5th Congress Autism-Europe
Artículos / Proceeding
Autism-Spain

APPLICATIONS OF THE THEORY OF MIND APPROACH TO ASSESSMENT AND INTERVENTION OF CHILDREN WITH AUTISM

Juan Carlos Gómez, Beatriz López y Enrique López

School of Psychology, University of St. Andrews, St. Andrews KY 16 9JU, UK

Equipo PAUTA

Centro PAUTA, c/Patrocinio Gómez 1 bis, Madrid 28022

INTRODUCTION

In this paper we report some preliminary results of a project whose aim is to explore some practical implications of the theory of mind approach for the assessment and intervention of persons with autism. During the last years this approach has produced important advances in the theoretical understanding of autism (see Frith, 1989; Baron-Cohen, Tager-Flusberg and Cohen, 1993; Happé, 1994). According to it, persons with autism suffer from an alteration in their ability to understand the nature of mental representations and their role in determining the behaviour of people. This approach has generated a wealth of empirical findings and theoretical debates in relation to autism. However, its possible impact upon the practical issues of assessment and intervention in persons with autism has received less attention. The project reported in this presentation was conceived as a collaboration between the theory and the practice. It was carried out at PAUTA, a school for children with autism and generalized developmental disorders. The general aim was to develop new or modify existing techniques for both the intervention and the assessment of the children attending this center.

The project was focused upon three main areas of impairment related to theory of mind:

a) Understanding that behaviour depends upon mental representations, i.e., that people act on the basis of their beliefs and desires. This ability is usually diagnosed by means of "false-belief" tasks (Baron-Cohen, Leslie and Frith, 1985), in which children with autism tend to fail to predict the behaviour of people with inaccurate representations of reality. They seem to find it extremely difficult to understand that other people may have a wrong representation of the world and act in accordance with this misrepresentation. In normal development, this ability typically appears at around 4 years of age and it is deemed to involve the capacity to form metarepresentations.

  1. Pretence and imagination. People with autism also tend to have great difficulty with pretend play and imagination in general. Pretend play is generally present as early as at one and a half years of age in normally developing children. For example, they pretend that a wooden block is something different (e.g., a car), or they imagine an object that does not exist (e.g., that they are holding a glass with "water"), or attribute an inexistent property to an object (e.g., the table is "wet" because the "water" fell out of the glass). The Theory of mind approach defends that the same basic cognitive mechanism is used to understand pretence and false belief, since in both cases it is a matter of understanding the difference between representation and reality. Indeed, in the case of pretence, what the child does if playing with this ability.

c) Joint attention behaviours. Before pretend play, normally developing children are capable of communicative behaviours such as showing or requesting things by means of gestures (e.g, pointing). These behaviours imply some understanding that one can make other people notice one's desires or notice one's interest in particular things in the world. This is why they are usually considered to imply some kind of theory of mind or a precursor thereof. Many of these behaviours are typically absent or rare in children with autism.

The project was divided into two phases. In the first phase, we developed and applied techniques to assess the developmental level of children in each area. In the second phase, new intervention techniques or modifications of existing procedures were developed and applied to the children.

In this report we are going to concentrate upon one example of assessment and one example of intervention.

JOINT ATTENTION IN COMMUNICATION

Traditionally it has been considered that people with autism have more problems with protodeclarative communication (e.g., pointing for showing things) than with protoimperative communication (e. g., pointing for requesting things) (see Curcio, 1978; Baron-Cohen, 1989). However, it has been shown that when the joint-attention components of protoimperative communication are taken into account, most children with autism also have problems with requesting (Gómez, Sarriá  and Tamarit, 1993; Phillips et al., 1995). Good requesting behaviours consist not only of producing gestures to which others can respond, but also of checking and manipulating the attentional states of the people who are expected to respond to the requests.

