils with visual impairments (including those described as 'educationally blind') although mostly they have had additional co-occurring disabilities. From the school for children with autism she went to teacher training in special educational needs, linguistics and the psychology of education at Hertfordshire College of Higher Education (which later became part of the University of Hertfordshire). On 1st September, 1993 she took up the first UK post in an education department as a full-time Lecturer in Autism at the University of Birmingham. Rita has researched and published widely on the educational and linguistic aspects of autism. She is co-director with Dr Stuart Powell of the Educational Research into Autism Group which has centres in the University of Birmingham and the University of Hertfordshire. She is a Chartered Psychologist.