| Abstract: | In two studies, we compared young children's performance on three variations of a nonverbally presented calculation task. The experimental tasks used the same nonverbal mode of presentation but were varied according to response type: (1) putting out disks (nonverbal production); (2) choosing the correct number of disks from a multiple-choice array (nonverbal recognition); and (3) giving a number word (verbal production). The verbal production task required children to map numerosities onto the conventional number system while the nonverbal production and nonverbal recognition tasks did not. Study 1 showed that the performance of 3-, 4- and 5-year-old middle-income children (N = 72) did not vary with the type of response required. Children's answers to nonverbally presented addition and subtraction problems were available in both verbal and nonverbal forms. In contrast. Study 2 showed that low-income children (3- and 4-year-olds; N = 48) performed significantly better on both nonverbal response type tasks than on the verbal response type task. Analysis of individual data indicated that a number of the low-income children were successful on the completely nonverbal calculation tasks, even though they had difficulty with verbal counting (i.e., set enumeration and cardinality). The findings suggest that the ability to calculate does not depend on mastery of conventional symbols of arithmetic. |
