Artist’s Statement
Let’s be real— we’ve all been there. That moment when you’re with your friends and someone asks a question that makes your stomach drop because the honest answer? It’s a little too real.
Secret Society was born from how much we— Adi, Ari, Kelly, and Nhu— love these moments: the humor born from the awkwardness and the deep trust and connection that comes from that. We wanted to create a game where that feeling becomes the main event, not something to shy away from. Those spicy prompts asking about your villain origin story or the pettiest reason you’ve ghosted someone? They’re designed to make you squirm a little, laugh a lot, and save your society along the way.
If you’re the type who loves strategy and detective work, you’ll find your home with the Truthers. There’s something so satisfying about picking up on subtle cues – the way someone hesitates before answering, or how they conveniently shift the topic. You might just uncover the secret agendas the Traitors are keeping from you. You’ll get to be a part of something bigger, a team working together to uncover deception threatening your society.
And if you’ve ever felt the rush of getting away with a perfect lie? The Traitors are waiting for you. There’s an art to weaving just enough truth into your story to make it believable, to cause just enough chaos to derail suspicion without drawing it to yourself. It’s mischievous. It’s chaotic. And yes, it’s insanely fun.
What makes Secret Society special isn’t just the gameplay— it’s what happens after. It’s the “I can’t believe you lied about THAT” and the “How did you know I was hiding something?” It’s the shared references that become inside jokes for years, and the surprising truths that make you discover more about your friends.
So gather your friends and get your stories straight. You might share some hilarious lore or lie to their faces. Either way, we promise you’ll leave knowing each other a little better than before in this secret society.
Concept Map
Initial Decisions about Formal Elements & Values
Formal Elements
In the initial phase of our game’s design, our core objective was to create a game that encouraged strong social interaction and strategic gameplay through the aesthetics of fellowship, challenge, and expression. We focused on establishing the following formal elements to structure our game and achieve our goals:
Players
- Multilateral competition (players individually answering prompts and making independent voting decisions)
- Team competition (Truthers and Traitors functioning as united groups encouraged to cooperate because of the shared win objectives within each group)
Rules
- Players take on roles as “Truthers” or “Traitors” in The Truth Consortium
- Truthers must answer prompts truthfully
- Traitors must answer deceptively according to secret agendas
- 30-second speaking time limit per player
- Voting to identify suspected Traitors
Objectives
- Truthers: Answer prompts truthfully and identify Traitors
- Traitors: Complete secret agendas while avoiding detection
- The Oracle: Facilitate gameplay while participating
Procedures
- Setup and role assignment
- Rounds of prompt answering from cards
- Discussion after each round
- Voting and banishment of suspected Traitors
Player Interaction
- Social deduction and observation
- Strategic deception (for Traitors)
- Group discussion and voting
- Response to prompt cards
Resources
- Information (player responses)
- Time (30-second speaking limit)
- Votes
- Truth Seeker Token
- Prompt cards
Conflict
- Direct opposition between Truthers and Traitors
- Dilemmas faced when making decisions
- Hidden information about roles and agendas
- Careful balance between honesty and suspicion
Boundaries
- Game ends when all Truthers are banished
- Game ends when all Traitors are banished
- Game ends when one Truther and one Traitor remain and agenda is guessed (correctly or incorrectly)
Outcome
- Determined by successful identification of roles
- Dependent on completion of secret agendas
- Creates emergent narratives through gameplay
Values
| Values | Description |
| Social connection | The game creates meaningful interaction between players, creating an environment that brings friends and strangers together. |
| Strategic tension | Players must balance honesty with skepticism, creating a tension that drives gameplay forward. |
| Creative expression | Prompt cards encourage both vulnerable honesty and creative deception, allowing players to express themselves. |
| Psychological insight | The game rewards observation and deduction, challenging players to read social cues and behavior. |
| Humor and honesty | “Spicy/Messy” and “Chaotic/Silly” prompt categories create an unserious emotional environment that invites players to be transparent. |
Theoretical Framework
Our game design incorporates several key theoretical concepts:
- Bartle’s Player Types: Truthers function primarily as “achievers” whose goal is to figure out who the Traitors are, while Traitors blend the “explorer” and “killer” roles by deceiving Truthers while completing their secret agendas (Bartle, 1996).
