Enter Brainport, a seemingly matrix-inspired piece of technology that translates visual information from a video camera atop the head into a series of electrical impulses delivered to the tongue. The idea of sensory substitution behind Brainport has been around since the 1960s. The user learns the variations in pulses as a new language of sight, and translates them into information about the surrounding visual scene. The learning process is surprisingly fast - most users are able to recognize objects within hours. Erik Weihenmayer, who has been completely blind since age 13, has successfully used this technology to climb a rock wall.
We can take this one step further and arrive at a brain in a vat scenario where a brain deprived of all sensory input from the body is fed a portrait of sensory reality through pre-processed electrical impulses (We don't yet have computational power or knowledge enough to translate real world data into something parse-able by the brain that doesn't have to be learned as a language like in Brainport, but the thought experiment is valuable).
So what's the difference between sensory substitution and the brain in a vat? I think it lies partly in the location at which each technology intercepts the creative process of human perception (the creative element is also open to debate). Taking vision for example, we receive far more bits of information than the brain uses to construct a functional image of the world; much of the excess information is lost to meet the capacity of the optic nerve, and then a process of reconstruction occurs once the information reaches V1 (the slightly higher-capacity primary visual cortex). There is a creative aspect to this process of reconstruction. What heuristics govern it? To what extent is the process biased towards previous experience (do those heuristics come from probabilities based on other visual encounters)?
 |
| The bottleneck in visual information's pathway from the eye to the brain: |
The idea that sensory substitution intercepts information processing somewhere in the middle of the hourglass raises an important question regarding the heuristics governing reconstruction. Could a sensory substitution method work if someone had never possessed the sense necessary to learn those rules of the world? Climber Erik Weihenmayer was not blind from birth - he possessed ideas of space and shape from his partially sighted years.
A question that gets me is whether or not these heuristics are learned (they are ontologically independent principles of the outside world to which the brain molds) or whether the biology of the brain predetermines this structure and imposes it on otherwise chaotic data. Space as we know it: an artifact of the way information is processed in V1? I think sensory substitution may provide a means to bring us a little closer to answering that question.
An experiment: use sensory substitution to deliver condensed data to a patient who has been blind from birth. If the brain's inherent biology governs the construction of a portrait of reality, then unless this patient's biology is pathological, the creative process should theoretically still occur. If the heuristics are derived entirely from previous experience then a different picture entirely or no picture at all would result..
Thoughts?
S
