The Talos Principle - Designing for Player Error

This blog post was originally published on my Medium.

Player errors are important. Much of game design is making decisions about error space. Do you expand the error space, like with hazards in platformers? Do you contract it, like with aim assist in shooters? Platformers like Uncharted might not need hazards, and, likewise, shooters like Counter-Strike don’t use aim assist. The decision can seem straightforward in many situations, but it requires a greater understanding of what your game should be.

Take Portal. The main conceit is that you can only open portals on white surfaces. Then what if every wall, floor, and ceiling in the puzzle is white? It may be interesting from a conceptual standpoint, but from the player’s perspective, it’s distracting. How do they know which surface is important? The level design lacks direction because there’s too much positive space, not enough negative. From an error point of view, there are too many inconsequential mistakes.

Contract error space by using non-white surfaces and targets to aim at [Portal 2007]

The Talos Principle, on the other hand, is perfectly OK with large error spaces. The reflectors that are used to redirect the light can be placed anywhere on the ground. The player can immediately grasp their simple functionality, so their error design is for the player to learn how to spot ideal locations. The game even goes to great lengths to ensure that the player can stay focused on this task. The most pertinent quality of life is that when the player needs to put a reflector on top of a box, the reflector snaps to the top of it. They never need to worry about the reflector falling off (bonus points for never needing to worry about knocking reflectors over).

The HUD shows what would be connected before the player places the reflector [The Talos Principle 2014]

The assisted jumps work the same way. The "footprints" signal indicates to the player that if they press jump, the game will “jump” for the player and place the avatar right where the signal is. Sure, the game could have let the player do their own platforming. After all, the avatar can already jump, so why not just let the player jump unassisted? Jumping onto a platform is intuitive, it gives play to the controls, and platformers are a popular genre. Why not?

The footprints signal an assisted jump [The Talos Principle 2014]

The Talos Principle is purely a puzzle game. The types of errors the game wants the player to make are mistakes--not slips. The difference, as explained in Don Norman’s The Design of Everyday Things, is that “a slip occurs when a person intends to do one action and ends up doing something else [italics added]”. This is like if the player wants to jump on a box, but they end up falling short and missing it entirely. On the other hand, “a mistake occurs when the wrong goal is established or the wrong plan is formed [italics added]”. This is like if the player wants to jump on the box to get somewhere higher, but it does not lead them to the puzzle’s solution. The player’s intention is key here. If they want to jump on something, the game will do it for them with no error. The assisted jump assures that no slippage occurs. Without getting distracted by platforming, the player can stay focused on their mistakes and solve the puzzle.

Puzzle games are about players correcting mistakes. Action games are about correcting slips. This is one reason why the two genres do not exist on the same continuum. Designing your game’s error space is crucial to the player’s experience. What kinds of mistakes can the player make? What kinds of slips? What is your game about? There is no right answer, only a well-designed one.

-- Stephen Trinh (@stephentrinha)