Measuring fluid intelligence at age four

Individual differences in the broad ability dimension General fluid intelligence (Gf) are typically assessed with visuospatial problem solving tasks, and particularly so among young children. Such tests may, however, contribute construct-irrelevant variance due to the figural content of the tasks. The main aim of the study is, therefore, to investigate if measures of working memory capacity can add to the measurement of Gf. A sample of 364 children aged four were given a test-battery which included 10 tests designed to measure different aspects of Gf: three visuospatial problem solving tests, two visual short-term memory (STM) tests, two verbal STM tests, and three verbal tests of working memory capacity. Confirmatory factor analysis showed that an oblique model with four factors fitted the data well, as did a bifactor model with a general factor, along with verbal and visual modality factors. The bifactor model was the preferred solution, and it was concluded that the general factor in this model represents Gf. It was also observed that the visuospatial problem solving tests were influenced by both Gf and the visual modality factor, the latter contributing construct-irrelevant variance to the intended measurement of Gf.