The relationship between competence and performance in early development in children with Down's Syndrome

This thesis investigated the effects of performance deficits on the development of cognitive competence in children with Down's Syndrome (DS). Fifty-one DS subjects ranging in age between 3 months and 16 years participated in 5 cross-sectional and 2 longitudinal studies. Performance was studied both in a standardised assessment situation and in response to a set of discrimination learning tasks. Each of the 4 assessment studies revealed a substantial differential between levels of performance and true competence in this group of DS subjects. Although overall levels of 'IQ' found were similar to those reported in previous studies of cognitive ability in DS children, item-item analyses and analyses of test behaviour revealed a strong tendency to underperform. Task avoidance was frequent, resulting in very poor levels of test-retest reliability. A pattern emerged from the longitudinal data suggesting that there may be a causal relationship between this testretest unreliability and the apparent loss of skills at later age levels. It was found that children often withheld demonstration of optimal levels of performance on tasks with which they had encountered difficulty at younger age levels. The implications of the results from the assessment studies are three-fold: firstly, the repeated evidence of unreliable performance and of developmental instability suggests that psychometric methods of cognitive assessment should be used with great caution with this population of mentally handicapped children; secondly, and relatedly, it suggests that theories of development in DS based on outcome measures obtained from such tests may be inaccurate; thirdly, the frequency with which children were seen to 'lose' skills suggests that teaching effort should be focussed not only on encouraging the development of new skills but also on ensuring the consolidation of these skills once acquired. Three discrimination learning studies investigated whether an errorless learning procedure might enhance motivation to perform to full competence and increase performance reliability. Results from the first two studies were highly encouraging: with this technique DS subjects quickly learned the target discrimination and this skill was reliably demonstrated over 4 testing sessions. Results from a third study indicated, however, that the technique was of little value in teaching learning skills per se: learning did not transfer to a second, very similar discrimination task. It was concluded that while errorless learning procedures have a useful role to play in teaching DS children that they can learn, they may be most effective when used in conjunction with conventional trial-and-error teaching methods.