Designing for Embodied Interaction: Integrating Studies of Body Actions That Support Scientific Reasoning

• In Event: 65.041 - Design-Based Research: Coming of Age for a Versatile Methodology Supporting Diverse Research and Learner Needs

Abstract

Objectives
In this design-based research (DBR) project we are developing and refining enhancements to existing science simulations such that they allow learners to control simulation parameters using embodied interactions (e.g., hand gestures). Computer simulations have several affordances for science learning (Lindgren & Schwartz, 2009; National Research Council, 2011) and recent work has shown additional benefits of incorporating body movement and physical interaction with digital simulations (Black et al., 2011; Johnson-Glenberg et al., 2014). Unfortunately, learning technology designs are often driven by current technological capabilities rather than learning theory or research findings. In this project we precede design and implementation with focused studies seeking to identify units of behavior associated with successful scientific reasoning—what we are calling Embodied Explanatory Expressions (EEEs)—that can be fed directly into the interface design of simulations covering topics such as molecular interactions and heat transfer.

Theoretical Perspectives
DBR methodologies call for initial designs that are rooted in theory and a robust review of the relevant research literature (e.g., Hoadley, 2004; The Design-Based Research Collective, 2003). DBR methodologies specifically targeting learning technologies call for designs that are pragmatic and grounded with respect to theories of how people learn and the specific contexts within which the technologies are implemented (Wang & Hannafin, 2005). Given the particular type of technology we are designing for—interactive simulations accepting gesture as input—we interpret this call as requiring initial studies that result in findings at the appropriate level of specificity such that they can be integrated directly into a prototype interface and assessed for its effectiveness.

Methods
Our approach to identifying EEEs that can be leveraged effectively in simulation design involves conducting clinical interviews with middle school students who are prompted to reason about science phenomena and encouraged to use their hands to support their reasoning. Specifically we utilize interview protocols designed to examine learners’ explanatory models (e.g., Cheng & Brown, 2010). For example, the participant in Figure 2.1 (see consolidated document) was prompted to explain why the piston in a compressed syringe filled with gas will push back out when released. The participant uses her hands to represent gas molecules impacting the piston and causing increased pressure.

Data sources
Frequently occurring EEEs are being extracted through a process of video coding for learning interactions (Derry et al., 2010). Video events are coded on 2 dimensions: the physical characteristics of the embodied expression and (2) the role that the expression plays in the student’s explanation.

Results
Preliminary data collected reveals the occurrence of embodied expressions that have strong potential for incorporation into interactive science simulations, such as the action being performed in Figure 2.1 being used as way to increase the number of molecules present in a Gas Law simulation.

Scholarly significance
This project models a DBR process that meets the criteria of being grounded and pragmatic by executing the focused studies necessary to identify design-ready embodied interactions.

Authors

• Robb Lindgren, University of Illinois at Urbana-Champaign

• David E. Brown, University of Illinois at Urbana-Champaign

• Daniel L. Hoffman, Kamehameha Schools