Developing learning model and software for rehabilitation of autistic reasoning

Project objective

Resent psychological studies have revealed that autistic children can neither reason properly about mental states of themselves and others nor understand emotions. Autism is a multifactor disorder that is characterized by impaired social interaction and communication, combined with repetitive and stereotyped patterns of behavior, and affects up to 1% of school-aged children in some geographic areas. In this proposal we are concerned with the strategies of rehabilitation of reasoning to improve communication skills of children with autism.

     It has been confirmed by multiple clinical studies that the properly timed treatment is essential for the autistic patient to increase the chance for recovery. An early behavioural intervention is highly beneficial for autistic children. There is an opinion with increasing support by multiple researchers that intensive behavioural intervention may result in a dramatic improvement of communication skills of children with autism. A majority of autism experts believe that there is no alternative to behavioural intervention thought as the only way to facilitate compensatory learning. Teaching proper reasoning about mental states is one of the essential components of behavioural intervention, in accordance to the Theory of Mind, which is quite popular today.

     Multiple technologies have been suggested to support autistic learning, including software and hardware system developing particular behaviour patterns. However, a general framework of reasoning rehabilitation with explicit orientation to the mental world is currently unavailable for most autistic patients. A majority of current software tools does not directly cure the reasoning and decision–makings skills for living among humans with intentions and beliefs. Instead, they teach certain forms of behaviour without explaining their simplest components, the basic mental actions and mental states, which are easily perceived and understood by normal children.

      We propose to develop from existing prototypes and integrate a set of intelligent software tools with the clear focus: operating with mental attitudes. Over last seven years we have deployed a variety of means to enforce a patient attention to the limited set of mental entities, including the basic (want-believe-know) and derived (pretend, deceive, offend, forgive, ask, answer, etc.). The next step for the reasoning rehabilitation software we are currently proposing is to demonstrate that this limited set of entities allows describing the major part of mental world. Having acquired the skills of operating with these entities (which is indeed reasoning), an autistic patient becomes capable of understanding themselves and others, as our studies have demonstrated.  We have also verified that the learners can apply the acquired patterns of mental entities to real-world scenarios.

       Various modes of interactions with a trainee should be employed in software components. Keeping attention of autistic patients by its intelligent behaviour, the system will be able to predict what will happen with agents (their future mental states), given an initial mental state specified by the patient. The system will also be capable of understanding children’s questions about mental states of characters of a scene or of a short story. Moreover, the system will ask questions to children concerning these characters and analyse their answers. In addition, there is an existing tool where children "build up" the definitions of above mental entities; it has been proven to be efficient but should be implemented online.

       Various ways of specifying mental states are helpful in developing the skills to operate with the mental entities. Typing words into an interactive form, selecting appropriate mental entities from a list box, outlining a meaning of a given mental entity via a set of list boxes – all these exercises assist the children with autism in their adaptation to mental world. 

 

 

 

The main beneficiaries of this project are children with autism who are capable of using computers, and their parents. Usually even pre-verbal autistic children are strongly attracted to computers. It is important, however, that autistic children are occupied with exercises with clear focus on development of a particular skill rather than an arbitrary game. It is well-known that autistic children can develop a strong interest to a particular domain which may not be of interest to a majority of others. The task of rehabilitation software is therefore to transfer a passion of autistic trainee to developing creative skills which would be appreciated by others, including the cabability o freasoning about mental world.

     The software is intended for a wide audience of trainees with communication problems, or those who intent to develop their communication skills further. Currently, the difficulty of exercises can be significantly varied, from very basic to those which require operation with quite complex mental states. We propose to develop the online version with integrated tools targeting children from 6 years old to adults up to 16 years old; for autistic children the respective mental age range is assumed.

The software to be develop will benefits a wide spectrum of children:  from preschoolers to teenagers, from classically autistic to Asperger's Syndrome. Children with intellectual disabilities (like Autism, Down syndrome, Pervasive Developmental Disorders, developmental delay) are targeted as well.