they way i understand it now, is that with passive perception the character isnt really looking for "something" but might spot it anyway..as opposed to active perception where you are actively looking for "it".
It's more variable than that... "Passive" rolls are more like a metagaming construct to avoid certain players asking a DM to let them roll for trapfinding every couple of meters and thus slowing the game down. They are sometimes quoted as having also the purpose of letting a DM secretly determine if a PC notices something, although this is not really more useful than the DM
rolling perception secretly in the PC's place.
Unfortunately, trying too much to link this metagaming-motivated rule to a narrative always caused unwanted consequences since Take10 in 3e. So go there if you want, but be aware that you may open a can of worms. There is already enough ambiguity between Perception and Investigation, that we don't need more discussions between
eight different options such as Active Wisdom Perception, Active Intelligence Perception, Active Wisdom Investigation, Active Intelligence Investigation, Passive Wisdom Perception, Passive Intelligence Perception, Passive Wisdom Investigation, Passive Intelligence Investigation. You don't need to use them all, and you don't even need to care too much about which one to use. Remember that at the end it's only a matter of
probability of success or failure, and players won't even notice the difference, unless they have very a good perception of probabilities... or is it investigation? :/
Just in case you wish to know, I usually do it as a mix bag, without the pretense of being consistent:
- roll Wisdom Perception for
noticing (when not actively searching) noises, visual details, smells, environment anomalies, as well as hidden things, but also when
actively listening to a door or visually focusing on something; the key idea is to "see if there is anything to see here" without knowing what you're looking for
- roll Intelligence Investigation when you know what you're looking for, which pretty much implies
actively
I don't normally use passive checks at all, because even with
repeated/routine trapfinding I prefer to let the PC roll, and then just apply the result to a large window of time instead of a single instant and place (so I might let you roll
one Investigation check for a whole hour in the caves, or for a whole locale).
A final suggestion: if you don't want to trap (!) yourself into the situation where passive checks means autosuccess/autofailure, use them only when there is an
contest i.e. opposed checks, and obviously let the other side roll.
Say you are on a road and hear a group of Orcs marching on the same road towards you. You jump behind a tree and try to hide. Say the Orcs arent looking for you. The DM lets you make a stealth check and compares this against the passive perception (which is DC10+wisdom ability) of the Orcs. Is there a difference between 1 Orc and 5? Does the player for instance has to make 5 checks in that last case? Of is it still one roll but with a modifier of some sort?
And how would the above scenario play out if the Orcs were active looking for elves.
When you have a group AND the consequences of the check are supposed to affect the whole group, you can always do a (properly named)
group check, which means the whole group succeeds if half of its members succeed.
OTOH, if everyone in the group has the same modifier to the roll, it might be just as good to simply roll
once for the whole group, especially if the group is large.
The bottom line is that the actual statistical distribution can be very different, but since the outcome is merely success/failure, all you may want to care about is the
average. But still, why you should even care about a few % difference in an average, when it would only really matter over a large number of occurrences of exactly the same situation?