Genes, Brains, and Human Potential The Science and Ideology of Intelligence

174 HOW THE BRAIN MAKES POTENTIAL

there is another prob lem. It takes thirty to one hundred milliseconds for
sense receptors to respond and pass the signal on. So by the time we have
perceived an object in motion, it is no longer exactly where we now see it.
Th is makes actions like catching a ball more diffi cult than you think—
not to mention catching prey or avoiding predators, like our ancestors did:
actions need to be ahead of vision, as it were.
Fi nally, a single scene may contain dozens of objects. Th is is why
William James, in 1892, famously referred to the “blooming, buzzing
confusion” of sensory input. It has been described more recently as “an
onslaught of spatio- temporal change.”^5
For all these reasons, what we see (our percepts) cannot be reduced to
the sensory input. One argument is that some kind of inferential con-
struction must be involved. Irving Rock called it perceptual “prob lem-
solving,” or “intelligent perception.” Like the other intelligent systems we
have discussed so far, this means going beyond the information given.
So investigators have been astonished as the extent of the work done
by the visual system has become more obvious. In a 2012 article in the
journal Neuron, Richard Masland noted that “although the ret ina forms
a sheet of tissue only 200 ࢆm thick, its neural networks carry out feats of
image pro cessing that were unimagined even a few years ago.” Th ey sup-
port Donald Hoff man’s view that those of nomal vision are, in fact, vi-
sual geniuses naively unaware of their rich talents.^6 But how are such feats
achieved?

BEYOND THE INFORMATION GIVEN

Th e traditional mechanical model is based on a linear convergence of

recorded visual features or ele ments. And it demands some built-in rules

of pro cessing of some sort. However, the “inferencing” just mentioned

seems to be based on something other and more than that. Again, the

demand is for predictability (e.g., that a fuzzy input is really from a spe-

cifi c object with all its implications). However, as with cell metabolism,

and the intelligence of development and physiology, it is not so much the

surface pre sen ta tion of inputs that are informative. Rather it seems to be

the structure beneath the surface that yields the predictability.

This content downloaded from 139.184.14.159 on Tue, 17 Oct 2017 13:54:00 UTC

http://www.ebook3000.com