A New Differential Access Hypothesis: Exploring the Effect of Working Memory Demands on the Validity of Knowledge Acquisition Techniques

Abstract

The differential hypothesis predicts that some knowledge acquisition techniques are most efficient at capturing either declarative or procedural knowledge. The type of knowledge captured may also be impacted by the type of working memory evoked by the knowledge acquisition technique and by the performance task. In the present study, the type of working memory evoked by a knowledge acquisition technique and by a spatial or verbal task were matched or mismatched. It was hypothesized that a match between the type of working memory evoked by the knowledge acquisition technique and by the performance task would improve the predictive validity of models captured by knowledge acquisition. Card sorting and a verbal task were expected to evoke echoic working memory, while conceptual graphing and a spatial task were expected to evoke iconic working memory. The predictive validity of each knowledge acquisition method for each task type was evaluated by calculating correlations between knowledge acquisition data and task errors. Study results indicated that there was a significant positive correlation between card sorting and verbal task errors, and between conceptual graphing and spatial task errors. The correlations between conceptual graphing and verbal task errors, and between card sorting and spatial task errors were not significant. These findings indicate that card sorting was valid for capturing knowledge of verbal but not visual information, while conceptual graphing was valid for capturing knowledge of visual but not verbal information.