Once upon a time, science fiction was full of new hardware “plugged into the brain” providing wonderful capabilities. This fashion has died down in recent years, but I was always irritated that the stories glossed over what I thought was the most difficult and interesting bit of the mind-brain-hardware interface.
A good, simple example is a clock. What do we need to do to “interface” a human mind with a digital watch? Let’s be more precise: we want a person to have an innate and instant sense of what time it is. Note that this is an absolute, not a relative thing. We are not talking about accurately sensing a duration of some minutes or hours, but of having a direct sense of the time in the home time zone (for me this would be UTC). The “sensor” is trivial: a watch and 10 year battery probably cost less than 5p, but how would we “wire it in”?
Let’s look at another sense: hearing. This is admirably simple in that a number of hairs of different lengths resonate in the cochlear, and a few hairs of the same length active imply one frequency in the sound input. Job done? No, there is a whole auditory cortex that processes the basic inputs and provides expectations which then interact with input data. And that’s not all, lots of other bits of the brain get in on the act too. This isn’t all to do with processing a noisy signal, it is also to do with what the sounds represent to an animal living in the real world. Let’s say that we don’t need all that: we just want a sense of the absolute time. We have the plug, now where is the socket ?
Absolute time is likely to be a learned thing: mammals generally probably have no sense of it. We learn to tell the time from clocks as infants, so there is no pre-existing bit of brain to plug into. We could, of course, hack into the visual cortex and put little red digits at the bottom of the field of view (in principle), but that’s not a direct sense of time at all. We can re-purpose any existing sense, and just hijack the lobes that exist already (such as using a matrix of vibrating pins on the skin to simulate a visual sense), but the goal is to create a completely new sense.
We could just spray a handful of electrodes over some likely bit of cortex in a new-born, connect them to a clock, and then train the child to integrate the weird sensations it gets from the electrodes together with clock time when it learns to tell the time. [Obviously morally indefensible, but this is a thought experiment.] That might well work, but it’s not going to get us very far as a stepping stone to creating “wonderful and new senses”. It does give a useful clue though: development. The development of the brain in infants is a generic tool of great power.
What we need is a new lobe in the brain, a “time lobe”, which accepts the sensor input and presents it to the mind (y es: the mind) in a way that is compatible with the whole worldview. If we knew how to do that, could we surgically insert such a lobe into an adult to give this new capability ? It doesn’t look like we could, because the mind develops as a whole, but maybe an adult to learn to use the capability, in the same way that some people do recover (to some extent) from strokes.
OK, now the interesting bit: how would we go about designing such a lobe ?
The thought experiment with newborns gives one route: we artificially evolve the design using neural nets in silico, and then [somehow] make or grow the lobe in hardware/squishware. But we still have a complete unknown in that we don’t know what we are interfacing with in the rest of the brain. There isn’t a little homunculus in there, there really isn’t. The mind is a collective emergent property, and we can see now that it’s not surgery or electronics which is the issue here, it is something philosophically rather deeper.
The Whole Starfish 