- MDA Framework: We incorporate fellowship through various forms of social dynamics, challenge through deduction, and expression through prompt answers, emergent narratives, deception, and the satisfaction of discovery (Hunicke, 2004).
- Information Economy: The game creates a rich environment of hidden information that players must navigate, similar to games like Blood on the Clocktower.
Overall, we believe our design choices emphasize an environment of social connection, harking on the tension between honesty and deception, creative expression, and balancing humor with moments of vulnerability, all of which we aim to serve as key ingredients to a game that brings friends and strangers alike together. Our prompts encourage both laughter and revelation, and this creates an extremely rich experience for all players that transcends basic competition. This reflects something similar to how Tim LeTourneau views design – creating interlocking systems that respond to player actions and make space for transparent self-expression.
Future Development Considerations
As we continue to develop “Secret Society”, we might consider:
- Adding feedback systems that increase tension as the game progresses
- Introducing more roles with different powers to enrich the “information economy”
- Implementing mechanisms so players eliminated early remain engaged
Testing & Iteration History
Before iterating, each team member wanted to understand the dynamics that emerged upon play. Thus, Adi and Nhu conducted playtest 1, Ari and Kelly conducted playtest 2, and only then did we begin iterating.
Playtest 1 & 2 – April 15 & 17
Playtesters: Mostly strangers with a mix of extroverted and introverted people.
| What Didn’t Go Well? | Why Is This Important? |
| Each round required individual distribution of prompts to players, which was overly time-consuming. | A problematic dynamic emerged of constant game flow interruption, disrupting two important aesthetics:
|
| Players reached a strategic deadlock: non-imposters provided vague answers to protect the prompt, while imposters gave minimal responses to hide their identity. | A problematic dynamic emerged of players optimizing for safety rather than the intended bluffing and social deduction, disrupting three important aesthetics:
|
| Prompts were task-oriented, largely general, and/or impersonal, eliciting brief, simplistic answers that lacked the depth needed for meaningful discussion or imposter detection. | Hinders design patterns for building friendships:
|
| Minimal responses from all players to the prompt made it difficult to identify the imposter during discussion rounds, resulting in low engagement. |
|
| The imposter’s position in the turn order significantly impacted their chances of being discovered, especially when they had to answer first. |
|
| The transparent voting system prevented meaningful collaboration between multiple imposters. |
|
Key Findings
Our key finding was pretty simple: our game’s prompts left people with nothing real to say. Even when the prompts were funny or “interesting,” they were so random and impersonal that players just didn’t have anything to talk about. We kept thinking that if we just sped up how we handed out prompts, or if we wrote “better” ones, it would fix the problem. It didn’t. The issue wasn’t the quality of the prompts—it was the nature of the game. When the questions don’t connect to people’s real lives or experiences, even the best-written prompt falls flat. The result? Every discussion round felt awkward and cold, no matter what we tried.
Comparative analysis made it clear why this was happening. We looked at other games and saw three things we were missing:
- First, prompts need to let people draw from their own lives. When players can pull from personal experience, they have real stories and opinions to share. That’s what keeps conversation flowing (see “We’re Not Really Strangers”, “Jubilee’s Odd One Out”, and “Amigo” from CS 247G Spring 2025).
- Second, you need mechanics that force people to talk. If the game doesn’t require you to speak, you won’t— especially when you’re put on the spot with a random question (see “Filibuster” from CS 247G Spring 2025).
- Third, the game has to force players to interact with each other, not just answer in isolation. The best social games make you talk to other people to move forward (see
We realized that lively discussion isn’t an extra— it’s the main event in a social game. If the structure doesn’t support that, no amount of clever prompt writing or faster game flow will fix it.
