Procedual content generation that fits the player

This study examines how it is possible to create a game that uses procedural content generation to create personalized levels aimed for the three player types: Immersion-Exploration, Action-Social and Achievement-Mastery, as defined by Yee and Ducheneaut (2017), by using quantitative data gathered from the players’ behaviour. Research on procedural content generation, user- centred design and different player-types, their expectation of video games and how to measure desire to continue, have been done for in the analysis of the project.

A/B split testing

Two prototypes have been created to evaluate on the problem statement. One which takes user behaviour as input (A), and one which does not (B). The hypothesis is that if players mainly interacts with the game elements that have been designed towards their player type branch, they will have, what we call, specialized. If they specialize more in prototype A than B, the implementation of user behaviour as input, is necessary to withhold higher grade of desire to continue from all three player types who are playing the same game.