Beyond the Ball-and-Stick: Students’ Processing of Novel STEM Visualizations

• In Event: Cognitive Processes in Comprehension of Visual Representations: Art, Diagrams, Graphs, and Models

Abstract

Objectives or purposes
This interdisciplinary experiment explored student use of Electrostatic Potential Maps (EPMs, Figure 1), a type of visualization that is enjoying growing support in chemical education. We sought to determine if EPMs are useful to students in chemical reasoning, if and how students come to rely on EPMs during learning, and whether the use of EPMs depends on prior knowledge or cognitive abilities.
Perspective(s) or theoretical framework
Visualizations are not automatically converted into knowledge, and learners differ in the degrees to which they comprehend and rely on them (Rapp & Kurby, 2008). In contrast to the popularity of visualizations in STEM fields (e.g. Phillips et al, 2010), there is little research on whether and how students attempt to use them during learning experiences. To inform both design and instruction, it is crucial to know (a) the nature of individuals’ unfolding interactions with visualizations, (b) whether cognitive abilities mediate their use, and (c) whether different types of scientific concepts are differentially supported by visualizations.
Methods, techniques, or modes of inquiry
We took a multi-method approach to this experiment utilizing eye-tracking, verbal explanations, and responses to problems. Prior knowledge and cognitive abilities measures served as potential moderators.
Data sources, evidence, objects, or materials
Thirty first-semester organic chemistry students were recruited as high-prior--knowledge (HPKs) or low-prior-knowledge (LPKs) based on a chemical pre-test. After a short introduction to EPMs, students viewed 24 randomly presented molecules simultaneously represented by both EPMs and ball-and-stick visualizations. For each molecule, participants answered two interpretation questions and two application questions, which required inferences beyond simply mapping the visualization’s depiction to features of the molecule. EPMs should be useful for both types of questions.
 
Results and Conclusions
Accuracy (Figure 2). Application questions were significantly more difficult than interpretation questions, and LPKs performed significantly worse overall than HPKs. These main effects were qualified by an interaction which suggested that the advantage for HPKs was most apparent on the more difficult application questions.
Eye Tracking. Relative visualization use was calculated based on total looking times for the two representations. Suggesting that EPMs are useful for reasoning about the chemistry questions, greater attention to EPMs was significantly associated with higher accuracy (r=0.60).
Representation use also varied as a function of prior knowledge and question type (Figure 3). LPKs relied more on ball-and-stick representations than HPKs, and both HPKs and LPKs relied more on ball-and-stick representations for the application questions than for the interpretation questions.
Figure 4 shows representation use across practice trials. For both types of questions, significant interactions indicated that, while HPKs and LPKs both initially relied on the familiar ball-and-stick representations, only HPKs learned to rely more on the novel EPMs over the course of the experiment.
Scientific or scholarly significance of the study or work
While visualizations can foster learning, the effective use of novel visualizations proves neither automatic nor universal. Only by understanding when, how, and why different learners are able to comprehend complex visualizations will we be able to fully realize their pedagogical potential.

Authors

• Scott R. Hinze, Northwestern University

• David Rapp, Northwestern University

• Vickie Williamson, Texas A&M University

• Mary Jane Shultz, Tufts University

• Kenneth C. Williamson, Texas A&M University

• Ghislain Deslongchamps, University of New Brunswick