Let's start with a basic premise: the player says, "Can I tell if the NPC is lying?" and the GM says, "Let me have a Wisdom(Insight) check...". Some people feel that the player should have to abide by the result.
But let's run a thought experiment: suppose the player says, "Eh, never mind." Do they still have to roll?
Or, what if the player never asks? What if the player just listens to the GM and nods. The GM might say, "Do you believe him?" And the player says, "Not sure." Should the player be forced to roll, anyway, because the GM wants to know what the player (and the character) are thinking? That doesn't make any sense to me.
This is why I think rolls should only be called for in response to an action declaration that has consequences.
So I could see the player saying, "I'm going to keep asking questions and see if I can get the NPC to contradict themselves, so that I know if they are lying." And the GM might say, "Um, ok, but I'll need a roll, and if you fail you are going to piss them off..."
I don't follow the context of what these hypotheticals mean for your argument. Can you explain your thought experiment better?
Base premise, I would say there is an expectation that if the result of the game play results in something affecting the character, the player is at least expected to have the
character acknowledge that result. If that expectation isn't in place, then everything falls over. So I think it is at least that "my character (does|doesn't) tell that the NPC is lying or not" is default stakes in your first example. Not that the
player has to
abide. The player can know that I am hamming this NPC up as the shadiest guy in town despite their character's failed Insight roll and act accordingly, but acting accordingly is still through the lens of the character. Maybe the player decides to roll with it and play their character as the fool because that's interesting, or maybe the player starts scheming for another way to get better information, because the player still has overall agency on their character's actions and they still want more information. That's fine, as long as we're agreeing that an in-game thing happened (even if it's abstract), and we should proceed play under the agreement that it changes something about the character (in this case, their confidence in what they see or hear).
Before the "didn't roll, didn't ask" situations (because this is where I don't understand the thought experiment), let's also admit that the game rules solve a situation where maybe the player can't read me, or
I'm a really good liar or actor. The player has the right to try to have their character have some in-game knowledge that isn't/can't be conveyed by the people at the table, it's right there in this game's rules. And, again the implicit contract is everyone follows the cause and effect of the game rules
in-game. Or further still, maybe I'm just a terrible actor (which I, personally, am) or there's no direct information for the player to make a decision about, because I just narrated the gist of a conversation rather than starting an in-game volley of conversation. Valid situations, right? Thus, it's valid that the character should be allowed to have information the player does not. So, Insight (for example) has a game purpose. It'd be wrong to deny players that chance, if they're playing a game that has made those accommodations (or concessions, or however you want to think about it).
If the player asks the original question and chooses not to roll for some reason, that's outside of the scope of the game rules. Same with just nodding along and never testing the situation. The player can think whatever they want, same as before, and again, act accordingly through the lens of the character --- a character who does not have any particular insight (heh) into this situation. So, back to the thought experiment, I agree with you that the GM asking the player to roll, to understand what the player is thinking, makes no sense --- the game rules test the character, not the player, and the character has opted out, basically
. I also agree that rolls should only be called for* in response to an action declaration that has consequences, but I don't see where these situations have bearing on the validity of these skills as a means of conveying information to the character.
(*): I am fine with passive skills, myself, like if a master investigator is trying to figure out if a guileless child is lying, it would be obvious to them. But that's just a kind of test where we believe we can skip the roll, because failure here should be impossible.
I guess you could hypothesize a system where a character's beliefs and thus consequent actions
must be
defined by the results of rolls, removing the player, but I don't think anyone here is advocating for such a game. Now I'm wondering if you are hypothesizing that. Is that a real perception of how people use Sense Motive/Insight/etc.?
