Thursday, April 23, 2009

Perception, Senses, and Searching in Roleplaying Games

I've been thinking over perception and searching in gaming. The way I figure it, there are a few ways to go about it:

1. Negotiated: if there is a dagger +1 in the pile of moldering clothes, a character who rifles through them is going to find it. The DM might rule that a character who merely prods the pile of rags with his trusty 10' pole (to make sure that there's no monster hiding in there) might not find it. If there's a key in the third desk drawer, a character who says he's opening said drawer (or says he's searching the whole desk) finds it. End of story. I like this method because it rewards clever players, but it can also become tedious if the party stops to pull and prod every brick, book, and candle in whatever location they happen to be exploring. I've noticed that older games tend to at least partially treat perception like this, as do games that don't address the issue mechanically at all.

2. A flat probability. I'm not sure I've ever seen this outside certain rules in older versions of D&D: characters have a flat chance of finding a secret door, or a party has a certain chance of being surprised. (Which I think it a matter of perception) This tends to put characters on more or less equal footing,

3. Perception is a stat. Games I've seen that operate like this include all the old White Wolf games, "classic" Deadlands, In Nomine, Gamma World 4th edition, etc. I'm sure I could think of more if I tried. I have some misgivings with this method, because players get lazy and just ask if they can make checks with their perception stat. The DM can also get lazy, just setting up rooms and encounters to be contingent on someone passing a stat check.

4. Perception is a skill. This started showing up in the form of Thief skills (Detect Noise, Find/Remove Traps), and mutated horribly in the d20 days to be three separate skills. (My least favorite handling of perception in all of gamedom.) I have the same problems as above with #3, except this also forces the players to spend skill points/levels/whatever and makes them less capable at doing other things.

5. Perception is a roll you make with another stat. The new World of Darkness and it's various offshoots do this. I find that games often co-opt the system's equivalent of "intelligence" for this. I'm wary of game systems that overemphasize certain attributes by making them "too good."

Personally, I like negotiated perception the best, though I do see a need for a mechanical system to determine things such as surprise, and I can see the value of the old rules for finding secret doors. (Lest your session turn into "Okay, I tug every book on every bookshelf. I twist and pull all the torch sconces, I push ever brick in the floor.") I am also not a fan of adventures that hinge entirely on a successful search/perception/cognition roll to find a clue, key, or MacGuffin without which the party cannot progress further.

The more I think about it, the more I find that I dislike a mechanically standardized approach to perception. I think that it simplifies things, but ultimately takes away from the experience. However, there should be rules in place for things like ambush, or a character using "hide in shadows" while other creatures pass by.

I'll have to think on this further.


  1. Don't think it either or. You can combine the negotiated approach with the mechanical approach. Limited the time you need to roll.

    For example if there is a chest covered by garbage then the player "says I will undercover the garbage" they will find the chest.

    If however it is a jewel in that pile. Then I would have a roll because it may be missed despite the player digging through it.

    If the player insist on standing at the door and looking around the room. Then you would roll for the chest in the garbage (the garbage is piled up unusally) but there would little or no chance to spot the jewel.

    This the approach GURPS takes to avoid incessant rolling of skills. And I pretty much apply this technique to any skill based system

    Now for older editions of D&D what I would do it make sure that everybody can to a base series of actions. (Perception, climb, stealth, etc). The thief class would sacrifice combat ability in order to be good at something else.

    The other classes are not prevented doing these things. But because they are focused on fighting, praying, or spells they never get as good as the thief.

    The problem with the original Thief that it implies that only the thief can do certain things. Just only the fighter get the high HD and good to hit bonus. Only the magic-user get to cast wizard spells.

    If they instead laid out how everybody could climb, perceive, jump, etc, and gave thief a bonus then I think everyone would have a better feeling about the thief.

  2. I agree with Rob that it is not an "either or" situation. I like to think of the percentage chance approach as an abstraction of the "open ended puzzle" that is your "Negotiated". I had to try and describe it briefly when I put together a Quick Start Adventure for OSRIC last year, which might bear reproducing here:

    Instead of following predetermined mathematical formulae, all tasks are resolved by deferring to the concepts of reasonable probability and environmental interaction. In the case of the former, the game master simply assigns a reasonable probability of failure and the consequences thereof, often taking into account the attributes, race, class, level, background and circumstances of the character. For many actions, perhaps even the majority, there need be no risk of failure and so no die is rolled; similarly, some tasks will have no chance of success and so require no randomisation. The only reason to assign a probability to a task is to randomise the outcome, a measure that is rarely necessary, except to heighten tension or resolve uncertainty.

    Whilst assigned probability provides an abstract means of resolving tasks, environmental interaction takes a more literal approach; it encourages players to think carefully about their imaginary surroundings and make intelligent decisions based on the flow of information between them and the game master. For instance, players who think to have their character inspect a chest for a false bottom, look behind a specific tapestry, or seek to trigger a suspected trap should be rewarded with the logical outcome of such actions, given that there is something to find or a mechanism to trigger. The environment thus becomes something of an open ended puzzle that is as complex and challenging as the game master cares to make it.

  3. Matthew- Thanks for posting that. This is exactly what I have in mind, a sort of combination of methods 1 and 2.

    I might also mention that, next time I run D&D, I am interpreting thief skills as "ordinary abilities on steroids." Anyone can sneak, but a thief or ranger can move with total silence. I interpret the detect noise skill as something passive; anyone can try to listen down a corridor, but the thief might suddenly shush his companions as they are in a discussion because his observational abilities are trained. I mention this since many of the thief's skills are so closely tied to perception, be it the thief's or the perception of someone he is trying to evade.

  4. No problem, glad to be of help. I also consider thief abilities to work the way you describe. Most have "magical" or "semi magical" analogues in AD&D, so it is no great stretch (indeed, it is arguably "by the book").