Abstract

Young children's understanding of indeterminacy was investigated in an extension and partial replication of Fay & ICIahr (1996). They asked preschoolers to respond know or guess to various evidence patterns that included matches between the target and a potential source of that target (+), mis-matches (-), or unknown matches(?). Fay & Klahr found that children were able to correctly respond know to indeterminate problems such as (- + - +) in which it was clear that more than one item matched the target. However, most pre-schoolers incorrectly responded know to problems in which there was a single match plus some remaining unexplored evidence(e.g., - + ? ?). Fay & Klahr called this phenomenon positive capture. One possible source of positive capture is children's mistaken assumption that all items in each problem set were unique. In the present study, we explored this possibility by providing kindergartners with different models of the uniqueness or sameness of items in each problem set, focusing on problems like (- + ? ?), to which positive capture responses are common. Four models were compared. (1) No Explicit Model： Problems were presented as in Fay & Kiahr with no additional information about the relationship among possible items. (2) Explicit Exclusive Model- Children were told that each item was unique. Only one item could match the target. (3) Explicit Target Redundant- Children were told that two of the items in the set were the same. The target item matched the two identical items. This model was expected to counteract the uniqueness assumption by making explicit the indeterminacy of the evidence, leading more children to give the correct answer guess, And, (4) Explicit non-Target Redundant: Same with ETR, but the non-target item matched the two identical items. After each positive capture problem children were also asked whether any of the unrevealed items could match the target (probe question). Performance in the No Explicit Model condition replicates Fay & Klahr; the majority of children committed the positive capture error on (- + ? ?) problems. In response to the probe question, 67,7% of kindergarten children indicated that there were no more matches among the unseen items (uniqueness assumption). Even when it was clearly stated that there were two matches (Explicit Target Redundant), younger children made positive capture errors. In this condition, 90.6% of kindergarten, children showed positive capture, and 31.3% of them claimed there was only one match, in contrast to the given model. Thus, even when children recognized that there were two matches, many were still unable to overcome the positive capture error.