Information integration by children: the identification of rules by an alternative method

Two empirical studies were conducted with 136 preschool children, aged 5 to 6, and 89 adults. Using rectangles of varying sizes, a paired-comparisons procedure revealed two integration rules that determined the children's responses: a multiplying rule and a centration rule characterized by a preference for the maximum of height and width. The adults applied the multiplying rule, but if this rule failed, they used rules similar to those applied by the children. In none of the age groups was there evidence for an adding rule. In a second study, these results were replicated for two other types of response scales, rank ordering and rating, indicating that the results of the first study were not due to the use of the paired-comparison procedure. The present results differed from results of previous studies that used different techniques for the analysis of integration rules.