Information Technology Reference
In-Depth Information
is no path to get there, and it is conceptually disconnected from my existing net-
works. If it strikes the right balance with respect to the receiving individual, the
new work extends his networks too, and forms a foundation for further extensions,
further curious explorations in our continuous strive for novelty.
Novelty in creativity is often divided into Boden's private P-creativity and histori-
cal H-creativity (Boden
2004
). Is it new for me, or new for a whole cultural domain,
or even all of humanity? Novelty in relation to myself includes the expansion of
my networks to gradually encompass new areas. Tool networks and conceptual net-
works both contribute to this process. For some artists the tool networks lead into
new areas. For others, the ideas take the lead, and the tool networks are expanded as
needed. The former is more common in music, as an abstract art, where tools and
techniques play an important role. The latter may be more common in contemporary
fine arts, where real-world concepts often are of primary importance.
When a conceptual representation develops, it expands to new areas in interplay
with continually accumulating cultural input, pushing the individual artist to ex-
pand further, past what has been covered by others. Nobody has access to a global
database, but only to the fraction of human culture contained in the artist's mem-
ory, defining his cultural networks and subspaces. To put it simply, his exploration
of spaces happens in interplay with what he remembers from what he has seen or
heard before.
8.3.6 The Model in Context
There are empirical studies of creative processes within psychology research (e.g.
Barron
1972
, Konecni
1991
) and abundant recollections on the subject from artists
(e.g. Barron
1997
, Klein and Ödman
2003
). These accounts from artists are some-
times contradictory and personal, and concentrate on rhapsodical and very personal
details of particular processes. Artists not aware of existing psychological theories
of creativity, may not be able to give a systematic account of what is happening.
They sometimes reconfirm well-known phenomena and myths, but hesitate, con-
sciously or not, to reveal their creative techniques, or are not able to verbalise the
mechanisms of their own creativity. Some seem to preserve the romantic mystery
around creativity. And since not all researchers have first-hand access to these pro-
cesses (since they are not professional artists themselves) computational implemen-
tations directly derived from artists' processes are rare, with a few notable excep-
tions. Harold Cohen's autonomous painting program
AARON
, is based on his own
analysis of how he went about composing and painting a picture, from sketching
the first lines down to colouring. It works very well, within a limited domain of cer-
tain motives (McCorduck
1990
). In the field of music, David Cope is well-known
for his advanced computer-generated pastiches of classical music. Recently, he has
changed direction, and developed a composing program called
Emily Howell
(Cope
2005
), which develops its own musical style in a dialogue with Cope himself. In this
case external musical input and human feedback gradually helps form the stylistic
