The Role of Representation and Working Memory in Diagrammatic Reasoning and Decision Making

Abstract

In this chapter we present research into the role of visual and verbal working memory in visual reasoning. Eighty subjects participated in an experiment where 34 different gain-loss problems were represented in either text or graphical form. In order to test the role of each component of Baddeley's model of working memory, subjects performed secondary verbal, visual, and mental suppression tasks while reasoning about the problems represented in text or graphics, yielding six different conditions. In two control conditions, no suppression tasks were performed. Interference and preference reversals occurred in all six conditions involving suppression tasks, even though no interference was predicted in conditions involving the graphical representation coupled with the verbal suppression task and the text representation coupled with the visual suppression task. In the control conditions subjects made responses consistent with prospect theory with little interference. Response times were consistently slower in the four text conditions compared to the four graphics conditions. The visual and mental secondary tasks resulted in increased response times, respectively, but the verbal secondary task did not. The data suggests that certain graphical and text representations require both visual and verbal resources and the taxing of these resources results in perceptual-cognitive biases that sometimes favour a minimal inference strategy