Abstract:

Agent Based Modelling (ABM) is an area of computational study that constitutes studying emergent phenomena, via agents that display autonomous interactions (inter and intra) based on a given ruleset. The core of such a simulation lies in the simplistic rulesets the drive the autonomous behavior of the agent. The current advancements in this field have lowered the barrier to entry for novice learners of ABM, however once the scale (and complexity) of the problems start increasing and learners explore more complex questions, it can become quite a challenge to ensure that the fundamentals are being adhered to. This research presents a workflow to develop a learning tool (“LearnABM”) aimed to preserve the low barrier of entry to ABM education and facilitate manageable scalability (in terms of complexity) for novice users to transition into intermediate users. The workflow is comprised of three stages of simulation – a basic cellular automaton simulation (Conway’s Game of Life) to represent medium complexity, an implementation of a traditional ABM (Schelling’s Model of Segregation) to represent low complexity, and a multi-agent simulation (by modifying the base Schelling’s Model) to represent high complexity. Participants then construct various test cases in the simulation and record their observations in terms of system behavior using the Cognitive Learning Scale (pretest and posttest) with respect to its alignment with the goal of ABM and rank their perceived complexity as a learning tool. The results displayed significant improvement in higher order cognitive learning dimensions and diverse topic exploration approaches when using the tool. Section 2 provides the investigation methodology and provides insight into the scaffolding nature of the three simulations. Section 3 presents the results of the experiment. Section 4 highlights the limitations of the current research. Finally, Section 5 presents the conclusion and discussion on LearnABM’s future.

Keywords:

Agent based modelling, cellular automata, complexity simulation, Schelling’s Model, education theory, inquiry-based learning.