(The title of this article is a reference to an expression often uttered by bewildered players in "The Crystal Maze" - a television series broadcast in the UK in the 1990s that is considered a precursor to the modern escape room game format).

Escape Room (ER) games contain a variety of puzzles and tasks that players must complete within a themed environment, with the aim of achieving some ultimate objective (often, but not always, to "escape the room"). Unlike in some games, where the rules are explicit and fairly constrained, much of the thrill of an escape room is often derived from exploration and experimentation; players are given freedom to discover puzzle elements, deduce their significance and function, and determine the effect of using them in different ways.

("Time Travel" mobile escape room by Crux Scenica. Image used with permission.)

In common with other sorts of puzzle games, a good ER game is one that makes its players feel smart. To feel smart, players need to be given choices and the freedom to make decisions - the result of which should not be obvious or trivial, but nor should it be unobtainable. Delight can result when players take a chance on "it’s a long shot, but it might just work!", and their risk pays off with the intended reward.

In this article we’ll consider how to provide the "sweet spot" of guidance to the player - enough information that they are able to progress without becoming stuck or frustrated, but not so much that they don’t feel challenged.

Raph Koster, A Theory of Fun for Game Design (2004)

Identifying Puzzle Elements

One of my favourite game design definitions is taken from Bernard Suits' The Grasshopper (1978):

Using the language of this definition, in order for a player to progress through a game they must first identify the obstacles that are blocking them from achieving their goal - i.e. the puzzles that must be solved and tasks that must be completed. The first generation of escape rooms were fairly basic in design, consisting of a few boxes in a minimally-styled room, locked with padlocks. The obstacles in this case are clear. However, as escape room designers have begun to create ever more ambitious and more immersive rooms, the act of identifying what constitutes a puzzle element has become less obvious. Some escape rooms now feature scenic design, lavish backgrounds and elaborate props comparable to that on a movie set. While these elements assist in creating immersion and transporting the player to another time or place, they don’t have any purpose in the puzzle flow. How then do players identify what is a puzzle element, and what is just set dressing?

ER players enjoy discovery, and escape rooms often involve undirected search activities - akin to the player of a point-and-click adventure game scanning the mouse cursor across the screen until finding an object or game element with which they can interact. In a videogame, a successful search typically causes the located item to become highlighted with a coloured outline, identified by name on the command bar (“Look at mysterious statue”), or produce a positive response when clicked. However, such contextual metainformation and feedback is not available in a real-world ER. Instead, designers should be careful to ensure that any object that players find in an escape room that is important, looks important. Focussed lighting effects can draw attention to particular objects or elements, as (to a less precise degree) can directional sound design. If an object becomes available midgame (e.g. dropping from a trapdoor), then it can be accompanied by a sound effect and triggered by an action such that players are likely to be nearby and notice its movement. 

Conversely, designers should also make sure that any element that looks important, is important. Items that appear to be significant but actually have no purpose are red herrings, either created intentionally or not, and players may spend effort trying to interpret meaning or function that was never present - a frustration that is one of the most common complaints among ER players. Any elements that are not relevant to the game should be removed from the room or, if this is not possible (e.g. it may be necessary to have a “Break Glass” fire alarm switch installed on the wall for safety reasons), then these should be clearly marked with some indication to let players know that they are not part of the game. 

While obscurity is a valid game mechanic (on which the entire subgenre of “Hidden Object” games are based), and many players enjoy the activity of physical searching (as evidenced by this survey), it needs to be done in moderation and balanced with other types of activity. It is often the case that directed searching - i.e. in which the players know what, and how many items they are looking for, is more enjoyable than undirected searching. Crucially, a directed search can be ended at the point that players know they have succeeded in finding all required items, rather than an undirected search of potentially unlimited duration for an unknown number of items. 

Determining the Action Space

Having identified the puzzle elements, the next step is for players to decide the ways in which those elements can be manipulated or interacted with. In a videogame, all interactions within the game are explicitly defined within the program code, and (generally speaking) are deliberately designed to occur. If a tower topples over when hit by a projectile, it’s because the physics engine specifically allows it. If a red keycard opens a red door, it’s because it was coded that way. In early text adventure games, this set of possible actions could be enumerated from the list of commands understood by the parser, which was normally limited to common verbs such as "push"/"pull"/"use"/"look" etc. As such, there were only a finite number of ways in which puzzles could be solved. 

In escape rooms, however, the opposite is true. Players may attempt to perform any actions of which they are physically capable. A computer terminal placed in an escape room can not only be turned on and off, but dismantled, picked up and thrown, juggled, or have marmalade smeared in its disk drive. Rather than create lines of software code to explicitly allow certain player actions, escape room designers need to impose rules to limit the action set of the players and remove these undesirable choices. Pre-game briefings frequently explain "you don’t need to lift anything that weighs more than a book, you don’t need to unscrew anything, and you don’t need to touch anything with a warning sticker". If you don't want players to pick up and move a normally portable item, you need to screw it down.

