Civilization Creator Explains Why Everything Game Devs Know Is WrongS

Firaxis Games co-founder and Civilization creator Sid Meier expanded the minds of GDC attendees by talking about the psychology of game design and breaking some bad news to us. Everything we know, Meier argues, is wrong.

Meier's keynote at this year's Game Developers Conference, titled "The Psychology of Game Design (Everything You Know Is Wrong)," touched on the concept that hard facts, math, science, and historical research are not the foundation for a successful video game. Understanding the psychology of the player is the most important thing.

Gameplay, he says is a psychological experience. It's all in our heads. And it makes us egomaniacs. If you play Meier's Civilization series, in which you're a god-king tasked with building a society that stand the test of time, you're an egomaniac.

Another facet is what Meier calls The Winner Paradox, explaining that while sports leagues like the NFL and NBA have only one Super Bowl winner or NBA championship winner, in games, everyone is supposed to win.

Meier touched on the balance of reward versus punishment for players, a mechanic in which players feel good, sometimes intelligent that they've beaten a system. Similarly, game designers have to be cognizant of the psychological impact of making a player suffer a loss.

Difficulty levels help explain that argument, that players feel rewarded by their ability to overcome increasing challenges. To help prove this Sid points out that Civilization IV features an impressive nine difficulty levels. Everyone wants to be above average, Meier concludes.

Sid's next argument, his trademark-worthy (he believes) Unholy Alliance theory, the bond between player and game designer. He talks about the concept of flight simulators, that players feel good about themselves based on their ability to pilot a complex piece of machinery. Players need to also suspend their disbelief, in both movies and video games, something that game designers need to factor into their work.

Meier talked about the limitations of suspending the player's disbelief with nothing more than a 16-color display.

Players also need a sense of moral clarity, Meier said. He believes it's more satisfying to win, for example, against a vicious character like Genghis Khan than it is a sympathetic enemy, like a destitute lone village filled with women and children.

The "unholy alliance" with players also includes the use of style. Meier defines that as a sense of consistency, that designers can't have a cartoon world that suddenly turns violent, or vice versa. That's when the player loses their suspension of disbelief and moves on. The only way to keep a player in the game is to make use of humor, style, music and atmosphere in a consistent, homogenous way.

Meier's background is in math and science and the mathematical rules don't always translate well to player expectations. The example Meier provided was from Civilization Revolution, that a team of 1.5 versus a team of 0.5 is expected by the player to win every time, as opposed to the 0.5 battle participant winning one time out of four.

That's math upsets player expectations, that a 20 versus 10 battle does not necessarily equate to a battle of two-to-one. Players find that the psychologically of the math does not always match the psychology of the gameplay.

Meier talks about his own mistakes—his "My Bad." moments. One of them was originally making Civilization as a real-time strategy game. Another was to include the idea of "rise and fall."

Wouldn't it be cool to have this moment where you've risen to the top, then fall, then come back from that crumble and rise again, Meier though. No. Players would simply reload a previous save once their society started to fall. The draw for players was to rise, re-rise and rise further.

The next "bad" was the Civilization tech three, that the carrot on the end of the tech tree stick—basically the evolution of weapons of war, from gun powder to nuclear weapons—was what players strove to achieve. After discovering gun powder in the game, players simply wanted to find out how to get gun powder faster in their next play through.

Meier says there's a bit of paranoia from players who experience randomness. He says that players feel that the computer is sometimes "out to get them."

One title that Firaxis is working on is Civilization Network, which taps into all types of gameplay changes. One is the ability to let players exchange gold between civilizations. "There would be this fascinating diplomacy dynamic in the game," the team theorized.

That never happened, Meier said, that players never generously gave gold to other less fortunate players.

On the concept of making "AAA" games on a shoestring budget, Meier says that designers should use the player's imagination. "We don't have to literally show players everything cool," he argues. Let them fill in the holes for themselves.

He says that designers should also "go with the flow." Players will be inclined to agree with the concepts that games introduce, as in the Civ Rev example of foreign diplomacy. He shows on screen a text box that says that foreign rulers seek your favor, discussing the exchange of exotic treasures, like a gift of dancing bears.

There are no dancing bears in Civilization, but the player envisions them as components of the game without having to see anything more than a text box.

Also, tap into what players already know, Meier says. Using a familiar theme, like pirates, already evokes a familiar aesthetic, that players can identify with the roles, characters and designs of things they already know.

Meier says another facet of psychological game design is the role of artificial intelligence. Players want the AI to feel like a person, for the AI to outsmart them, for it not to appear dumb. If the AI is doing something brilliant, however, players may suspect that the AI may have "peeked behind the curtain" and cheated.

That breaks a player's suspension of disbelief.

Artificial intelligence is an "improvement metric," a benchmark for the player's own abilities, something to strive to overcome. AI also provides feedback to the player, letting them know that they're improving, that they're understood by the system.

Meier moves on to talking about protecting the player from themselves. On load and save issues, Meier has observed that players are very cautious with their saves, pausing to save their progress before every battle, reloading if they save. That, Meier said, doesn't help a player build a better strategy, a handicap of sorts.

That's not the way Civilization was designed to be played, he said. In Pirates! the team limited the ability to save the game only at ports, trying to prevent that type of overly cautious saving.

Cheat codes can also damage the gameplay experience for themselves, he says. One Civilization game had a cheat option from the main menus, something that Sid didn't agree with, a player's ability to bring tanks into the middle ages. Meier wanted that option buried, so as not to damage the core Civilization experience.

But modding, Meier says, is a "cool thing."

As a designer, Meier says that game developers need to listen to the player, "what they're really saying." It's our jobs, he says, to understand what causes negative emotions in gamers and strengthen what inspires positive emotions.

What is the point of all this? Meier says that developers are trying to create "the epic journey." How do we use psychology to make the journey more epic, he wonders. One, interesting decisions, a term that Meier says that are the type that encourage players to envision the future, to contemplate what alternate paths they can take. Learning and progress is fundamental to that journey, that players must feel that they're evolving, moving forward.

In "the epic journey," players should be drawn into the "one more turn" phenomenon curious about what's just around the next corner. Likewise, the lure of replayability also helps to make the epic journey in any game the coolest of all time.

Now, Meier says, we know everything. So go make some epic journeys, folks!