Speed- and accuracy-related measures of an intelligence test are differentially predicted by the speed and accuracy measures of a cognitive task

Abstract

Raw scores on time-limited multiple-choice intelligence tests are determined by incorrect responses and missing answers. Both these error types were previously found to be negatively related to each other. Individual differences in the emphasis on speed or accuracy can explain this finding. But even though individual differences in the emphasis on speed or accuracy have been identified not only in intelligence tests but also in cognitive tasks, little is known about their interplay. Therefore the aim of the present study was to investigate to what degree speed- and accuracy-related performance scores of an intelligence test can be predicted by speed and accuracy measures of cognitive tasks, respectively. For this purpose, 200 participants completed Cattell's Culture Fair Intelligence Test (CFT 20-R) and performed the Swaps Task, an experimental cognitive task. To investigate the interplay between the speed and the accuracy measures of both kinds of task, a latent variable approach was used. Overall, the emphasis on speed or accuracy was not systematically related to the intelligence score. However, closer inspection of the data revealed that reaction times, but not errors rates, in the Swaps Task predicted the number of not-reached items as an indicator of speed in the CFT 20-R. At the same time, error rates, but not reaction times, in the Swaps Task predicted incorrect responses as an indicator of accuracy in the CFT 20-R. Taken together, speed- and accuracy-related performance scores of an intelligence test were predicted by speed and accuracy measures of a cognitive task, respectively. Most important, however, the finding that the emphasis on speed or accuracy was not significantly related to intelligence scores clearly indicated that this emphasis does not interfere with the validity of the intelligence test.