Background. The Representational Redescription model of cognitive development (Karmiloff-Smith, 1992) identified a level when a child's representation of a task may be incorrect but difficult to access or change. On the balance beam task children at this level have been found to hold a fixed belief that all objects must balance at their geometric centre, even those which have more weight at one end and actually balance off-centre.
Aims. In this study we sought to determine whether children who hold this 'centre theory' about balance can alter their beliefs when exposed to an adult teaching them the correct solution. Since the RR model stresses the top-down, impenetrable nature of this type of representation our aim was to see whether particular interventions might bring about progress in these children.
Sample. The participants were 42 children, from 4, 11 to 7,4 years, who demonstrated at pre-test that they held a centre theory by being unable to balance any beams which were unevenly weighted.
Methods. The children were randomly assigned to one of three conditions where they either saw an adult modelling and explaining the correct solution, or an adult worked with them offering guided participation, or the child worked alone.
Results. Progress was measured by improvement from pre-to post-test in ability to balance unevenly weighted beams. Children who had seen an adult modelling and explaining the task showed greater improvement than those who worked alone, although this difference was not significant at a delayed post-test one week later.
Conclusion. Although many children with a fixed naive theory about balance were identified this study demonstrates that, with the right type of input, it is possible to facilitate progress.