Animation as Feedback in a Computer-Based Simulation: Representation Matters

The purpose of this study was to explore how users interact and learn during a computer-based simulation given graphical and textual forms of feedback. In two experiments, university students interacted with a simple simulation that modeled the relationship between acceleration and velocity. Subjects interacted with the computer simulation using a discovery-based approach: no formal instruction on the science concepts was presented. Subjects had control over the acceleration of a simple screen object--a ball-in a game-like context. Three simulation conditions were studied, each differing on how feedback of the ball's speed, direction, and position was represented: graphical feedback, textual feedback, and graphical plus textual feedback. Results showed that subjects learned more tacit knowledge when provided with animated graphical feedback than with textual feedback, although gains in explicit understanding of these science principles did not depend on the way the feedback was represented. Patterns of interactivity and frustration are also discussed. 1 Please note that standard units of measurement (i.e., centimeters and seconds) are being used here to help the reader understand the nature of the simulation activity even though there was no attempt to calibrate the simulation to these standards. The speed of the animated object was based on units of distance and time that were unique to the simulations. For example, distance was measured in pixels and time measured in simulation cycles. Of course, all units of measurement are arbitrary conventions and the actual choices for these units do not matter so long as they remain consistent. 2. Obviously, an adequate understanding of the relationship between acceleration and velocity is prerequisite to fully understanding the nature of the task that subjects experienced in this study. An adequate explanation of the underlying physics principles is beyond the scope of this article. Unfortunately, the static medium of a journal article does a poor job of representing such a dynamic system.