[Paper Review] Perspectival Computational Thinking for Learning Physics: A Case Study of Collaborative Agent-based Modeling

Amy Voss Farris, Pratim Sengupta|arXiv (Cornell University)|Mar 15, 2014

Innovative Teaching and Learning Methods15 references18 citations

TL;DR

This study investigates how middle school students develop perspectival computational thinking through collaborative agent-based modeling of motion graphs. By integrating multiple perspectives—physical motion, computational agents, peer collaboration, and graphical representations—the students co-constructed a shared understanding of physics concepts, demonstrating that collaborative modeling fosters deeper conceptual integration through perspective-taking in computational design.

ABSTRACT

We examine the process through which computational thinking develops in a perspectival fashion as two middle school students collaborate with each other in order to develop computational models of two graphs of motion. We present an interaction analysis of the students' discourse and computational modeling, and analyze how they came to a joint understanding of the goal of the modeling activity. We show that this process involves bringing about coherence between multiple perspectives: the object in motion, the computational agent, the other student, and graphs of motion.

Motivation & Objective

• To examine how computational thinking emerges in a collaborative, multi-perspective context during physics modeling tasks.
• To investigate how students negotiate and integrate diverse perspectives—physical motion, computational agents, peer interaction, and graphical representations—during model development.
• To understand the role of discourse and joint problem-solving in achieving coherence across perspectives during agent-based modeling.
• To identify the mechanisms through which students achieve shared understanding in collaborative computational modeling of physics phenomena.
• To contribute to the design of pedagogical frameworks that support perspectival computational thinking in science education.

Proposed method

• Conducted a case study with two middle school students collaborating on agent-based modeling of motion using a visual programming environment.
• Collected and analyzed students' discourse during modeling sessions using interaction analysis to trace perspective negotiation.
• Tracked the evolution of computational models over time, focusing on how changes reflected shifts in perspective integration.
• Mapped the students' modeling decisions to specific perspectives: the object in motion, the computational agent, the peer, and the motion graph.
• Employed qualitative analysis to identify key moments of perspective alignment and conceptual coherence in the modeling process.
• Used the framework of 'perspectival computational thinking' to interpret the students' cognitive and collaborative processes during model development.

Experimental results

Research questions

• RQ1 How do students negotiate and integrate multiple perspectives—physical motion, computational agents, peer collaboration, and graphical representations—during collaborative agent-based modeling?
• RQ2 In what ways does collaborative discourse support the development of computational thinking in a physics context?
• RQ3 How does the joint construction of a computational model lead to a shared understanding of motion phenomena?
• RQ4 What role does perspective-taking play in the emergence of coherent computational models during collaborative learning?
• RQ5 How does the integration of multiple perspectives contribute to conceptual understanding in physics through computational modeling?

Key findings

• Students developed a shared understanding of motion by aligning perspectives: the physical object, the computational agent, the peer, and the graph.
• The modeling process was characterized by iterative perspective negotiation, where students shifted between viewpoints to resolve inconsistencies in the model.
• Discourse played a critical role in aligning perspectives, with students using verbal explanations to coordinate their understanding of agent behavior and graph interpretation.
• Coherence emerged not from a single perspective, but from the dynamic interplay between the physical motion, the agent’s actions, the peer’s input, and the graphical output.
• Students demonstrated perspectival computational thinking by intentionally designing agents to reflect specific viewpoints, such as the agent’s local perception of motion.
• The joint modeling process led to deeper conceptual integration, as evidenced by the students’ ability to explain discrepancies between agent behavior and motion graphs through coordinated perspective shifts.