I didn't see the post itself, but it sounds like Unconscious Thought Theory. The experimental evidence is pretty weak, and imo the theory as it stands is just too poorly specified to really test experimentally.
There is some evidence that offline processing matters for eg motor learning or statistical learning. I haven't looked in enough detail to know whether to trust it or not.
Given that the last Open Thread that was intended to be weekly didn't get much traffic, let's try a monthly version.