How We Iterated & Why
We threw out almost all of our old mechanics and started fresh. We wanted prompts that let people draw from their own lives, and that would spark real conversation. Questions like “Name a song you would play on a road trip” or “Write a headline for a lion escaping the zoo” only led to one-word or generic answers, leaving little to discuss. So, we switched to prompts like, “If your FBI agent saw your recent search history, what would they say?” These gave players more to talk about.
We also needed a mechanic that forced people to talk, so we required everyone to speak for 30 seconds when responding to a prompt. Further, we also wanted players to interact. To do this, we gave Liars secret agendas. Truthers could work together to spot these agendas and immediately eliminate Liars. This mechanic created a dynamic where players were intensely engaged, looking to catch suspicious behavior.
We renamed our game from “Odd Ones Out” to “Realish Friends” since the prompts help players get to know each other, but there are always a few liars in the mix. We used bright, pop colors—light blue, magenta, and yellow—to match the playful spirit of the game and the laughter it brings.
Playtest 3 – April 22
Playtesters: Mostly friends, extroverted people, and only one more reserved player.
| What Went Well? | What Didn’t Go Well? |
| Social interaction grew naturally: By Round 2, players started “yapping” and getting to know each other without needing prompting | Voting materials were inconsistent. Players didn’t all have matching sticky notes and writing utensils, which disrupted the flow. |
| Players self-organized timing: In later rounds, players took ownership of the timer, showing they were invested in maintaining the game’s structure. | 30-second prompt answering felt too long: Players often stalled or “killed time” instead of naturally speaking for the full duration |
| Favorite part was discussion: Players really enjoyed the discussion/debate phase — strong indicator that the core social deduction mechanic is working. | Subtle liar cues weren’t effective. Secret agendas like “scratch your head” were too subtle to create meaningful gameplay moments. |
| Liked knowing who other liars were: This added a layer of coordination for traitors that was satisfying and fun. | Dead players lacked engagement: Once players were eliminated, there wasn’t much for them to do, which hurt momentum. |
| Replayability potential: Players said they’d play with both strangers and friends, meaning the social dynamics worked across different types of groups. | Unclear rules about communication: It wasn’t clear if players could talk to each other during the 30-second answering round. |
| Confusion around role names: “Liar” didn’t feel right because players felt they weren’t technically lying, just blending in. | |
| The possible secret agendas were unclear to the players. They needed clarification that secret agendas could be nonverbal as well as verbal actions. This negatively impacted their ability to detect a traitor. | |
| There was no rulebook for the players to refer to. They sometimes forgot to start the timer or didn’t fully understand mechanics, showing that relying on verbal explanation isn’t enough. Clear, readable rulebook needed. |
Key Findings
At this point in our iteration history, we had finally landed on mechanics that were creating the intended kind of fun for our players. However, the game had a lot of loose ends. These can be categorized into four main key findings.
- Our game needed a more cohesive premise. The game needed a narrative to carry the gameplay dynamics along and enable the players to immerse themselves in their role and mission.
- Our game needed a very set of tools to play with; at this point, we only had cards with prompts and roles. This was a call for us to standardize our game in order to maintain momentum and professionalism.
- The nature secret agendas needed to be solidified. At the time, the players were never instructed that the agendas might be non-verbal. Moreover, we received feedback that non-verbal agendas were too difficult to decipher.
- Dead player re-engagement is an opportunity. Adding mechanics that allow eliminated players to come back (like cracking a secret or achieving a lifeline challenge) could keep everyone involved and add exciting comeback opportunities.
How We Iterated & Why
Moreover, we brainstormed a new premise for our game: Secret Society. A secret society naturally evokes themes of hidden loyalties, secret missions, and suspicion— all of which align perfectly with the mechanics of uncovering Liars, which were now called Traitors to align with the premise. The mysterious, high-stakes atmosphere of a secret society makes the experience feel more immersive and better matches the strategic deception players loved during the discussion rounds.
