Puzzle solving behaviour represents an important domain within educational research, offering insights into cognitive strategy formation, spatial reasoning, and decision making processes during interactive learning activities. This study investigates cognitive patterns underlying problem solving performance using trajectory based sequence analysis of a 10 coin triangular puzzle exhibited at a museum. Despite the growing use of computational analytics in education, understanding cognitive processes remains challenging as outcome evaluation alone cannot capture strategy. Using behavioural interaction data collected from puzzle solving sessions, the study applies data mining and sequence analysis techniques to examine movement trajectories, coin selection frequencies, and transition structures across solving attempts. The research focuses on identifying critical early stage decisions, high frequency coin usage patterns, and terminal sequence formations that influence solution success or failure. Each coin position is treated as a discrete cognitive decision node to extract dominant behavioural pathways associated with effective or ineffective problem solving strategies. Key research areas include computational detection of critical coin transitions for modelling puzzle solving dynamics, analysis of behavioural insights in psychology oriented puzzle tasks, temporal gap analysis and behavioural pause patterns as predictors of problem solving outcomes, and sequential decision behaviour modelling in grid based cognitive tasks. Temporal cognition is examined through pause interval analysis to evaluate cognitive load variation, uncertainty resolution, and strategic reflection during interaction sequences. This interdisciplinary study contributes to educational data mining and cognitive psychology by providing a computational framework for understanding insight related problem solving behaviour, with implications for learning analytics and cognitive assessment modelling.

Authors:
Subeksha Shrestha, London Metropolitan University, United Kingdom
Wendy Ross, London Metropolitan University, United Kingdom
Arnabi Modak, London Metropolitan University, United Kingdom
Upendo Daudi Manya, London Metropolitan University, United Kingdom
Ganegodage Chathuranga Madushan Dharmarathne, London Metropolitan University, United Kingdom
Prapa Rattadilok, London Metropolitan University, United Kingdom

About the Presenter(s)
Dr Subeksha Shrestha is currently a Deputy Course Leader of MSc in Data Analytics at London Metropolitan University, London, United Kingdom.

Connect on Linkedin
https://linkedin.com/in/subeksha-shrestha

See this presentation on the full schedule – Wednesday Schedule

Bookmark