It’s not really an age old debate, but it’s something I have struggled with a bit ever since I started DMing for 5e games. Much like perception vs search in the older D&D systems, sometimes it’s difficult to determine what you actually want the players to roll in order to find something. Now, it’s pretty clear cut that in order to find a creature that is hiding you use perception, but what about finding secret doors and hidden compartments? The default book answer is perception, but then you’re left asking yourself “what is investigation actually useful for?” Sure, there are those outlaying instances where it’s obviously what you want to use, but I still have a hard time justifying perception over investigation when searching for traps, hidden doors, and other items for the following reasons:
- It hurts anyone proficient in disarming traps. Arcane trickster rogues even uses intelligence primarily for its spellcasting, so you know that’s going to be a high stat. It seems a bit unfair for that to go to waste if they can’t use their naturally high investigation to assist them in any way when looking for traps which is still a big part of a rogue.
- It may be hidden to the senses. Sometimes based on the description of the trap it’s impossible to spot it. Literally impossible. In that case you may have to use clues of its activity in order to determine that a trap is here. Hmmmm … that sounds kind of like investigation, doesn’t it?
- I hate having perception be the king of everything. Nothing gets more boring than having everyone roll perception over and over again for everything. I wouldn’t cry too much if active perception went away completely, but for now I’d rather mix it up a bit.
The official material has gone back and forth on what to use; sometimes requiring perception and sometimes requiring investigation. Sage Advice has said there was overlap, which doesn’t really help at all. There are several approaches I have tried over multiple sessions to varying degrees of success. Consider the following scenario with a ranger with good perception and a rogue with good investigation:
You enter the dimly lit stone room. This is obviously a lived-in apartment as you see common furnishings, a wide open closet, and the occasional piece of clothing thrown about. An almost completely melted candle sits besides a pile of notes and pen on a table.
Now, if you’re my players you’ll look around at the immediate stuff, probably read the notes, and then determine that there has to be some sort of secret around here and begin looking around. At this point I’ll be looked at me expectantly to tell them to roll something and find the legendary items obviously hidden in this commoner’s closet.
“Roll investigation or perception – your preference.” I’ve tried giving them the choice of the skill they want and setting a DC that either of them can meet. The explanation for this is that they may both find it, but they’ll go about it in different ways. The perception guy may spot some creases in the drawer that shouldn’t be there while the investigation guy may observe that the inner dimensions of the drawer are too small based on the outside dimensions.
The result of this is mixed but probably the best out of all of the approaches. It seems to make players happy since they can choose their good skills, but it also slows things down since they’re not used to having a choice like this and will fumble with their sheets or just be indecisive in general. One thing I don’t like, however, is that on occasion one of those skills just plain won’t work for the situation. In that case, when I call for a single skill check, I’ve already given some information away. If I call for both, one of the players get frustrated because although his 20 didn’t see it, the 15 with the other skill did. I started giving them the “uneasy feeling” for those with the skill that couldn’t spot it, which seemed to help, but that just slowed the game down more than I would have liked.
“Roll investigation.” In some sessions, I’ve just made investigation the default for detecting objects like traps, secret doors, etc. This speeds things up and made the rogue happy, but the ranger is constantly grumbling because he feels useless being unable to spot things with his eagle eyes.
The result of this I kind of liked since investigation got a good use and I was able to describe things in more detail than “you spot a trap … I don’t know how you know it’s a trap, but it’s a trap.” Instead he could see the slight acid spills around the keyhole and determine there was some sort of acid trap there, or see that the doorknob was positioned in a rather odd way that might hint at something being tucked behind it.
“Roll perception.” Now the rogue is feeling pouty because other people have to point out traps and secret doors that he supposedly specializes in, but can’t seem to spot. The ranger feels more useful because he can help with his sharp spotting skills.
Out of all of them I like this one the least. It seemed rather anti-fun to have the trap master depend on others to be better at seeing things he should be able to pick out in the first place due to actually knowing what he was looking for.
I’m not really sure what the best approach is yet. It is probably none of these. I think I’ll be going with approach number one for now and try to get a bit more creative in cases where there is no overlap; perhaps have one or the other grant advantage to certain things based on what they’ve done.