D&D 5E How do you handle insight?

[Swarmkeeper]

As far as "punishing" players for asking for a check ... I don't. In the case of insight and a few other skills I use the better of passive or the roll.

Perhaps I was unclear. At our table, players declare an action and goal. If they jump ahead and start rolling, I gently ask them to describe what their character is doing and what they are trying to accomplish. No one is being "punished". Players learn to drop the habits of other games and adapt to the 5e mechanics. When I do ask for a roll as DM for an ability check, I also relay the stakes involved (i.e. the DC and the "meaningful consequence of failure"). The player can usually then decide if their PC would proceed with the potentially risky action or rethink things.

Having a PC in the party who is naive and gullible is quite common in my experience. If a PC's flaw is to be gullible and they're they only one questioning an NPC, the NPC can lie their ass off.

Agreed, if that is how a DM decides to play that particular NPC based on their bonds/flaws/etc. Then again, perhaps that NPC is overconfident and goes way over the top in his lie, even twirling his mustache as he's lying through his teeth, rejoicing in his ability to pull one over on these naive adventurers who think they are so special - and not realizing suspicious Brog is about to break him in half. I'm sure we can think of other ways to play it out, too.

As a DM I need some indication and reminder that they aren't going to be suspicious.

Absolutely. Having players role play their action and goal is a great way to get this out.

Player (Susan): Suiza trustingly asks the proprietor "We found a body in the alley out back, do you know anything about it?"

DM, now reminded that Suiza's character is the gullible type, might even grant inspiration to the PC. Suiza still might auto-fail (i.e. no roll called) since she's so gullible, but the player is reinforcing what her character is really like to everyone at the table.

So maybe a different question: a PC has the flaw "I put too much trust in those who wield power within my temple's hierarchy. " They're questioning someone superior in the hierarchy that is lying but is proficient at deception.

To me this is a scenario where I would like to call for an insight check, but I don't know my player's characters to that level of detail.

I agree - no DM should be expected to know any player's character completely. Over time, through role play, the DM will gain familiarity with the personality of each PC, but the DM has more than enough on his/her plate to keep the details perfectly straight. It is therefore up to the player from time to time to remind the DM (and everyone at the table) through first or third person role play what their PC is like - traits, ideals, binds, flaws, etc.

Player: With a deep bow of respect to my temple superior, my PC asks her...

For me it's simpler and more fair to the player's vision of their PC to let them ask for an insight check now and then.

To me, the table learns nothing about the character - or really what they specifically want to do and why - when the player just asks to make an Insight check. To be honest, I just don't find that to be an interesting style of play. Works just fine for some, but not for others.

 

[MarkB]

I’m not denying players a chance to apply proficiency to anything. I ask that my players relate their actions to me with a reasonable degree of specificity (reasonable being a key word here) so that I can adjudicate those actions without making assumptions.

So, when a player says "I'd like to make an Insight check" in response to a statement from an NPC they're questioning, what is it that you find unreasonably non-specific in this instance?

 

[5ekyu]

Perhaps I was unclear. At our table, players declare an action and goal. If they jump ahead and start rolling, I gently ask them to describe what their character is doing and what they are trying to accomplish. No one is being "punished". Players learn to drop the habits of other games and adapt to the 5e mechanics. When I do ask for a roll as DM for an ability check, I also relay the stakes involved (i.e. the DC and the "meaningful consequence of failure"). The player can usually then decide if their PC would proceed with the potentially risky action or rethink things.



Agreed, if that is how a DM decides to play that particular NPC based on their bonds/flaws/etc. Then again, perhaps that NPC is overconfident and goes way over the top in his lie, even twirling his mustache as he's lying through his teeth, rejoicing in his ability to pull one over on these naive adventurers who think they are so special - and not realizing suspicious Brog is about to break him in half. I'm sure we can think of other ways to play it out, too.



Absolutely. Having players role play their action and goal is a great way to get this out.

Player (Susan): Suiza trustingly asks the proprietor "We found a body in the alley out back, do you know anything about it?"

DM, now reminded that Suiza's character is the gullible type, might even grant inspiration to the PC. Suiza still might auto-fail (i.e. no roll called) since she's so gullible, but the player is reinforcing what her character is really like to everyone at the table.



I agree - no DM should be expected to know any player's character completely. Over time, through role play, the DM will gain familiarity with the personality of each PC, but the DM has more than enough on his/her plate to keep the details perfectly straight. It is therefore up to the player from time to time to remind the DM (and everyone at the table) through first or third person role play what their PC is like - traits, ideals, binds, flaws, etc.

Player: With a deep bow of respect to my temple superior, my PC asks her...



To me, the table learns nothing about the character - or really what they specifically want to do and why - when the player just asks to make an Insight check. To be honest, I just don't find that to be an interesting style of play. Works just fine for some, but not for others.

Perhaps unclear...

Well let's be specific, you lead off a post with this.

"I'll echo some advice I've heard before here on ENWorld: Why would a player ask to make a specific ability check in 5e? Given that the 5e DMG points out that the DM should "Only call for a roll if there is a meaningful consequence for failure", a player requesting to make a specific ability check is asking for a chance to fail and thus harm the party in some way."

