The Ultimatum Game, Reimagined: A Serious Game Design
Timeframe: 3 Months
Tools: Unity, Paper sketching
My Role: UX Designer, Game Designer, Usability Tester
Project Overview
This mobile game was developed as part of my master’s thesis, where I led the design and evaluation of an interactive experience exploring how design choices influence player decision-making and emotional responses in the Ultimatum Game (UG).
In the UG, one player (the proposer) is given a sum of money and must offer a share to another player (the responder). If the responder accepts the offer, both players receive the agreed amount. If the responder rejects it, neither player receives anything. The game reveals rich emotional and ethical dynamics around fairness, trust, and decision-making.
I created a mobile Serious Game where player interactions were shaped by two key design variables:
Fairness of offers made by NPCs
Emotional tone (polarity) of NPCs, conveyed through facial expressions and dialogue
Drawing on principles from behavioral economics and affective computing, I designed a game that not only engaged players but also offered insight into the emotional and cognitive factors behind their in-game choices.
Problem Statement
Players often make emotionally driven decisions in social dilemma games, yet these interactions are mainly happening through real-life conversations or hypothetical scenarios. There is a lack of interactive experiences that explore how design factors, such as offer fairness, character behaviour, and facial expressions, influence decision-making and emotional response in moral choice scenarios.
Design Process
Empathize
->
Define
->
Ideate
<->
Prototype
<->
Test
...........
Empathize -> Define -> Ideate <-> Prototype <-> Test ...........
Empathize
The foundation of this project was academic research, not traditional UX user research. Rather than identifying user needs through interviews or observations, the requirements for the game were defined through a rigorous literature review and the formulation of research questions. This meant that the design process was shaped not by personas or user journeys, but by the methodological demands of scientific inquiry.
Research Questions
Building on established research on framing effects, fairness, and emotion in gameplay, I identified two central research questions that would drive both the structure and content of the game:
RQ1: What are the motivating factors for making decisions across player roles in the Ultimatum Game (UG)?
RQ2: How do the interaction parameters of the UG influence the valence (positive or negative tone) and arousal (emotional intensity) of players' emotional responses?
Because the goal was to generate valid and interpretable data to address these questions, the game design was essentially a response to these research objectives. It needed to support a controlled yet engaging experience that varied specific parameters (e.g. offer fairness, NPC expression, dialogue polarity) while allowing naturalistic play. In this way, the game itself became an experimental tool, purpose-built to study decision-making and emotion through interaction.
DEfine <-> Ideate <-> Test
Goals
The game aimed to simulate variations of the Ultimatum Game (UG) to explore the impact of interaction parameters—such as offer fairness and NPC affect—on players' decisions and emotions. However, transforming a behavioural experiment into a meaningful and engaging mobile experience required careful iteration.
Initial Ideas
Initial ideation was guided by behavioural game theory literature and early sketches that mapped UG interactions onto a narrative structure. The decision to embed UG offers within a fruit-collecting game created a rhythm between task and dialogue, avoiding cognitive overload and redundancy.
To evoke emotional responses, several design choices were made:
NPCs were given facial expressions and affective dialogue (positive, neutral, negative) to reflect varying social cues.
Self-reporting of emotions followed each UG interaction, using Ekman’s six basic emotions on a scalable intensity slider.
A perspective switch (third-person to first-person) was introduced to deepen immersion, encouraging players to empathise with the character.
GAME Rules
The objective of the game was to collect five fruits each round. Subsequently, the village master provided the player and another villager with $20 to split. The player was assigned the role of either proposer or responder, while the other villager had the remaining role. In each interaction, the proposer decided on the split. If the responder accepted the offer, the money was split accordingly. If the responder rejected the offer, no one got the money. The money was accumulated in the player’s pouch.
GAME MECHANICS & INTERACTION PARAMETERS
The game was developed using Unity (C#) and built around sequential UG interactions. Through further refinement, the game structure solidified:
17 interactions were selected to balance coverage of conditions (player role, NPC affect, fairness) without causing fatigue.
Each round followed a consistent loop: fruit collection → UG offer → emotion self-report.
Interactions were scripted to simulate fairness dynamics found in real-life negotiations, e.g. rejecting 90:10 offers or being influenced by NPC tone.
Think-Aloud Usability Test
A think-aloud usability study was conducted with 4 participants, which is sufficient to identify 90% of usability issues. During the think-aloud session, participants were encouraged to verbalise their thoughts as they navigated the game, helping me identify any usability issues.
The gameplay session was conducted using an Android tablet. Following the gameplay session, participants were asked to complete a Game Engagement Questionnaire (GEQ). After completing the GEQ, a brief semi-structured interview was conducted, during which targeted usability questions were posed.
This surfaced several usability challenges:
Confusion around the accept/reject buttons (too close, too similar in appearance).
Joystick controls felt unresponsive or nauseating.
Emotion reporting scale needed clarification.
Some narrative elements (e.g. “In this round, I get to propose”) were found redundant.
Players misunderstood the concept of shared money, assuming all money was theirs.
Feedback also pointed to gaps in emotional engagement. Some participants reported feeling disconnected from the characters or confused about game progress.
FInal Iterated Prototypes
The following changes were implemented:
Accept/reject buttons were colour-coded (green/red) and repositioned for clarity.
Dialogue by NPC was made clearer.
Role-switching was made predictable (every 3 rounds), with on-screen indicators.
Emotion reporting instructions were refined to emphasise flexibility in response.
A progress bar and fixed fruit count per round provided clarity on structure.
Joystick controls were made smoother.
The game rule panel was made accessible mid-play for quick referencing.
These refinements ensured the game met its primary function: supporting the collection of emotional and behavioural data for academic analysis, while still maintaining enough engagement to retain participant attention across all interactions.
Reflection
Designing this game offered a unique challenge: creating an interactive system that was both scientifically valid and engaging. As the designer, I developed a deep familiarity with the game’s rules, structure, and underlying logic. I could predict how NPCs would react, what offers made sense to me and how the game would unfold.
In a sense, I mastered it.
But this mastery also made it difficult to see the game from a new player’s perspective. What felt intuitive to me was not always intuitive to others. During testing, I noticed that new players often misunderstood the rules of the Ultimatum Game, became confused by interface elements, or misinterpreted the emotional tone of NPC interactions.
This reminded me of a core principle in game design: players don't enter with your mental model — they build their own.
Most of all, this project highlighted the delicate balance between experimental control and player experience. Every game mechanic had to serve both a design and research purpose, and this required thoughtful iteration and user testing. Bridging that gap pushed me to think like both an academic and a designer, and I wrapped up this project with a stronger appreciation for how subtle interface and interaction details can shape complex human behaviour.