In hopes of standardizing our game and improving the player experience, we created physical materials, including:
- Role cards (Truther or Liar+Secret Agenda)
- Prompt cards
- Voting slips
- Rulebook
The rulebook was designed to contain all necessary information for players to set up and play the game without any external instruction from the creators.
Based on the feedback we received, we also decided that secret agendas should only be verbal. This makes it easier for Truthers to detect suspicious behavior during discussions, and ensures that gameplay is centered around listening and storytelling rather than subtle body language.
However, although some players struggled to fill 30 seconds during earlier rounds, we chose to keep the 30-second timer. This decision gave Liars enough time to blend in naturally and build convincing answers, while also maintaining a consistent game rhythm.
Lastly, we constructed a new mechanic that allowed banished Truthers to re-enter the game if they correctly guessed a Truther and their secret agenda. This keeps eliminated players engaged, adds new opportunities for strategic comebacks, and increases the tension for Traitors who think they are safe.
Playtest 4 – April 24
Playtesters: 8 mostly extroverted people, all Stanford Co-op residents, everyone knew one or more other people in the game.
| What Went Well? | What Didn’t Go Well? |
| There was lots of table talk and lively debate. During discussions, players were actively pointing fingers, defending themselves, and making arguments, which is the kind of engagement we wanted. | Moderator did not know they had to choose the traitor and truther cards before dealing them out.
This could make them rule the wrong ratio of truther to traitor cards. |
| Players enjoyed the theme and liked the storytelling dynamic of truthers vs. traitors. | Traitors looked at their cards a lot more than truthers since they had to read their secret agenda
This could give away their secret identity to the truthers. |
| The balance of Traitors to Truthers felt good; no one team steamrolled the other and both sides felt like they had a fighting chance. | Players were not sure if they should read the prompt out loud; the rulebook did not specify.
If they don’t read the prompt out loud, then the players would be confused by the response. |
| Replayability felt strong: Players immediately started brainstorming how they would play differently next time. This is a good sign that the game hooked them. | Players weren’t sure if the Truth Seeker should respond to the prompt; the rulebook did not specify.
If the Truth Seekers do not respond to the prompt, then the players would not be able to gather enough information about the other players. |
| A banished truther was able to leverage the accusation rule to banish a truther and re-enter the game. This showed that the new rules in the rulebook were clear and easy to follow. | In the first couple rounds, players would only say a sentence or two instead of using the full 30 seconds.
This made it difficult for traitors to not stand out, since they needed to fulfill their secret agenda. To fix this, the players impromptu created a rule that players must speak for 30 seconds. |
| Players understood the objective of the game and began strategizing within their teams; the traitors sneakily tried to put suspicion onto other players to survive the necessary number of rounds. | Players had to close their eyes many times because the moderator did not know they needed to know the secret agenda.
This really slowed down the momentum of our game during the first round, which is when we want users to really start to get into it. |
| The players got to know each other and develop relationships and dynamics with each other; for instance, some players began bonding and others began being fake petty to each other. | Players said they didn’t expect the questions to be so silly since the color scheme was serious and mysterious (black and white).
This could create a mismatch in expectations vs. delivery with our target audience. |
| There was an instance in which there was a tie in voting, and we found there was a possibility that there could continue to be a tie no matter how many rounds of voting went by.
If players were stuck in this scenario, the rulebook would not specify how to get out of it. |
|
| Occasionally, players really did not like the category of the card that they chose.
This negatively impacted their experience while playing, as they would have preferred to answer a “Spicy” card as opposed to a “Vulnerable” card. |
Key Findings
This last gameplay really helped to iron out the last few kinks of our game. We were able to verify that the kinds of fun we were hoping to create were indeed being created, and that the new mechanics we incorporated into the game flowed smoothly during gameplay. However, we also noticed the places that needed tidying up. These can be delineated as three key findings.