Even having eliminated some of these options, a great many possible actions remain. Players therefore work out the likely methods of interacting with an item based on its perceived affordances. "Affordance" is a term first used by psychologist James Gibson in 1966 to refer to the set of possible actions between an individual and their environment. In game design terms, the "perceived affordance" of any object is the set of actions that a player readily infers about the possible ways in which it may be used, implied by virtue of its physical design. 

Creating Affordances

Consider a desk drawer. It has the affordance of being opened, since it can slide along a rail mechanism. That affordance of “openability” is conveyed by placing a handle on the front panel that can easily be pulled - the handle is the cue that allows players to perceive that the interaction is possible. Many ER props have such affordances: a large, hexagonal-shaped key has the affordance of being inserted into a similarly large, hexagonal-shaped keyhole, for example. In cases where affordances are not obvious, they may be explicitly signified through use of symbology, shape, colour, or other design elements placed on the object; players will naturally infer a relationship between objects sharing such common elements or attributes. 

False Affordances

Now suppose that you were to place a handle on a simple wood panel. That would generate the perception that the panel was the front of a drawer that could be pulled open, but that perception would be misleading. This is a false affordance - when something doesn’t behave the way that you would expect it to. False affordances are the interaction equivalent of red herrings, and they similarly can be a cause of great frustration in escape rooms. If players repeatedly attempt to perform an impossible action, designers should consider whether they have created a false affordance that has misled players into thinking that this action was possible. 

Hidden Affordances

Instead of placing a handle on a wooden panel, now suppose that we have a working drawer, but we remove the handle. The drawer can still be opened (perhaps with some effort and a little prising), but there is no longer any visual cue that this action is possible. This creates a hidden affordance. Hidden affordances can be used intentionally as a puzzle mechanic - many escape rooms use technology like magnets of RFID sensors to create "magic" interactions between objects, which can result in delight if executed well. However, if players cannot reasonably discover the correct method of interaction with an object, they may exhaustively attempt to find other affordances through brute force, and this may lead to damage. Rather than be completely hidden, it may be desirable to simply obscure certain affordances so that they are non-obvious, but still discoverable.

Outside Knowledge and Learned Behaviour

Use of "outside knowledge" is generally frowned upon in escape rooms; nevertheless, it is inevitable that players will perceive certain affordances and recognise associations between items based on prior conceptual models or previous experiences. If players find a flashlight that doesn't work, and separately find some batteries, it is reasonable for players to expect that inserting the batteries into the flashlight will make it work, since batteries have an expected affordance to make electronic items work.

A clever designer can exploit this learned behaviour by deliberately setting players up to think they know how to solve a problem a certain way based on previous experiences, when in fact a much better way exists. This behaviour is known as the Einstellung effect, and is often used in magic tricks. It should be applied with extreme caution though - if implemented badly, it can make players feel as if they have been tricked (which they have...).

Action Execution and Feedback

ER players do not compete against each other, but rather they are playing against the room. As in most games, there therefore needs to be two-way interaction between these parties. The players take their “turn” by attempting to perform some action, and the game then responds with appropriate feedback.

The reward for solving a puzzle is often the code required to unlock a padlock, information or items required to assist in solving another puzzle, or an electronic trigger that changes the status of some item in the room. Whichever of these occurs, the outcome should be explicit to the player - there should never be uncertainty as to whether a puzzle has been solved or not.

Incremental feedback can also be used to reward correct steps taken towards the solution - this can include lighting or sound effects to indicate to the player they are taking the right approach. Likewise, incorrect actions can also be pointed out.  This forms a dialogue of sorts between the players and the game - the player tries something, and the game replies “Yes, good.” or “No, that’s not quite right”. 

To return back to the very first point of this article thought, players like solving problems - feedback mechanisms should not be so explicit as to allow players to solve puzzles by trial and error - there should be a logical way of deducing the correct solution, and the feedback mechanism serves only to guide players onto that path, not to reveal the answer. The ideal feedback should be elastic - providing stronger guidance the further away players’ actions are from the desired path, and gentle nudges for those only slightly astray from the flow channel, where their skills perfectly match the level of challenge.

Finally, designers should always maintain a responsible attitude towards the player actions required to solve a puzzle, and how they are rewarded. It is incredibly irresponsible of ER designers to create puzzles that require players to stick a fork in a fake electrical socket (yes, they exist) because, by rewarding that action in one escape room, it will likely be something that players will consider trying again; only next time the electrical socket might not be fake...