I agree with trying to get the players more involved beyond rolling dice, but not all of your conclusions about the skill.
Here's an article from Paper and Pencil dot com from 2012 that discusses (at length) the Perception skill in PF.
OVERVIEW OF PATHFINDER’S SKILLS: PERCEPTION
I doubt there is any skill quite so divisive as perception. I hardly feel qualified to assess it. So many dozens or hundreds of men and women much more experienced than I have spent so many forum threads and blog posts debating back and forth over the issue of perception that I feel presumptuous even attempting to resolve it. But that’s what I signed up for when I said I was going to begin this series of posts, and I’m not going to back out now, so here we go.
If you recall from earlier in this series, I believe disguise and escape artist are two skills which should be house ruled out of the game because they are needed so incredibly infrequently as to be useless. Well, if those skills commit the sin of a deficient frequency, then perception commits the sin of excessive frequency. The skill is so frequently called for, and so necessary, that -C uses the term “Skill Tax” to describe it. Putting points into perception is more of a requirement, than an option. And if something is a requirement for effective play, why clutter up the skills list by giving players the illusion of a choice?
But I’m getting ahead of myself. If you’d like to have perception torn to pieces in front of your very eyes, read -C’s two posts linked above. He’s done a superb job highlighting the problems with this skill, and I feel no need to repeat that task. As daunting as it is to attempt to resolve the problems with perception, I have no qualms about stating quite firmly: perception does have problems.
Pathfinder uses perception in two different ways: one which I’ll call active, another which I’ll call passive. Active perception checks are generally requested by the player. “I’d like to make a perception check to search this room,” or “I’d like to listen intently for someone sneaking up on me.” Active perception checks occur when the character is primarily engaged in the act of looking, listening, smelling, feeling, or tasting. Passive perception checks (sometimes called “reactive”) most often occur without the player’s knowledge. They are rolled by the GM, in secret, to determine whether or not a character is able to perceive something. This might happen when there is a faint scent of cooking coming from behind a door, or if someone is attempting to move without the player’s noticing them. Passive perception checks are useful for information which the players might not be looking for specifically, but which the characters might nevertheless notice and find useful if they’re alert.
Pathfinder is actually pretty vague about when a perception check should be rolled, and what information a player should be able to gain from it. As such, we can’t exactly blame the game for the fact that many GMs and players use it as a substitute for creating & interacting with an actual environment. However, the game would function better if it presented a more focused version of this skill. In the hopes of correcting this oversight, I've broken perception down into its component parts. Below are the various uses I've seen for perception, and my attempts to work out what the skill should apply to, and what it shouldn't.
Active Perception: Primarily used to find hidden treasure, discover secret doors, and avoid traps.
If a room contains treasure, that treasure is either obvious, or not. Obvious treasure, such as that found in a chest, on the body of a foe, or simply laying on the floor, should never require a perception roll to find. So long as the players say they’re searching the body, or opening the chest, they should be allowed to find the treasure.I would hope that was self evident. If a treasure is not obvious, then the players ought to be forced to look for it if they want to find it. The GM should describe the environment the players are in, and if they feel there may be treasure present, they can describe to the GM where they look for it.
What could possibly be gained by having the players roll to discover non-obvious treasure? If their roll succeeds and they find the treasure, then it’s not functionally different from obvious treasure. It simply required the extra step of rolling a die. If, on the other hand, their roll fails and they don’t find the treasure, then the treasure effectively does not exist, because they can’t find it. It’s not as though it’s fun for a player to simply be aware that treasure was hidden before they found it. What is fun is the process of finding that hidden treasure. Coming up with the idea to search for loose bricks on the wall, and being rewarded by finding a bag of 30 gold is a lot more fun than entering a room, rolling a die, and being told you found 30 gold behind a loose stone.
Locating secret doors is somewhat different from finding non-obvious treasure in two important ways. First, the hiding place of even the most well hidden treasure ought to be mentioned in the room’s description. If it’s not, then the players have no way to know where to search That doesn't mean you need to drop obvious hints, simply that it’s not fair to hide treasure under a bed without telling the players that the room contains a bed. On the other hand, secret doors are most commonly built directly into the walls or floors. Since those are present everywhere, there’s no good way for players to search for secret passages intelligently. The second difference is that treasure is an end unto itself. It, along with experience points, is how players are rewarded for successful play. Secret doors, on the other hand, exist only as a means to an end. That end being whatever lies beyond the door. Though the joy of discovery shouldn't be discounted.
