Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

From Frustration to Fun: An Adaptive Problem-Solving Puzzle Game Powered by Genetic Algorithm

Matthew McConnell, Richard Zhao

TL;DR

This work introduces the Adaptive Problem-Solving Game (APSG), a Unity-based system that uses genetic algorithms to generate real-time, pathfinding-based puzzles tuned to individual players via a flexible player-modeling framework. The GA encodes puzzles on grids, applies column-based crossover with repair steps to maintain solvability, and uses a weighted fitness function to map puzzle features to a 1–10 difficulty scale. A pilot study compares Standard, Increasing, and Time-based adaptive modes, finding that time-only adaptation underperforms and that a richer, analytics-driven approach yields better perceived difficulty and progression, with frustration reductions across modes. The results support the practicality of integrating PCG and online adaptive difficulty for engaging educational games and point to future work on emotional-state metrics and broader educational applications.

Abstract

This paper explores adaptive problem solving with a game designed to support the development of problem-solving skills. Using an adaptive, AI-powered puzzle game, our adaptive problem-solving system dynamically generates pathfinding-based puzzles using a genetic algorithm, tailoring the difficulty of each puzzle to individual players in an online real-time approach. A player-modeling system records user interactions and informs the generation of puzzles to approximate a target difficulty level based on various metrics of the player. By combining procedural content generation with online adaptive difficulty adjustment, the system aims to maintain engagement, mitigate frustration, and maintain an optimal level of challenge. A pilot user study investigates the effectiveness of this approach, comparing different types of adaptive difficulty systems and interpreting players' responses. This work lays the foundation for further research into emotionally informed player models, advanced AI techniques for adaptivity, and broader applications beyond gaming in educational settings.

From Frustration to Fun: An Adaptive Problem-Solving Puzzle Game Powered by Genetic Algorithm

TL;DR

This work introduces the Adaptive Problem-Solving Game (APSG), a Unity-based system that uses genetic algorithms to generate real-time, pathfinding-based puzzles tuned to individual players via a flexible player-modeling framework. The GA encodes puzzles on grids, applies column-based crossover with repair steps to maintain solvability, and uses a weighted fitness function to map puzzle features to a 1–10 difficulty scale. A pilot study compares Standard, Increasing, and Time-based adaptive modes, finding that time-only adaptation underperforms and that a richer, analytics-driven approach yields better perceived difficulty and progression, with frustration reductions across modes. The results support the practicality of integrating PCG and online adaptive difficulty for engaging educational games and point to future work on emotional-state metrics and broader educational applications.

Abstract

This paper explores adaptive problem solving with a game designed to support the development of problem-solving skills. Using an adaptive, AI-powered puzzle game, our adaptive problem-solving system dynamically generates pathfinding-based puzzles using a genetic algorithm, tailoring the difficulty of each puzzle to individual players in an online real-time approach. A player-modeling system records user interactions and informs the generation of puzzles to approximate a target difficulty level based on various metrics of the player. By combining procedural content generation with online adaptive difficulty adjustment, the system aims to maintain engagement, mitigate frustration, and maintain an optimal level of challenge. A pilot user study investigates the effectiveness of this approach, comparing different types of adaptive difficulty systems and interpreting players' responses. This work lays the foundation for further research into emotionally informed player models, advanced AI techniques for adaptivity, and broader applications beyond gaming in educational settings.

Paper Structure

This paper contains 20 sections, 1 equation, 12 figures, 3 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Related Works
  3. Methodology and Implementation
  4. APSG as a Puzzle Game
  5. Genetic Algorithm and Puzzle Generation
  6. Data Representation
  7. Crossover Function
  8. Fitness Function
  9. Adaptive Difficulty and Player Modeling
  10. Puzzle Evolution and Metrics
  11. User Study Results and Discussion
  12. User Study
  13. Results and Discussions
  14. Frustration (RQ1)
  15. Difficulty and Progression (RQ2)
  16. ...and 5 more sections

Figures (12)

  • Figure 1: Example of a difficulty-5 puzzle.
  • Figure 2: Example of a difficulty-5 puzzle with a solution path drawn by the player.
  • Figure 3: Example of a solved difficulty-5 puzzle, where all the cargo boxes are dropped off in the correct destinations.
  • Figure 4: Difficulties 1 (top left), 5 (bottom) and 10 (top right) puzzles.
  • Figure 5: Example of large grid puzzle.
  • ...and 7 more figures