Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Gamifying Compassion: Mitigating Dialect Prejudice Through An AI-Driven Serious Game

Sicheng Lu, Erick Purwanto, Hong Liu, Aini Li, Adel Chaouch-Orozco

Abstract

Dialect bias is pervasive yet often unconscious, normalized, or obscured by masking. Existing HCI interventions primarily audit disparities and propose reactive fixes. We present CompassioMate, a dialect-aware serious game that nurtures perspective-taking through AI-mediated play. Players listen to audio samples to identify regional dialects, engage in simulated social interactions involving dialect discrimination, and explore branching narratives that reveal how changes in wording or stance can influence the outcomes. In a three-week field study with 20 university students, participants reported feeling comfortable when observing region-tailored dialogues; several described experiencing perspective change. We contribute: 1) a formative study identifying goals for safe action consequence modelling, 2) the design and evaluation of a serious game integrating dialect audio, region-mapping play, bias; and 3) design implications highlighting listener-side training, transparent evaluation, and narratives maintaining psychological well-being.

Gamifying Compassion: Mitigating Dialect Prejudice Through An AI-Driven Serious Game

Abstract

Dialect bias is pervasive yet often unconscious, normalized, or obscured by masking. Existing HCI interventions primarily audit disparities and propose reactive fixes. We present CompassioMate, a dialect-aware serious game that nurtures perspective-taking through AI-mediated play. Players listen to audio samples to identify regional dialects, engage in simulated social interactions involving dialect discrimination, and explore branching narratives that reveal how changes in wording or stance can influence the outcomes. In a three-week field study with 20 university students, participants reported feeling comfortable when observing region-tailored dialogues; several described experiencing perspective change. We contribute: 1) a formative study identifying goals for safe action consequence modelling, 2) the design and evaluation of a serious game integrating dialect audio, region-mapping play, bias; and 3) design implications highlighting listener-side training, transparent evaluation, and narratives maintaining psychological well-being.
Paper Structure (52 sections, 5 figures, 2 tables)

This paper contains 52 sections, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Accent and Dialect Bias Mitigation in HCI
  4. Empathy and Intervention in HCI
  5. Understanding Accent and Dialect Bias
  6. Research Aim
  7. Formative Study
  8. Initial Design
  9. Methods
  10. Participants
  11. Setup
  12. Procedures and Data Collection
  13. Data Analysis
  14. Findings
  15. Game Mechanics Guide Players to Confront and Reflect on Bias
  16. ...and 37 more sections

Figures (5)

  • Figure 1: Overview of CompassioMate. a) The game's core design is grounded in the four root causes and three manifestations of accent bias that we identified. b) Level 1's core gameplay mechanics challenge players to identify and recognize bias triggers. c) Gameplay in Level 2 requires players to analyze and differentiate reasons for bias. d) Gameplay in Level 3 challenges players to rewrite dialogues to intervene and resolve conflicts.
  • Figure 2: Visual examples of the game’s interactive world and its character-driven narrative
  • Figure 3: The game's AI system is comprised of two core components: the AI Panda Guide, which provides objective feedback, and the AI-driven NPCs, which enable players to gather clues and understand characters' inner thoughts.
  • Figure 4: A branching narrative in Level 3, showcasing three possible different endings.
  • Figure 5: Overview of Local AI Interactions in CompassioMate (see supplementary for predefined prompt templates).