Saturday, August 20, 2005

Working memory - more on Gf connection

As indicated by a number of prior posts ("Berlin BIS model of intelligence--material to review"; "g, working memory, specific CHC abilities and achievement") regarding the relationship between working memory (Gsm-MW) and fluid reasoning (Gf) or g, there is no shortage of contemporary research that continues to investigate the interesting relationship between higher-level cognitive processing and working memory. Below is a brief summary of yet another research article that sheds light on the possible reasons for the working memory/Gf/g relation.

Buehner, M., Krumm, S., & Pick, M. (2005). Reasoning=working memory ≠ attention. Intelligence, 33(3), 251-272.

  • The purpose of this study was to clarify the relationship between attention, components of working memory, and reasoning. Therefore, twenty working memory tests, two attention tests, and nine intelligence subtests were administered to 135 students. Using structural equation modeling, we were able to replicate a functional model of working memory proposed by Oberauer, Suess, Wilhelm, and Wittmann (2003) [Oberauer, K., Suess, H.-M., Wilhelm, O., & Wittmann, W. W. (2003). The multiple faces of working memory: Storage, processing, supervision, and coordination. Intelligence, 31, 167-193]. The study also revealed a weak to moderate relationship between the "selectivity aspect of attention" and working memory components as well as the finding that "supervision" was only moderately related to "storage in the context of processing" and to "coordination". No significant path was found from attention to reasoning. Reasoning could be significantly predicted by "storage in the context of processing" and "coordination". All in all, 95% of reasoning variance could be explained. Controlling for speed variance, the correlation between working memory components and intelligence did not decrease significantly.
Major findings
  • Oberauer, Suess, Wilhelm, and Wittmann’s (2000, 2003) model of working memory hypothesizes that working memory can be separated into two facets: a content facet (contains verbal/ numerical material and figural/spatial material) and a functional facet (separated into the components of storage in the context of processing, coordination, and supervision.
    • A characteristic storage task is a dual task, where participants have to remember words, then perform another task and finally recall the remembered words. This factor is similar to the updating and working memory capacity of Miyake, Friedman, Emerson, Witzki, Howerter, and Wager (2000) and Engle, Tuholski, Laughlin, and Conway (1999).
    • Coordination is the ability to build new relations between elements and to integrate relations into structures (Oberauer et al., 2003, p. 169).
    • Supervision involves the monitoring of ongoing cognitive processes and actions, the selective activation of relevant representations and procedures, and the suppression of irrelevant, distracting ones.
  • The variance explained by working memory components (especially storage in the context of processing and coordination) on Gf was 95% regarding the latent factors. Storage in the context of processing was the best predictor of Gf. It was revealed that coordination is also a significant predictor of Gf. Supervision and the selectivity aspect of attention had only little or no impact on Gf.
  • The excellent global-fit confirmed the structure of working memory found by Oberauer et al. (2003). However, the content factors could not be confirmed. This might be due to the reduced standard deviations and (consequently) lower reliabilities of some working memory tasks.

No comments: