Elementary Cognitive Tasks and Their Roles in g Estimates Dasen Luo Indiana University of Pennsylvania, Indiana, PA, USA Stephen A. Petrill Case Western Reserve University, Cleveland, OH, USA In this study, we examined the relationship between elementary cognitive tasks (ECTs) and the general factor, or g, of intelligence. We investigated the pattern of correlations of these tasks with estimates of g, and their roles in g estimates when g estimates were used to explain scholastic performance. We found that, in addition to a general information processing component, there is a memory processing component, nonchronometric in nature, that substantially contributes to g estimates. We also found that using ECTs to define g, along with traditional psychometric tasks, does not significantly alter the axis of g in relation to the axis of the scholastic performance factor in the factorial space. In other words, the predictive power of g will not be compromised when g is defined using experimentally more tractable ECTs. The theoretical and practical implications of such findings were discussed. Within the last 30 years, there has been a dramatic renewal of interest among individual difference researchers in elementary cognitive tasks (ECTs). One topic that has been at the forefront of such a research interest is the relationship between ECTs and a more general index of intelligence, or g. ECTs have been credited for being more tractable in their psychological mechanisms, and it has been hoped that an established relationship between ECTs and g can resolve to a certain extent the problem of a circular definition for intelligence. Recent studies have demonstrated that ECTs primarily tapping the speed of processing have reliable correlations with g (Vernon, 1987). These correlations, after correction for attenuations due to restriction of range and measurement error, can usually show a moderate and strong relationship between ECTs and g. These findings notwithstanding, certain important questions regarding the relationship between ECTs and g remain. One question that has been discussed recently is whether ECTs relate to g through a unitary process or through multiple components (Detterman, 157 Direct all correspondence to: D. Luo, Indiana University of Pennsylvania, Department of Psychology, 704 Whittier Drive, Indiana, PA 15701, USA. INTELLIGENCE 27(2): 157±174 Copyright D 1999 by Elsevier Science Inc. ISSN: 0160-2896 All rights of reproduction in any form reserved.