International Meeting for Autism Research: Training Children with Autism Spectrum Disorders to Pass Theory of Mind Tasks Using Thought Bubbles

Training Children with Autism Spectrum Disorders to Pass Theory of Mind Tasks Using Thought Bubbles

Friday, May 21, 2010: 11:30 AM
Grand Ballroom E Level 5 (Philadelphia Marriott Downtown)
9:45 AM
J. M. Paynter , School of Psychology, The University of Queensland, St Lucia, Australia
C. C. Peterson , Psychology, The University of Queensland, St Lucia, Australia
Background: Impairments in theory of mind (ToM), in autism spectrum disorders (ASD), have received much attention (see Baron-Cohen, 2001). Recently, research has investigated whether these impairments may be amenable to change. This has relevance both practically and theoretically in terms of informing understanding of the mechanisms by which ToM may be acquired. Previous research has used a range of visual strategies to assist children with ASD to pass ToM tasks. Early research using photo or picture-in -the-head strategies (e.g. Swettenham, Baron-Cohen, Gomez & Walsh, 1996; McGregor, Whiten & Blackburn, 1998) demonstrated some success in improving pass rates in the trained task, but showed less success in generalisation to untrained tasks. More recently, the “thought bubble” method proposed by Wellman et al (2002) showed promise in improving pass rates on both the trained task and some generalisation. This method taught children with ASD that mental states were like thought bubbles, a convention universally recognised in children’s cartoons.  Thought bubbles were used to demonstrate and teach theory of mind concepts.  However, despite promising findings, with improvements on the trained changed location false belief as well as on a deceptive contents false belief task, this study suffered from limitations. Perhaps most notably, it lacked a control group. Given that in a more recent training study Fisher and Happe’(2005) found significant improvements in their control group in a similar deceptive contents false belief task, further research into thought bubbles is required.
Objectives: The current study sought to extend upon Wellman et al’s (2002) study by including a control group, adding a delayed follow-up test session to assess durability of training effects and including an empirically validated developmentally scaled ToM scale (Wellman & Liu, 2004) as an outcome measure to assess generalisation.
Methods: Twenty-four children with ASD participated; 17 completed training and seven comprised a non-intervention control group. Training was based around Wellman et al’s (2002) method around five stages, over two to four training sessions. All children completed pre/post standard false belief tasks and Wellman and Liu’s (2004) scale tasks. Nine children from the experimental group and the entire control group also completed a three week delayed follow-up session.
Results: The trained group showed significant improvements on false belief total score, as well as on Wellman and Liu’s (2004) scale. Improvements were maintained at follow-up. The control group did not show any significant changes on either scale.
Conclusions: This research lends further support to the growing body of evidence that children with ASD can be taught to pass ToM tasks. It adds specific support to the thought bubble method as a promising strategy to promote social-cognitive reasoning, which allowed for some generalisation beyond the trained task. This was maintained, at least in the short- term. Although in its preliminary stages, the thought bubble method shows great potential as a valuable tool worthy of further research to increase understanding of the mechanisms by which ToM may be acquired, and ultimately as a practical resource.
See more of: Cognition 2
See more of: Cognition
See more of: Autism Symptoms