Teaching how to make requests is an important component of intervention with children with autism. Some degree of control for visual attention is usually included as a part of the teaching. We felt that it was necessary to create a systematic procedure to assess the degree of understanding of the role played by the attention of others in requesting siuations. Do the children take into account the attentional availibility of others when making their requests? If so, what are the cues they use to assess attention and inattention?

The assessment scenario was the following. In a small room an adult was sitting behind a table. Desirable objects (chosen in accordance with the known preferences of each individual child) were available on a shelf to the left of the adult. Neither the shelf nor the adult were directly accessible to the children (although they could touch the hands of the adult when these were resting on the table).

We recorded the spontaneous strategies, if any, used by the children to request the objects, noting if eye contact was or not used. After a few normal trials in which the adult was attentively responding to the chlld's attempts, experimental trials began to be interspersed among the normal ones. In these experimental trials the adult adopted a variety of "inattentive" postures that prevented him/her from seeing the gestures used by the child.

This is a list of the "inattention" conditions used:

a) Adult sitting with his/her back to the child

b) Adult's head turned to a corner.

c) Head to the child but eyes closed.

d) Head and eyes oriented to the target object

e) Adult oriented to the child but failing to respond for a while.

Since a favourite form of request for some children with autism involves taking the adult's hand and pushing it towards the object (which would intrinsically perform the function of calling the attention of the adult even if this were not intended by the child), we systematically controlled the accesibility of the adult's hands, keeping them under the table in some critical trials.

The aims of the situation were, first, to confront the children with a situation that was unusual in terms of the ones in which they were being taught at the school. (The positioning of the adult behind the table with the corresponding limitation in direct physical access was not a usual one for the children); and, second, to see if the children were sensitive to different signs of inattention and they were capable of modifying their request procedures resorting to some kind of attention getting behaviours. This could give us an idea of the extent to which the children had mastered the joint attention component of requesting behaviours or they were just using effective gestures without any insight as to the connection between these and the response by the adult.

All trials were videotaped and subsequently analysed. On the whole the requesting strategies of the children when the adults were attending to them could be classified in the following four categories:

a) No request: the child tries to use physical means to obtain the object.

b) Takes and uses the other person's hand without making eye contact.

c) Use of distal gestures and vocalization without eye contact.

d) Use of distal gestures and vocalization with eye contact.

None of the children stuck to the use of the first strategy which involves completely ignoring the human. Although some initially tried to obtain the target by themselves, all of them eventually resorted to the human in one way or another. Four of the children assessed were using the hand without paying any attention to the face of the person. Two of them used the hand but looked at the face of the adult. Three children used distal gestures and vocalizations without looking at the face, and seven did so looking at the face of the other person.

The reactions of the children to the different situations in which the adults were not attending were as follows:

--Same strategy as in the previous trials. No sign of having detected the lack of attention of the adult. This reaction was shown by five children.

--Absence of request when the adult is not looking, but fails to call his/her attention. (4 children).

--Calls the adult's attention but without making eye contact. (one child).

--Calls the adult's attention making eye contact. (two children).

As many as nine of the children were helpless when confronted to the situation where the adult was not responding because he was not attending to them. Some of them seemed to detect the lack of attention of the adult (especially in the situation when he was with his back turned to them), but did not know how to alter the situation. So, they simply stopped making requests. However, others would just repeat their request without regard of the other person's attentional availibility.

It is specially remarkable the reactions of the children whose main requesting procedure was taking the hands of people towards the target object. When they were deprived of this simple means because the adult hid his hands under the table, most of them were totally unable to produce an alternative strategy. Some would strive to get hold of the adult's arm and then take his hand to employ their usual procedure. Others would give up or try a physical strategy. None of them would request the hand of the adult, for example by means of stretching his/her own hand towards the other person.

