There's a lot that can be done purely mechanistically, if you're super concerned about consciousness interfering with accuracy. For example, in the extreme, fM

There's a lot that can be done purely mechanistically, if you're super concerned about consciousness interfering with accuracy. For example, in the extreme, fMRIs show different activation patterns for people with different developmental disabilities. It's extremely clear and not at all subjective.