A few years ago I was in a computational neuroscience research group. We built mechanistic models of specific brain areas and neural pathways and ran them in neurosimulators. We used them to explain psychophysical experiments and to predict which neurons would be active during them. At one point we even wired several of these models together to simulate an agent in a 3D environment that had to solve simple tasks. I don’t think anyone in the group ever believed what we’d built was conscious, or that the simulation was the real thing.