Given those differences, I think it is reasonable to allow players to roll perception checks to discover hidden doors. The area for such a perception roll should be relatively small–perhaps 15ft square, centered on the character. Some characters could even be allowed to find hidden doors with a passive perception check, if they passed within 10 feet of it. Traditionally this is an ability which was given to elves, but perhaps it would be better if anyone with 5 ranks or more in perception was given the chance to automatically discover hidden doors. Reducing the act of finding a hidden door to a roll does come with a danger, however. If the entirety of the interaction is rolling to look for a door, finding a door, opening the door, and going through the door; how is that any different from rolling to find non-obvious treasure?
I would propose that a perception roll allows players to, as -C puts it “learn the location of the secret door but not how to open it.” Once they know the door is there, the players are free to attempt to bash it down if they wish. Though in some cases that may be pretty difficult. Alternatively, the players can search for the mechanism within the room which opens the door. Perhaps a loose stone needs to be pushed in, or three worn-down keys on a piano need to be pressed simultaneously. And if the players notice the loose stone or the worn-down keys before they find the door in the first place, then the discovery of a secret passage will be made all the more exciting.
Using perception checks as a means of finding traps is something I've struggled with as of late. My party’s rogue has frequently complained of the tediousness involved in searching every door, every chest, and every trigger that they find in order to avoid the handful of traps in each dungeon. And I have to agree: it is tedious, and not at all fun. I suppose I could make it less tedious by simply adding more traps to my dungeons, but that seems like a lazy solution. The problem is not that there’s an insufficient amount of things to find, it’s that excessive rolling is boring.
Pathfinder has a rogue talent called Trap Spotter, which allows rogues to make an automatic, passive perception check whenever they are within 10ft of a trap. I propose that this be removed as a possible rogue talent, and instead, this effect be made part of the Trapfinding class ability which rogues receive at level 1. Additionally, any character with 5 ranks or more in perception could also be granted the ability to notice traps using passive checks.
Judgement: In general, active perception checks seem to sap potential fun from the game. I propose eliminating active perception checks from the game, except when it comes to searching for secret doors or traps within a 15ft square area centered of the player.
Passive Perception: Primarily used in opposition to a sleight of hand and stealth checks or to notice fine details in an environment.
Sleight of hand may be the single most well-designed skill in Pathfinder, and rolling perception against it is a reasonable method of conflict resolution. There is no need to alter the way in which perception interacts with that skill.The stealth skill, on the other hand, has a number of problems, which I will detail once I write my overview for that skill. Speaking strictly on how perception interacts with stealth, however, I can find no fault. If the stealth roll is being utilized properly, then rolling perception against it makes perfect sense.
That only leaves a roll for noticing fine details in the environment. This is anything which the character’s might not notice right away. A faint smell from two rooms over, an Orc's knee sticking out from behind the barrel he’s hiding behind, or a crack in a stained glass window across the room. I think this application of the skill works well enough. Whenever I go to use it, though, I ask myself one important question: is there any reason to withhold this information?
Using the examples above, it makes sense to roll a passive perception check to see if the players notice the Orc's leg sticking out from behind a barrel. The creature is obviously attempting an ambush, and their success or failure in that attempt will hinge on whether or not the players notice him first. That’s a good use of a roll. It also makes sense to make the roll to determine if the players notice the faint smell of cooking from two rooms over. If they notice it, they have an opportunity to prepare to enter the evil cult’s mess hall, if they don’t then they’ll be surprised when they walk into a room filled with enemies. The crack in the window, however, is essential information if the players are going to figure out the room’s puzzle. If they fail their spot check, then the puzzle becomes unsolvable unless one of them decides (out of the blue) to examine the window for cracks.
Note that you’re not rolling perception based on the type of information, you’re rolling it based on how that information will be used within the game. That may seem silly and unrealistic, and it is. But realism does not equal good game play.
Judgement: Passive perception is mostly fine. Just don’t use it to hide information the players ought to have.
Overall Judgement: I think perception has a valuable place in the Pathfinder game. However, the game benefits significantly from reducing this skill’s importance drastically.
