Attention and working memory as predictors of intelligence

The paper reports on an investigation of attention and working memory as sources of intelligence. The investigation was concentrated on the relatedness of attention and working memory as predictors of intelligence and on the structure underlying the prediction. In a sample of 120 participants, intelligence was assessed by the Advanced Progressive Matrices [APM; Raven, J. C. (1962). Advanced progressive matrices. London: Lewis and Co.] and Zahlen-Verbindungs-Test [ZVT; Oswald, W. D., & Roth, E. (1978). Der Zahlen-Verbindungs-Test (ZVT). Gottingen: Hogrefe]. Attention was restricted to sustained attention and measured by means of two versions of the Frankfurt Adaptive Concentration-Performance Test [FACT; Moosbrugger, H., & Heyden, M. (1997). FAKT Frankfurter Adaptiver Konzentrationsleistungs-Test. Testmanual. [Frankfurt Adaptive Concentration-Performance Test FACT Test Manual]. Bern, Gottingen: Huber]. The Exchange Test [Schweizer, K. (1996a). The speed-accuracy transition due to task complexity. Intelligence, 22, 115-1281 and the Swaps Test [Stankov, L. (2001). Complexity, metacognition and fluid intelligence. Intelligence, 28, 121-143] represented working memory capacity. Structural equation modeling revealed that attention and working memory predicted overlapping parts of intelligence. The data suggested different models for APM and ZVT as criterion variable. When APM represented intelligence, the final model suggested both working memory and attention as significant predictors. In contrast, when ZVT represented intelligence, the model included only attention as significant predictor. The restriction of the models to working memory as single predictor led to an insufficient result. (C) 2004 Elsevier Inc. All rights reserved.