ubiquitous to the relevant environment,
are extremely beneficial in learning the
environment. As such we will call them
“useful examples.”
Through specific interactions and
based on the previous tendencies of the
system certain general properties are
corroborated by “useful examples” of
these properties appearing in the envi-
ronment. This corroboration leads to
the formation of an achieved set of rep-
resentations in the system. Specific en-
counters with the environment start
with a deconstruction via a detection of
similarities to the useful examples, em-
bodied in the achieved set. After this de-
tection there is a refitting and fine tun-
ing to the specific elements of the event.
Having identified the specific elements
of the event, there is now a reconstruc-
tion of the event together with an asso-
ciation to other contextual factors to a
functional/meaningful event that can be
appropriately reacted to and added to
the memory of the system. [The two
phase are chronologically mingled;
both happen simultaneously. Every en-
counter is a specific encounter even
while the cognitive system is still build-
ing its understanding of the general
properties. Also in many cognitive sys-
tems it is not clear if the process of de-
fining new general properties ever
comes to a complete stop. “Young” cog-
nitive systems are less fluent in working
with the general properties, and it is
harder to teach an “old” cognitive sys-
tem new tricks, but all interactions with
the environment have elements of both
phases.]
In this discussion of cognitive sys-
tems, we will start with the visual
system. Visual systems are built ac-
cording to the niche in which they are
used. A bee’s view of flowers is different
from ours. The limits of sensitivity to
wavelength and contrast are to a great
extent built in. However, seeing is not
merely a matter of light sensitivity. We
see because we have learned how to do
so; we know what things to “look” for
and what they mean. This knowledge is
an understanding of the natural context
of vision and the general properties of
the visual stimulus that we encounter.
As an example, take a look at Figure 2
where our knowledge will not help us.
A painting from the brush of Salva-
dor Dali, this picture shows how com-
plicated it is to know what we are look-
ing at when we do not have the natural
context to tell us what we should look
for. The title may give you a hint, “Ap-
parition of a Face and Fruit Dish on a
Beach.” Once we reveal that the “appa-
rition” hides the figure of a dog, it is
doubtful if you could ever again look at
this picture without seeing man’s best
friend. Our preknowledge is not of spe-
cific aspects of every stimulus but
rather of their general properties. The
knowledge of the general properties of
visual stimuli is not something that we
are born with; rather, it is acquired
through the natural exposure to the en-
vironment. The process starts with the
natural visual tendencies of the visual
apparatus and the innate bias that we
are born with toward certain things,
such as complex stimuli and move-
ment; however, these innate aspects are
not enough. Given this tendency, the vi-
sual system, in its first stages of devel-
opment, builds up the tools that later
will allow the act of vision.
The need for this priming stage and
the existence of a critical period for its
occurrence is easily seen in people who
have had blocked corneas from infancy.
FIGURE 2
“Apparition of a Face and Fruit Dish on a Beach”: Salvador Dali (1970).
Such people, if their cornea is removed
too late, will remain functionally blind;
for although they can react to light, they
cannot resolve the images they see [5]
(much like living in a painting by Dali).
In those of us free to see the world, the
process of learning the general proper-
ties is unsupervised and is based on the
existence of useful examples for the
general properties of naturally encoun-
tered visual stimuli.
The general properties are those
things that are both ubiquitous to the
different stimuli and appear in many
different meaningful contexts. In vision
these would be things such as edges or
regularities of scale [6]. Their high fre-
quency and usefulness very naturally
cause their corroboration, until eventu-
ally we have a visual system with an
achieved set of representations based
on the “useful examples” (stage 1 in Fig-
ure 1). This “achieved set” is what en-
ables, from then on, the resolution of
visual stimuli: the deconstruction and
reconstruction of the second stage,
which allows the acquiring of the
proper functional understanding of the
image and memory.
Vision starts with the general prop-
erties of the image rather than its par-
ticulars. In our representation this is the
phase of detection (second half of Figure
1). Sight, as opposed to taking a picture,
16
CO M PLEXITY
© 2001 John Wiley & Sons, Inc.