NetLogo Models for Pattern Recognition in Problem Based Learning

Abstract

Problem-based learning and computational thinking, when integrated as a teaching-learning pedagogy, can provide a platform for both teachers and students to explore how to teach and how to learn effectively. Pattern recognition, one of the key aspects of computational thinking, can assist computer science engineers in modeling problem scenarios through accommodation and assimilation. This study proposes a research question to analyze the effectiveness of pattern recognition in modeling the problem scenarios using Schema Theory as a conceptual framework. A multi-method approach was adopted as the research methodology, and self-selection was used as the sampling technique among students who enrolled in the model thinking course jointly offered by industry Knit Space and KLE Technological University. Technology plays a significant role in designing such problem scenarios. For this study, two problems were designed using NetLogo, and reflection points were provided for students to design models for each problem. NetLogo is a platform that offers various simulation models across multiple domains. Both the problems selected emphasized on the pattern recognition. The Paths and Wolf-Sheep models were used for the study. The study analyzed 50 student answer scripts using both qualitative and quantitative methods, after an informed consent to use the data for the research study. Through appropriate descriptive measures and statistical techniques, such as paired t-tests, student feedback, in-vivo coding, and process coding, the collected data was thoroughly examined to derive results and discussion points. Alongside statistical measures, the study also explored themes generated from the findings, with a focus on technology's role in the process. The results align positively with the conclusion that pattern recognition significantly aids in building better models.