Eco-Collage is collaborative simulated game designed to help learner understand spatial reasoning principles important for environmental science and urban planning. Learners are confronted with the real-world challenge of effectively placing green infrastructure in an urban neighborhood to reduce surface flooding. Learners could try out different 2D spatial arrangements of green infrastructure on the virtual map and use the simulation to test each solution’s impact on flooding. The learners’ solutions and the solutions’ performances on different competing "rewards" are logged by the system. These could be used to analyze patterns of interactions and compare students' performances. As with many open-ended problem spaces, the solution space is large, i.e. there exist many solutions that could be termed as good with respect to one or more "rewards," and many alternatives that yield similar rewards also exist.
Logged spatial patterns and ecologically inspired metrics were used to device metrics that could guide the comparisons of different spatial- patterns. Ripley’s K measure is an extensively used ecology metric that measures differences in spatial arrangements. We used this metric to quantify learners' spatial reasoning skills and applied regression to characterize the different spatial arrangements in terms of the water infiltrations levels (reward) achieved as good and bad strategies. This work designed a dashboard (see right) that the learners' could use to compare spatial arrangements and corresponding reward outcomes. We used the metrics and the patterns to examine if the user interface design (see left image: how collaboration was mediated with the system) affected the way in which learners approached exploring the problem space: did they use different spatial strategies, or discover them more quickly or more slowly, when using different user interfaces?
Mallavarapu, A.,Lyons, L., Slattery, B., Shelley, T., Minor, E., Zellner, M. (2015). Developing Computational Methods to Measure and Track Learners’ Spatial Reasoning in an Open-Ended Simulation.Journal of Educational Data Mining 7(2), 49-82. PDF