The thing is, if I see a fire and put my hand in it, I'll experience pain. An MRI scanner would be able to detect certain parts of my brain being excited.
If my brain was then artificially stimulated in the same region, no doubt I'd feel the same pain again. However, none of this means that the fire wasn't real in the first place.
Of course, it's interesting because perhaps the fire isn't real and it's just some fringe part of the brain playing up, but I'm not sure how you'd go about designing an experiment to work that out (i.e. whether the circuits exist because the fire is real, or whether they've evolved some anomalous behaviour).
Most likely somatosensory thalamocoritcal networks (the anterior parts of the parietal lobe and the corresponding thalamus). Maybe the insula, which is involved in visceral sensations as well.
https://www.york.ac.uk/news-and-events/news/2015/research/ps...