Procedurally generating rules to adapt difficulty for narrative puzzle games (2307.05518v1)

Abstract: This paper focuses on procedurally generating rules and communicating them to players to adjust the difficulty. This is part of a larger project to collect and adapt games in educational games for young children using a digital puzzle game designed for kindergarten. A genetic algorithm is used together with a difficulty measure to find a target number of solution sets and a LLM is used to communicate the rules in a narrative context. During testing the approach was able to find rules that approximate any given target difficulty within two dozen generations on average. The approach was combined with a LLM to create a narrative puzzle game where players have to host a dinner for animals that can't get along. Future experiments will try to improve evaluation, specialize the LLM on children's literature, and collect multi-modal data from players to guide adaptation.