- The first key finding was that the rulebook clarity needs improvement. There were many things that were not explicitly mentioned which confused players. These included but were not limited to:
- How to choose the correct ratio of Truther to Traitor cards before dealing them
- The specific words that the moderator should say to put everyone to sleep, and what they should do during that time.
- Whether the Truth Seeker should read the prompt aloud, and how they should begin the round by answering the prompt themself.
- How long each player should be required to speak for.
- The order of discussion, accusations, and voting.
- What to do if the players voted a tie more than once.
- How many banishments were allowed per round.
- The second key finding was that there were some subtle tells that would tell who the Traitors were to the Truthers. Traitors having to look at their cards more frequently (to remember their secret agendas) could make them accidentally reveal themselves, giving away social cues that undermine the deduction challenge.
- The third key finding was that the visual and tonal components of our game were inconsistent. Players felt a mismatch between the serious, mysterious black-and-white visual design and the silly, lighthearted nature of the prompt cards. This mismatch could confuse or deter players who expect a more serious social deduction experience — or conversely, attract the wrong target audience.
How We Iterated & Why
We made several updates to the rulebook to eliminate ambiguity and ensure smoother gameplay. This included explicitly stating the following:
- The moderator needs to choose the correct number and ratio of cards before dealing them out
- The moderator needs to view the traitors’ secret agendas while the truthers are asleep in the beginning of the game
- The Truth Seeker should read the prompt aloud, and they should begin the round by answering the prompt themself.
- The players should speak for their full 30 seconds.
- There can only be one banishment per round.
- If voting ties more than once, the moderator flips a coin to decide who is banished.
Additionally, there was no clear order about the order of all the different ways that banishment could take place (voting, accusations, traitors who did not fulfill their secret agenda). Based on all the feedback we received, we decided on an explicit order and wrote rules for the moderator to be able to guide the banishment process.
With all of these new specifications, the rulebook grew significantly, so we organized it into distinct sections—Setup and Gameplay—to make it more digestible for players. We also shifted our visual theme to better match the energy of the game. We introduced fun, vibrant colors like purple, red, and blue, while still using mysterious symbols such as eyes and silhouettes to keep the secret society feeling strong. As part of this thematic overhaul, we renamed the moderator to the Oracle and the active speaker to the Truth Seeker to create a more cohesive and immersive secret society atmosphere.
Moreover, we separated the prompt cards into different themed categories. We made it a rule that the categories should be placed separately on the table, and each Truth Seeker may choose the category they feel most confident answering from. This change added an extra layer of player agency and strategy, allowing players to feel more comfortable and expressive during their responses.
Final Playtest
Final Prototype
Print-n-Play
The Secret Society Print-n-Play
Works Cited
Bartle, Richard. “Hearts, Clubs, Diamonds, Spades: Players Who Suit MUDs.” Journal
of MUD Research, vol. 1, no. 1, June 1996,
https://www.researchgate.net/publication/247190693.
Hunicke, Robin, et al. “MDA: A Formal Approach to Game Design and Game
Research.” Proceedings of the AAAI Workshop on Challenges in Game AI, vol.
4, no. 1, 2004, pp. 1-5.
Images Cited
Ament, J. (2019, September 10). They’re Watching You [Artwork]. GNU Blog. https://www.gnu.com/blog/2019/09/10/jennifer-ament-an-artist-for-progress/
Escher, M. C. (1921). Perfume [Canvas artwork]. Kanvah. https://kanvah.com/products/perfume-1921-by-m-c-escher-canvas-artwork
IREFAELS. (n.d.). The Eye Through Which I See [Image]. Pinterest. https://in.pinterest.com/pin/637329784801265260/
Legrand, J. (2020, November 3). Griboos! [Illustration]. Pinterest. https://in.pinterest.com/pin/810366526728201009/
Unknown. (n.d.). Eye Circus T-Shirt [Image]. Pinterest. https://in.pinterest.com/pin/735705289169324757/
Unknown. (n.d.). Untitled [Image]. Pinterest. https://in.pinterest.com/pin/397724210862274790/
