At some point, through mathematics or a sufficient resolution of simulation, we will be able to run software that achieves consciousness. Through rigorous proof, the first method will provide the basis for falsifiable conjecture and exploration of the phenomena of consciousness and subjective experience. The second path will require laborious, tedious, and methodical experimentation with the model to tease out the underlying math.

Consciousness exists. It ceases to exist when the brain is damaged or killed. We can predictably modify conscious systems with drugs and experiences. Particular types of injury and illness sometimes result in predictable consequences, letting us know things such as that the hippocampus is critical to memory, and that whatever hierarchy of semantic meaning is stored in the brain, neurons in the hippocampus represent ontological peaks in the mental landscape of constructed meaning.

With a sufficiently high resolution brain interface, you could record, replay, simulate, and share experiences between consciousnesses. If the mind is an emergent phenomena brought about within the confines of the skull, shared experience will be possible, up to the point that your subjective experience of being another person will be complete for the duration of the replay. If there's some outside factor playing a role in consciousness, it will be missing and likely prevent attempts at recording experience.

You haven't said how it can be falsified here, though. How sound a scientific footing is any of that really on then if we can't actually experiment?