Player Trust (Definitive Guide)
Players trust sense motive checks, can you blame them?
The referee must present the world truthfully, but that obligation ends once they begin speaking through an NPC. An NPC may be mistaken, or worse, a liar.
Desperate, grasping for purchase, the player calls for a wisdom roll on reflex. "Can I at least read their body language? Does my character think that it was sus when they bit their lip? Do we have a spell slot for a Zone of Truth?" - to me, this is a less than ideal development in the conversation.
For many years, I was a wisdom roll type of GM. Later, a body language description improvisor. Finally I just let the players stew in uncertainty. Hardass refereeing has its merits, but we can do better. We can write scenarios with logical, playable information to address this challenge just like any other.
It's all about whether the PC's understand the liar's motivations, or not.
If the NPC's behavior does not match known motivations, that should tipoff the players that they are being deceived.
Ex: I know Johnny is in $300,000 of debt. So why is he offering to take me on a yacht cruise?
From this baseline, you can weave more sophisticated lies.
Care should be taken NOT to build a labyrinth of mirrors. The more elaborate the con, the easier to stumble onto clues of the truth at the very least.
If the NPC's motivation is unknown, that lack should be conspicuous. (I've straight up told players, "You can't explain why this person is doing this.") There should be evidence of a cover up. 1
If the NPC presents a false motivation, it should not hold up to scrutiny. It should contradict something the PC's learned earlier or will learn soon. A good way to signal this is to suggest 2 possible motives that are mutually exclusive.
In all cases, it should be clear what the PC's risk by trusting the NPC.
"Okay, so you're going to get on this boat. You'll be alone with Johnny 100 miles away from the shore. Sound good?"
If it wasn't obvious, this is just a restatement of Information, Choice, Impact applied to social deduction. But rather than conversational vibes or body language interpretation, we provide straight forward evidence and logical inferences.
The corollary to this method is that NPC's whom the players know the motivations of and act as expected are extremely useful to the party: trusted confidants, workman-like partners, or pawns in the PC's hands.
The final step is to dial in the genre expectations by moderating how nasty the betrayals can be.
Heroes get betrayed all the time, but are adept at scrapping it out. A hero can afford to trust a sketchy source for the chance to save the day. Worst case, they waste time fighting some other bad guys.
Non-hero protagonists are vulnerable to betrayal, but have more incentive to verify anything out of the ordinary. The need to work with people aligned with your goals is acute. If it sounds too good to be true, it is.
Of course, the default state of most NPC's motivations is unknown or even undefined, though they are not necessarily trying to conceal anything. Not ideal, but such is the GM's burden. I think it's plain bad practice to lie with an NPC who's motives you have not defined and foreshadowed. That is just another form of "gotcha" GMing. Some effort will need to be put in presenting landmark and hidden tier info surfacing the NPC's motives. Eventually, players will learn to inquire about an NPC's motives directly.↩