Detecting Player Preference Shifts in an Experience Managed Environment

Anton Vinogradov, Brent Harrison

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


Players are often considered to be static in their preferred play styles, but this has been shown to be untrue. While this is not an issue in games that only attempt to craft a singular experience for the player, this can become a problem for games that use an Experience Manager (ExpM) to adapt the game to the player. As an ExpM observes the player’s actions within the game environment, it makes changes to the environment to bias it towards the player’s preferences, but if a player were to shift their preferences then any further ExpM observations are now based on this biased environment. These biased observations may not be able to show the player’s true preferences and cause the ExpM to form an incorrect player model. In this situation the ExpM should be able to detect these player preference shifts. In this paper we show that such a shift can be detected using only the observations of the player without further intervention. To evaluate these claims we use a text-based interactive fiction environment created to exhibit this bias caused by an ExpM. We compare our own method to detect such shifts to existing outlier detection methods and find that our method can detect shifts in player preferences.

Original languageEnglish
Title of host publicationInteractive Storytelling - 16th International Conference on Interactive Digital Storytelling, ICIDS 2023, Proceedings
EditorsLissa Holloway-Attaway, John T. Murray
Number of pages15
StatePublished - 2023
Event16th International Conference on Interactive Digital Storytelling, ICIDS 2023 - Kobe, Japan
Duration: Nov 11 2023Nov 15 2023

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume14383 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference16th International Conference on Interactive Digital Storytelling, ICIDS 2023

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.


  • Experience Management
  • Player Modeling

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Detecting Player Preference Shifts in an Experience Managed Environment'. Together they form a unique fingerprint.

Cite this