Lay the disagreement on the table,
and see how it gets flattened.
Information cascade (Bikhchandani, Hirshleifer & Welch, 1992) + the Asch conformity experiments (Asch, 1951): once the first person's answer is public, those after them abandon their own private signal to follow—at which point "consensus" is no longer consensus, but a cascade.
A meeting with no disagreement isn't consensus, it's a cascade
① Positioning in one line: Lay the real cognitive disagreements of a team on the table right now—and see how they get flattened.
02The game-theory core
Information cascade (Bikhchandani, Hirshleifer & Welch, 1992) + the Asch conformity experiments (Asch, 1951). Once the first person's answer is public, those after them abandon their own private signal to follow—at which point "consensus" is no longer consensus, but a cascade.
A meeting with no disagreement isn't consensus, it's a cascade.Reference: Bikhchandani et al. (1992); Asch (1951)
03Players / Grouping / Duration / Venue
8–40 players; 20 minutes; indoor + big-screen QR answers.
07Output · Three reporting layers
A team disagreement map (which questions really had disagreement, which were glossed over in a sentence) + cascade strength (how much the first responder shifted others' answers, shown as a percentage).
| Report | For whom | Hard constraint |
|---|---|---|
| Personal report | The individual | Visible to the individual only |
| Team report | Team / HR | Aggregated results only, no names |
| Leader report | Decision-makers | No names, no individual raw five-dimension values |
08Consequence mechanic
Changing your vote is affirmed positively (the changer earns a "correction point"), because changing a position is harder than holding one. This is the game's only reward design.
09Ending tiers
Mirror
≥20% changed back to their own answer after seeing the distribution.
There are still people on your team who don't live by the majority.
Iron Armor
Low vote-change rate, but no one gives a reason.
A silent disagreement isn't a disagreement, it's a hidden risk.
Unseen
Vote-change rate >60%, one voice across the room.
A meeting with no disagreement isn't consensus, it's a cascade—you only heard your own echo.
10Three lines of insight
Being the nth person to speak decides whether you are the you that you think you are.
Disagreement is the most expensive team data; flattening it is destroying data.
One person speaking first can decide the conclusion of the whole meeting—so speaking order is also power.
11Deployment checklist
B1 Fault Lines · Host Rulebook
8–40 players · 20 minutes · indoor + big-screen QR answers · 5 dilemma scenarios
04Rules & steps (with timing)
- 5 dilemma scenarios, each answered anonymously and simultaneously in 30 seconds; the big screen shows the A/B distribution.
- Question 1: fully anonymous (to capture a true baseline).
- Questions 2–4: first have 1 "public first responder" reveal their answer and give one sentence of reasoning, then the room answers.
- Question 5: reveal the previous question's distribution first, then answer (to observe the shift "after seeing the majority").
- After each question, invite two people of differing stances to each give 30 seconds of reasoning, then allow one vote change, and reveal the vote-change rate.
05In-game currency and payout table
| Item | Scope |
|---|---|
| In-game currency | None |
| Outcome measure | Vote-change rate (lower is more independent) |
06Observation points and signal table
| id | Behavior description | Weight |
|---|---|---|
first | Speaks first every time | Courage +2 |
wait | Waits for others to finish before speaking | Connection +1, Courage −1 |
cascade | Switches to the majority after seeing the distribution | Connection +1, Courage −2 |
anticascade | Changes back to their own answer after seeing the distribution | Courage +3, Growth +2 |
hold | Sticks to the original answer after being pressed and gives a reason | Responsibility +2 |
flip | Caves after one round of pressing and can't give a reason | Growth −2, Responsibility −1 |
reason | Says "why I think this" rather than "I just feel" | Wisdom +2 |
silent | Takes no position across all 5 questions | Courage −2, Connection −1 |
Dimensions: Courage / Connection / Wisdom / Growth.
—Host script
- "We're not judging right or wrong, we just want to know who changed, and why."
- The host states behavior only, never rates, never judges personality (the no-right-or-wrong principle).
- A vote changer earns a "correction point"—please affirm every vote change positively and publicly.
07Output · Three reporting layers
A team disagreement map (which questions really had disagreement, which were glossed over in a sentence) + cascade strength (how much the first responder shifted others' answers, shown as a percentage).
| Report | For whom | Content | Hard constraint |
|---|---|---|---|
| Personal report | The individual | Five dimensions / 16 types / six-decision-dimension radar / blind-spot list / ending tier / insight | Visible to the individual only |
| Team report | Team / HR | Team personality map + complementary network + cooperation decay curve + three metrics | Aggregated results only, no names |
| Leader report | Decision-makers | Three charts + five metrics + structural and risk conclusions | No names, no individual raw five-dimension values |