Fallacious but psychologically attractive inferences

I like to start with really simple theories and see where they go wrong. Recently this lead to an interesting combination of experiences

When I say:

People are justified in making those a priori inferences which are both necessarily truth preserving and psychologically compelling for normal humans.

People say: But reasoners do have initial justification for accepting certain attractive but ultimately fallacious arguments e.g. tricky arguments for the existence of God.

But when I take out the requirement of being genuinely truth-preserving and say...

People are justified in making those a priori inferences which are psychologically compelling for normal humans.

People say: But what about those bad but psychologically compelling inferences like inferring the consequent?

So which is it (do you think)?

When someone makes a psychologically compelling but invalid inference like the gambler's fallacy or inferences about naive set theory are they:
a) justified, (though presumably thinking about the right questions may later give them justification for changing their mind) or
b) unjustified
c) somehow there's a difference between the gambler's fallacy and naive set theory in this regard

I don't have a dog in this race, or think anything deep is going on here, but I'd really like to know which way normal language intuitions go.