That seems clear. It seems to clearly state that if a player asks for a skill check is asking for a chance to fail. Later on you go into more detail

If instead the results of asking for a skill check was just a reminder that it's done differently, how is this good advice? How do they bring risk of harm to the party thru a failure?

I ask cuz it seems a repeating loop - dont fo this cuz it might give you failure you would not have gotten but no I wont do that to my players.

 

[Ovinomancer]

So, when a player says "I'd like to make an Insight check" in response to a statement from an NPC they're questioning, what is it that you find unreasonably non-specific in this instance?

What they want to do? This request seems like asking for a roll so that it's acceptable for the PC to think something. Not my responsibility as a DM to determine or adjust what a PC thinks.

This is my fundamental issue with Insight as it's usually discussed. It's usually discussed in a manner of "is there a clue so that my PC is justified in having this opinion of the fiction?" The response is to use Insight to find this clue. The problem this creates was identified in the last thread on this topic -- what happens if the NPC is telling the truth and the PC fails the check? This gets into the GM presenting fiction specifically to mislead the players, only it's either known (the PC sees the roll) or always hidden (the GM always makes secret rolls, so the player isn't really getting any reliable information to begin with). Instead, I use Insight (and all skills) as active resolution mechanics. Insight, for me, is used when trying to find out information from an NPC via social interaction. This requires interaction with the NPC and isn't a check made against what the NPC has said to determine truthiness.

For example, using the OP example, the PCs may try to find out more about the owner, eliciting Insight checks to extract information (against the NPC's attitude DC). When they feel like confronting the owner, their approach may be trying to find out information from the NPC leveraging clues they've found, which would be an Insight check.

Mabye Bob's character starts by trying to intimidate the owner into telling the truth by looming and yelling that the owner's lying. Susan's character takes the opportunity of the owner's being distracted to look around the store for anything about the owner she can leverage if the intimidate fails. Which it does, and the shopkeep backs away from Bob's character sticking to his story and threatening to call the watch. Susan's character's perception check succeeds and she sees a child's drawings pinned under the counter and a few toys. Susan's character steps in between the shopkeeper and Bob's character and tries to defuse the situation by shooing Bob's character away and apologizing to the owner for the bad behavior (persuasion check success), then starting up a talk with the owner about their child, trying to improve attitude (Insight check with advantage for having good info, success). Now that the owner is placated and talking Susan tries to leverage concern for the child into finding out more information about the murder last night by promising protection for the owner's family and silence on where any information comes from. This is another Insight check (trying to elicit information), success meaning the owner talks and failure causing him to shut up and demand they characters leave the store.

But, at no point would I let a player ask for a check to determine if their character thinks something or not. The players are either suspicious, or they aren't. I also follow an 'active PC' mode of play where NPCs don't usually roll against PC outside of combat. NPCs do things, and players act. If an NPC is being intentionally deceptive, I'll use their passive Deception as the DC for any elicitation checks against them rather than the usual attitude checks. A successful Insight check uncovers the deception, but the player has to be engaging the NPC with a goal and approach (as above) that would result in asking for a check. Failure means the deception would be repeated. The owner just not being willing to offer information isn't an active deception, in my book, just recalcitrance. The owner being the murderer, on the other hand, would likely be an active deception.

 

[Charlaquin]

So, when a player says "I'd like to make an Insight check" in response to a statement from an NPC they're questioning, what is it that you find unreasonably non-specific in this instance?

“I’d like to make an Insight check” doesn’t give me either of the two things I need to be able to adjudicate the results: a goal and an approach. It just informs me that the player wants to take action, and thinks their proficiency in Insight will help them out.

Based on what sorts of tasks the Insight skill is typically applicable to, I think it would be reasonable to assume that the player’s goal is probably to learn something about the NPC’s emotional state. I don’t like making assumptions like that if it can be avoided, but in this case I think it’s a pretty reasonable one. The bigger issue in this example is that I don’t know what the character is doing that they hope will result in the revelation of whatever information they are hoping to glean.

So, my response to this player would be something along the lines of, “I’m hearing that you want to learn something about the NPC’s emotional state, but I’m not sure what, or what your character is doing to try to learn it. Could you be a bit more specific about what you hope to accomplish and how?”

 

[Charlaquin]

Perhaps it would help to lay out the typical flow of the conversation in my games.

1. I describe the scenario, usually in terms of what the PCs can directly perceive, and ask a player, “what do you do?”

2. The player announces what they do in terms of what they want to accomplish (their goal) and what their character does to try to accomplish it (their approach).

3. I assess whether the approach has a reasonable chance of succeeding at achieving the goal, a reasonable chance of failing to achieve the goal, and a cost for the attempt or consequence for failure.

3a. If the action has all three things, I decide a DC, and which Ability is most appropriate to resolve the action. Then I tell the player to “make a DX [X] [Ability] check. On a failure, [consequence.]

3b. If the player feels that one of their Proficiencies (skill, tool, weapon, armor, or language) might help them achieve their goal, they can ask me if they can add their proficiency bonus.

