Friday, May 20, 2005

Working memory, domain knowledge and higher-level cognitive performance

Hambrick, D. Z., & Oswald, F. L. (2005). Does domain knowledge moderate involvement of working memory capacity in higher-level cognition? A test of three models. Journal of Memory and Language, 52(3), 377-397.

What makes a person more competent on higher-level cognitive tasks? A larger reservoir of domain-relevant knowledge (software), larger working memory capacity (hardware), or the interaction of the two? Inquiring minds want to know. Contemporary cognitive research has implicated both working memory capacity (Gsm-MW) and domain knowledge (Gk) in higher-level cognition.

In their research article, Hambrik and Oswald investigated the interplay between Gk and MW. According to the authors, this study represents one of a few studies that have investigated the interplay between MW capacity and depth of Gk. The authors characterize MW and Gk in the following manner:

  • “Working memory capacity might be thought of as a stable component of higher-level cognition---a possible ‘hardware' aspect of cognition”
  • “Domain knowledge…might be thought of as a modifiable ‘software’ aspect of cognition”

Predictions regarding higher-level cognitive performance in a relatively large sample (n=381…yes…you guessed it…undergraduate students in an introductory psychology course) were based on three competing hypothesis.
  • Compensation hypothesis – Gk will attenuate the influence of MW capacity on higher-level cognition. In other words, greater Gk will reduce or eliminate the influence of MW capacity on domain-relevant tasks – there is an interaction effect.
  • The rich-get-richer hypothesis – larger MW capacity will enhance the use of Gk on higher level cognitive tasks. “In other words, people with high levels of working memory capacity tend to benefit from domain knowledge more than those with lower levels.”
  • The independent influences hypothesis – MW and Gk make independent “additive” contributions to higher-level cognition
The participants performance was measured on two tasks. One required tracking information in a dynamic environment (tracking the movements of baseball players on a baseball diamond) while the other was structurally isomorphic with arbitrary content (spaceship task).

The results revealed “greater use of baseball knowledge in the baseball task than in the spaceship task. However, even at high levels of baseball knowledge, this knowledge use did not alter the relationship between working memory capacity and task performance. This finding is inconsistent with compensation and rich-get-richer hypotheses. Instead, it suggests that working memory capacity and domain knowledge may operate independently under certain conditions”

Bottom line practical implications
  • Hardware (the working memory capacity an individual brings to a specific task) and software (domain-relevant knowledge an individual brings to a specific task) combine to facilitate higher-level cognitive performance. Simply having a larger CPU (MW) is not the answer. Having the most sophisticated software (knowledge base) is insufficient. Both hardware and software make unique contributions to performing at a high level on demanding cognitive tasks. At least in adult populations, both domains need to be assessed when trying to predict/explain a person’s performance.

No comments: