The characteristics of parent-child interaction and the impact it has on children's cognitive development has been an area of interest to developmental psychologists for many years, The seminal work of Vygotsky, which emphasised a distinction between the actual level of development and the zone of proximal development (ZPD), and proposed that learning proceeds most effectively when tutoring occurs in the ZPD, has provided a foundation for much subsequent research on interactive problem-solving. As part of this tradition, Wood and his colleagues (e.g., Wood, 1980; Wood, Bruner & Ross, 1976; Wood & Middleton, 1975; Wood, Wood & Middleton, 1978) have described the process of scaffolding which is used by parents and tutors to target the child's ZPD during interaction. The present study uses a Vygotskian perspective and the past work of Wood and Middleton (1975) to compare the characteristics of parent-child interaction on a problem-solving task between children with Down syndrome (DS) and typically developing (TD) children. Each parent­ child dyad completed a block model-copying task based on the one used by Wood and Middleton (1975) in their early research. Each session of interaction was then analysed and coded with regard to how general or specific the parent's interventions were, and how contingent they were upon the child's successes and failures, and level of ability. The frequencies of supportive and directive actions and vocalisations were also noted. In addition, each parent completed a measure of parenting style and the children were administered a test of expressive and receptive language development as a comparative measure of developmental level between the two groups. The results revealed many differences between the two groups. The parents of the TD children followed the pattern predicted by Vygotskian theory to be most effective, focussing more on the children's region of sensitivity to instruction (RSI) and responding contingently to their successes and failures. By contrast, the parents of the children with OS focussed more on their child's actual developmental level (ADL) than their RSI, and followed the contingent shift rule less consistently than the parents of the TD children. They also intervened in the task more often, using higher, more specific levels of intervention, and were both more supportive and more directive than the parents of the TD children. The implications of these findings are discussed with reference to past research, and the proposition is put forward that the parents of the children with OS may be employing a different interactional strategy which is more adaptive to the different needs and abilities of their children.