Review of Intrinsic Motivation in Simulation-based Game Testing
summary
Systematic review of intrinsic motivation for simulation-based game testing catalogs implementations, search keywords, exclusion criteria and gaps, and compares methods such as curiosity via prediction-error reward, count-based novelty for sparse-reward games (Montezuma’s Revenge) and empowerment as intrinsic reward. Finds competence, curiosity and empowerment appear in many papers, social relatedness and immersion are underexplored, models are mostly simple optimization agents, most lack validation against human play data, and the authors produce plans for datasets and synthetic “let’s play” videos with facial-expression labels.
AI-written summary · about
chapters (4)
related talks
51mThe alchemy of player motivation
16mThe Role of Uncertainty in Moment-to-Moment Player Motivation: A Grounded Theory
08mThe Effect of Gamification Mechanics on User Experiences of AdventureLEARN: a Self-driven...
06mEchoes of Player Experience: A Literature Review on Audio Assessment and Player Experience in Games
34mGames ≠ Subjective: How to craft engaging experiences by knowing your players
08mPlaying with Emotions: A Systematic Review Examining Emotions and Emotion Regulation in Esports...
03mInvestigating the Implementation and Effects of Personalized Gamification in Education
03mThe Cost of Reward: A Critical Reflection on the `What', `How', and `Why' of Gamification for...