The results of each individual child can be used to plan individualized intervention strategies. For example, in the case of the children using the hands of others, the intervention will consist of promoting the use of distal gestures as a way of obtaining the hands of people and/or as a substitute for the hand-taking procedure. In the case of children showing their sensitivity to the inattention of others by interrupting their requests, it is possible to try to make them notice how they can affect the attention of the other people by means of actions (such as touching) or vocalizations (such as calling the name of the person).

The results of this preliminary application of the assessment procedure show the real complexity of requesting behaviours and the need to assess the understanding of their joint attention components. The systematic variation in the subtlety of the available cues of inattention (ranging from gros cues, such as having the back turned, to subtler cues, such as having the eyes closed) could give an idea of what are the signs of attention to which different children are sensitive. This, in turn, can be used to help in designing the targets of the intervention in the area of joint attention.

UNDERSTANDING MENTAL STATES IN OTHERS

One of the most surprising findings of Theory of mind research has been the contrast between understanding pictographic representations versus understanding mental representations in high-level persons with autism. Children who remain unable to understand false beliefs and their connection to mis-directed behaviours (as shown by their failure in the Sally-Ann test) are, however, capable of understanding false photographs or pictures. For example, they can predict that the colour of a doll's dress on a photograph will not change although in reality the doll changes her dress after the photograph has been taken (see Happé 1994 for a summary of these studies).

The idea behind this intervention technique is to use this intact ability to understand pictographic representations as a basis to teach children with autism an alternative to a regular theory of mind. This possibility has been explored in collaboration with a British team. So far we have carried out a controlled experiment with 8 young children using the mind-as-photograph-camera analogy (reported in Swettenham et al., 1996), and the case study I'm going to present here using the "drawings in the head" analogy.

The aim was to teach a young girl the idea that people make drawings of things in their heads and they have a look at them before acting in relation to these things. To illustrate this idea, we resorted to the convention of drawing "bubbles" besides the heads of drawn characters. These bubbles -the child was told- show the "thoughts" of the character, what she or he has in the head. The girl was also told that a character can be "thinking" (or have in his/her head) the "truth" or a "lie". People had the "truth" in their minds when the contents of the bubble coincided with what was the case in the world; however, they had a "lie" when the contents of the bubble differed from reality.

This was first introduced as a game with a couple of dolls (Ana Roberta and Roberta Ana). The dolls were involved in different situations in which objects were changed in location. The girl was told, first, that the bubbles did not change when the changes in the real world occurred in the absence of the bubble holder. Then, she was also taught that the actions of the doll would correspond to the contents of the bubble, even when these were a "lie".

Progressively, the need for actually drawing the bubbles corresponding to each character was faded. The girl was encouraged just to think of the bubble each doll would have and tell what they would be doing as a consequence.

After several weeks of teaching in these informal situations, the girl was capable of passing the standard false-belief test posed with the same dolls that had been used for the training. This passing was accompanied by a verbalization in which she reasoned "ah, she has the lie in her head, therefore she will look there".

However, the aim of the teaching was not to enable this girl to pass Theory of mind tests, but to give her a tool that would be useful in her everyday life. Consequently, the teachers would encourage her to analyse spontaneous situations arising in her everyday life that involved some kind of false belief misunderstanding. She was encouraged to think of these situations in terms of the kinds of bubbles ("true" or "false") held by the people involved. For example, in an occasion she was punished for having hit another girl. Soon after, it turned out that it was likely that she did not hit that girl and, therefore, she had been punished wrongly. To ascertain what happened she was asked to draw what the punisher had in her head and to label it as "true" or "false". She drew a picture of herself hitting the other girl and labeled the bubble as "false".

In another occasion, upon hearing that a Fax had arrived from Juan Carlos, she said: "A fax from the king!". She was then explained that it was not the king's fax, but a fax from me. When asked to draw what she had thought, she again was capable of correctly analysing the misunderstanding.

We are now conducting an experimental study about the suitability of this technique with a group of children from Great Britain, in collaboration with the same team that camed out the "photos-in-the-head" study.

