The roles of labelling and abstraction in the development of cognitive flexibility

Three experiments were conducted to determine whether language contributes to the emergence of cognitive flexibility in preschoolers. The Flexible Item Selection Task (FIST), a measure of flexibility, was developed in Exp. 1 for use with preschoolers. On each trial, children were shown 3 items (e.g., a big red shoe, a big blue shoe, a small blue shoe) and asked to select 2 items that matched on one dimension (Sel. 1; e.g., size), and then to select a second pair that matched on another dimension (Sel. 2; e.g., colour). The results revealed that 2-year-olds did not understand task instructions and 3-year-olds performed poorly on both selections. In contrast, 4-year-olds did well on Sel. 1, but they did significantly worse than 5-year-olds on Sel. 2, suggesting that they had specific problems with switching flexibly between dimensions. It was then hypothesized that changes in flexibility between 4 and 5 years might be due to underlying changes in language. Exp. 2 was conducted to test this claim by using label manipulations on the FIST. The results revealed that Sel. 2 performance was not only correlated with receptive language development but it was also influenced by, labelling on Sel. 1. When 4-year-olds were asked to label the dimension on Sel. 1 that was relevant to that selection (the dimension on which items matched; e.g., size), their Sel. 2 performance improved significantly compared to children who were not asked to label, or who were asked to label the irrelevant dimension (the dimension that did not vary across the 3 items; e.g., shape). Also, children in the relevant-label condition who tended to label correctly on Sel. 1 also did better on Sel. 2 than children in that condition who made many labelling errors. Exp. 3 was then conducted to determine which kinds of labels improve 4-year-olds' performance on Sel. 2. In Exp. 3, the experimenter selected items on Sel. 1 and labelled them in some predetermined way. Results showed that labels that referred to the relevant dimension helped 4-year-olds on Sel. 2 whereas labels that referred to the irrelevant dimension did not. This pattern held whether the experimenter labelled the dimension (e.g., "size") or the cue (e.g., "big"). In general the results are consistent with the hypothesis that the emergence of flexible thinking in the preschool years may be mediated by language development.