Intelligence process vs. content and academic performance: A trip through a house of mirrors

The main purpose of modern intelligence tests has been to predict individual differences in academic performance, first of children, then adolescents, and later extending to adults. From the earliest Binet-Simon scales to current times, most one-on-one omnibus intelligence assessments include both process subtests (e.g., memory, reasoning) and content subtests (e.g., vocabulary, information). As somewhat parallel developments, intelligence theorists have argued about the primacy of the process components or the content components reflecting intelligence, with many modern researchers proposing that process constructs like working memory are the fundamental determinant of individual differences in intelligence. To address whether there is an adequate basis for re-configuring intelligence assessments from content or mixed content and process measures to all-process measures, the question to be answered in this paper is whether intellectual process assessments are more or less valid predictors of academic success, in comparison to content measures. A brief review of the history of intelligence assessment is provided with respect to these issues, and a number of problems and limitations of process measures is discussed. In the final analysis, there is insufficient justification for using process-only measures to the exclusion of content measures, and the limited data available point to the idea that content-dominated measures are more highly predictive of academic success than are process measures.