Diagrammatic Reasoning
Riccardo Fusaroli, Kristian Tylén · Pragmatics & Cognition · 2014
representations rather than from word-based representations, is pervasive in our society. Computers allow us to easily design and transmit diagrams that encapsulate a variety of information. Maps are specific instances of diagrams that are used to provide current and projected information (Chandrasekaran et al., 2002); for example, public transportation systems are displayed as color coded graphs. Other examples of diagrammatic reasoning include geometric problem solving in mathematics and free body diagrams in physics. Diagrams can offer cognitive shortcuts relative to verbal descriptions of certain kinds of information, notably relational and spatial information. Thus, diagrams can reduce the working memory load and make possible certain cognitive efficiencies. Challenges The overall goal of this research project is to produce cognitively-congruent models of diagrammatic reasoning in the Adaptive Control of Thought – Rational (ACT-R) architecture (Anderson et al., 2004). This study investigated perception and reasoning during a problem solving task that utilizes a diagram. The model presented here uses the architecture’s perceptual and motor modules; visual objects were created and placed in a virtual window on which the perceptual and motor modules could then act. There were two challenges in building the model: (1) ACT-R’s visual module is text- rather than diagram-based and (2) previous vision modeling efforts (e.g., Fick and Byrne, 2003) have focused on target search where the target is identified rather than using the visual information for subsequent decision making. To validate the ACT-R model, participants performed two diagrammatic reasoning tasks. Diagrammatic Reasoning Tasks