PICTOGRAPHIC REPRESENTATIONS AND PRETEND PLAY

The use of photographs and drawings has also proved useful in teaching aspects of pretend play. A crucial step in the development of this ability in the children from the school is the passage from playing with miniatures to using substitute objects instead of the miniatures. A good technique to facilitate this passage is to keep constant a particular scenario with which the child is already familiar while substituting an object for a single particular miniature. For example, if the child knows a script for playing the "Cafeteria" with a collection of bottle, dish, and cutlery miniatures, one possibility is to hand him a wood block instead of the ketchup bottle at the particular step in which he introduces the action of putting the ketchup bottle on the tray. Keeping constant the other objects and the action may help him to accept the pretence.

However, even in these conditions some children have difficulties accepting the pretend object and reject the block as a suitable object for the game. In these situations we have found that it can be helpful to keep a photograph of the absent object as reminder of the identity to be attributed to the block. At least in one case, the passage from rejection to acceptance was instantaneous upon recelving this help.

This suggests that the use of pictographic representations may be a highly suitable technique for trying to develop in children with autism some of the "metarepresentational" skills they seem to lack. Of course, what we are doing is to provide them with an analogue to deal with some aspects of this most important side of our world and life. This analogue is bound to have limitations and constraints in comparison with the original ability.

In summary, the discussion of new theoretical developments in the context of the practical issues of assessment and intervention with children with autism can be a useful source of ideas about new ways to carry out these functions. In this presentation we have discussed the practical derivations of a theoretical discussion about the role of joint attention in requesting behaviours. Also, we have shown how discoveries like the contrast between understanding pictographic versus mental representations can generate new ideas for intervention.

REFERENCES

* Baron-Cohen, S. (1989). Perceptual role taking and protodeclarative pointing in autism.

British Journal of Developmental Psychology, 7: 113-127.

* Baron-Cohen, S.; Leslie, A. M.; & Frith, U. (1985). Does the autistic child have a 'theory of mind'? Cognition, 21: 37-46.

* Baron-Cohen, S.; Tager-Flusberg, H.; & Cohen, D. (Ed.). (1993). Understanding other minds: perspectives from autism. Oxford: Oxford University Press.

* Curcio, F. (1978). Sensorimotor functioning and communication in mute autistic children. Journal of Autism and Childhood Schizophrenia, 8: 281-292.

* Frith, U.; & Happé, F. (1994). Autism: beyond theory of mind. Cognition, 50: 115-132.

* Gómez, J. C.; Sarriá , E.; & Tamarit, J. (1993). The comparative study of early communication and theories of mind: ontogeny, phylogeny and pathology. In S. Baron-Cohen, H. Tager-Flusberg, & D. Cohen (Eds.), Understanding other minds: perspectives from autism (pp. 397-426). Oxford: Oxford University Press. (Trad. castellana: "El estudio comparativo de la comunicación temprana y las teorías de la

mente: ontogénesis, filogénesis y patología". Siglo Cero Vol. 24 (6): 47-62.)

* Happé, F. (1994). Autism: an introduction to psychological theory. London: UCL Press.

* Phillips, W.; Gómez, J. C.; Baron-Cohen, S.; La , M. V.; & Rivière, A. (1995). Treating people as objects, agents or "subjects": How young children with and without autism make requests. Journal of Child Psychology and Psychiatry, 16(8): 1383-1398.

* Swettenham, J.; Gómez, J. C.; Baron-Cohen, S.; & Walsh, S. (1996). What's inside someone's head? Conceiving of the mind as a camera helps children with autism acquire an alternative to a theory of mind. Cognitive Neuropsychiatry .1(1): 73-88.

ACKNOWLEDGEMENTS

This project was funded by a grant of the Spanish Ministry of Social Affairs (Ministerio de Asuntos Sociales). J. C. Gómez also acknowledges the support of a DGICYT grant (PB92-0143-CO2-02) from the Spanish Ministry of Education.