place at some point after the reception of the visual signature and before
the words are uttered. One could claim that this funneling is not part of the
perception of Granny, but just part of verbalization. But it seems quite
unnatural to partition the process that way, for you could internally use the
information that it is Granny without verbalizing it. It would be very
unwieldy to handle all of the information in the entire visual cortex, when
so much of it could be thrown away, since you don't care about where
shadows fall or how many buttons there are on her blouse, etc.
Another difficulty with a non-funneling theory is to explain how there
can be different interpretations for a single signature-for example, the
Escher picture Convex and Concave (Fig. 23). Just as it seems obvious to us
that we do not merely perceive dots on a television screen, but chunks,
likewise it seems ridiculous to postulate that perception has taken place
when a giant dot-like "signature" has been created on the visual cortex.
There must be some funneling, whose end result is to trigger some specific
modules of neurons, each of which is associated with the concepts-the
chunks-in the scene.
Modules Which Mediate Thought ProcessesThus we are led to the conclusion that for each concept there is a fairly
well-defined module which can be triggered-a module that consists of a
small group of neurons-a "neural complex" of the type suggested earlier.
A problem with this theory-at least if it is taken naively-is that it would
suggest that one should be able to locate such modules somewhere within
the brain. This has not yet been done, and some evidence, such as the
experiments by Lashley, points against localization. However, it is still too
early to tell. There may be many copies of each module spread around, or
modules may overlap physically; both of these effects would tend to
obscure any division of neurons into "packets". Perhaps the complexes are
like very thin pancakes packed in layers which occasionally pass through
each other; perhaps they are like long snakes which curl around each
other, here and there flattening out, like cobras' heads; perhaps they are
like spiderwebs; or perhaps they are circuits in which signals travel round
and round' in shapes stranger than the dash of a gnat-hungry swallow.
There is no telling. It is even possIble that these modules are software,
rather than hardware, phenomena-but this is something which we will
discuss later. I
There are many questions that come to mind concerning these
hypothesized neural complexes. For instance:348Do they extend into the lower regions of the brain, such as the
midbrain, the hypothalamus, etc.?
Can a single neuron belong to more than one such complex?
To how many such complexes can a single neuron belong?
By how many neurons can such complexes overlap?Brains and Thoughts