Item

Abstract

I propose an experimental method to test various hypotheses on consciousness. Inspired by Sperry’s observation that split-brain patients possess two independent streams of consciousness, the idea is to implement candidate neural mechanisms of visual consciousness onto an artificial
hemisphere and test whether subjective experience is evoked in the device’s visual field. In contrast to modern neurosynthetic devices, I show that mimicking interhemispheric connectivity assures that authentic and fine grained subjective experience arises only when a stream of consciousness is generated within the device. It is valid under a widely believed assumption regarding
interhemispheric connectivity and neuronal stimulus-invariance, as described below. Interhemispheric
connectivity in low/mid-level visual areas is restricted in the sense that cross-hemispheric neuronal projection is observed only in neurons that retinotopically correspond to the vertical meridian. We may say that the retinotopic representation is only stitched together at the boundary
of two visual hemifields. Interhemispheric connectivity beyond the parafovea exists only in high-level cortical areas. On the other hand, granularity of visual information decreases in higher visual areas in the form of increased stimulus-invariance. The critical question is whether the
informational content in high-level areas is sufficient to support conscious vision. The “Intermediate Level Theory of Consciousness” by Jackendoff states otherwise and claims that it does not play a central role in conscious vision. If this is true, in combination with the hierarchical
properties of interhemispheric connectivity, we need to acknowledge that subjective experience and verbal report of bilateral vision arise, not because all necessary information for conscious vision is inter-exchanged between the hemispheres, but because two potentially independent
intra-hemispheric streams of consciousness are interlinked. The former scenario is denied because extra-parafoveal visual information represented in low/mid-level visual areas cannot be transmitted over to the other hemisphere in its original resolution. Likewise, under the above assumption, the only possible way we may subjectively experience authentic objects in the device’s visual field
is that a stream of consciousness is generated within the artificial hemisphere and is interlinked to our own. Simple influx of information from the artificial to the biological hemisphere would not be sufficient. Hence, we may construct a valid test for machine consciousness and use it to explore
the neural correlate of consciousness by means of analysis by synthesis. Interestingly, although fully replacing a cortical hemisphere is something of the far future, a minimal experiment can be conducted with today’s technology, for example, by establishing a brain-machine interface solely between populations of high-level face neurons. If a stream of consciousness is generated within
a device, we should be able to construct a case where two objects presented in the device’s visual field are distinguishable by visual experience, but not distinguishable by what is communicated through the brain-machine interface. Finally, I discuss the alternative assumption where high-level visual information is sufficient for conscious vision and show that the proposed test of consciousness can be adapted to cover this case by incorporating knock-out paradigms. Together, I provide
an exemplar neural mechanism of subjective bilateral vision that passes the proposed test.