3c. The player can decide to make the check, or to go back to step 2 if they feel the proposition is too risky.

4. I describe the results of the action, then return to step 1.

A player asking to make an Insight check, whether intentionally or not, is asking to skip steps 3 through 3c, under the assumption that their character takes an approach that has a chance of success, a chance of failure, a cost or consequence, Wisdom was the appropriate ability to use to resolve it, Insight was an applicable Proficiency for it, and the risk and potential consequences were acceptable.

And for some DMs - in my experience, a lot of DMs - that’s fine. Lots of groups like to leave the approach abstract, make the roll, and determine the approach retroactively based on the results of the roll. There is absolutely nothing wrong with running the game that way, if that’s the way you enjoy playing. However, for me personally, that often leads to an experience that I find dissatisfying. To each their own, but at my table, if you ask to make an Insight check, I’m going to ask that you instead tell me directly what you hope to accomplish and how, and I will decide if a check is necessary to determine the results.

 

[Ashrym]

Context is kinda key to that quote, especially the first sentence:

Finally, the Core Mechanic very explicitly spells out that the rules for action resolution are TOOLS used by the GM to determine the outcome of ACTIONS chosen by the players. It may not seem like a big deal – because we all know that’s how it’s supposed to be anyway – but that wording is very useful to new GMs and new players. And some experienced players and GMs need to hear that too. Under these rules, a player who asks the GM “can I make an Insight check” is not playing the game properly. They are playing against the rules.​

Again, everyone should feel free to use the rules however you like as DM. But when that causes some issue or awkwardness at the table (like the cascade of player-invoked rolls to accomplish the same task mentioned previously, for example), just refer back to p6 of the PHB for the intended general flow of 5e play and to p237 of the DMG for the 5e designers' intent for how ability scores are to be used in play.

The issue I have with this (and several other comments here) is that it seems to assume an action is necessary in order to make an insight check. I find the action of the players is the exception and the checks are made as an interpretation of NPC actions. That's like trying to force players to describe actions before checking perception. Yes, there are active perception checks and mostly passive. That's how I generally treat insight.

I don't have a problem with "I want to use insight to..." because the character would know how to apply the ability better (possibly worse; the point is player ability does not equal character ability) than the player would. At that point it's just semantics. It's less emersive but just as functional.

 

[Blue]

Ok, so we have:
A. The rules do say that the DM should only call for a roll if there is a meaningful cost for failure.
B. There is, in this hypothetical situation, no cost for failure.
C. The player asks to make a check.

Because there is no cost of failure, and the rules tell us only to call for a check if there is, the rules suggest that the DM should not allow the roll in this case.

Saying that “the player is asking for a chance of failure” is meant to express this, not to suggest that, because the player asked for a check, the DM ought to create a chance of failure where none existed before.

Which leads us to:

D. DM intentionally misinterprets a player's communication to be overly formalistic. He understands what the player is attempting to ask, but instead reacts like the grammer school teacher getting snooty when asked "Can I go to the bathroom" instead of "May I go to the bathroom", even though the intention is clear.

Or did you REALLY not understand what the player was asking for?

 

[Ovinomancer]

The issue I have with this (and several other comments here) is that it seems to assume an action is necessary in order to make an insight check. I find the action of the players is the exception and the checks are made as an interpretation of NPC actions. That's like trying to force players to describe actions before checking perception. Yes, there are active perception checks and mostly passive. That's how I generally treat insight.

I don't have a problem with "I want to use insight to..." because the character would know how to apply the ability better (possibly worse; the point is player ability does not equal character ability) than the player would. At that point it's just semantics. It's less emersive but just as functional.

It doesn't imply it -- it's how the game rules are written to work. This doesn't mean that you can't play it according to older editions -- clearly it works fine and you enjoy it, which is the only test that matters. But, yes, rolling insight can only happen if the DM asks you to, and then only in response to something a character's action declaration that has an uncertain outcome.

Personally, I find it odd that the PC's are asking for a check to get validation for having an opinion about something in the fiction, but okay.

 

[Ovinomancer]

Which leads us to:

D. DM intentionally misinterprets a player's communication to be overly formalistic. He understands what the player is attempting to ask, but instead reacts like the grammer school teacher getting snooty when asked "Can I go to the bathroom" instead of "May I go to the bathroom", even though the intention is clear.

Or did you REALLY not understand what the player was asking for?

No, it really doesn't, as I play this way and this doesn't happen at all, nor has it been reported as happening by those that follow this method. This is an assumption that relies on assuming poor play and bad behavior.

Requiring a goal and approach is not odd at all -- you're asking what the PC is doing in the fiction so you can adjudicate the reaction appropriately. I'll use an old chestnut -- imagine a door with a handle covered in contact poison. The PC asks for a perception check on the door. Did they touch the knob? I don't know, and neither could you -- any answer is an assumption about what the PC did. In asking for an approach, we'll be clear as to whether the assumed search of the door involves touching. Pretty simple, really. Everything else is being consistent about not assuming actions (even when it should be easily done). Surely, you're not suggesting that consistent DM techniques are poor play, are you?