Affording Design, Affording Redesign - Principia Designae: Pre-Design, Design, and Post-Design - page 93

and functional relations. Similarly, for the architects, the connections between the
continuing strings tended to be functional much more than for students. The dual
implications are that students are not adept at inferring the functional information
from the sketches, and that architects are adept at integrating the perceptual and the
functional when focusing to a new aspect of the design, and to continue thinking
about the new aspect in high-level functional ways.
Further analyses of the data provided a window on the origin and consequences
of unintended discoveries, that is, things designers see in their own sketches that
they had not intended. This phenomenon points to one of the key advantages of
diagrams (Suwa et al. 2001 ) . Designers drew them with one thing in mind, but later,
when inspecting their sketches, see new relations and properties and implications
once the diagram is in front of their eyes. They may see patterns in the locations of
structures that present a potential theme that will unite and integrate the compo-
nents, a perceptual discovery. They may see patterns of traffic or play of light,
conceptual discoveries.
Sketches are deliberately ambiguous early in the design process. Designers do
not want to be locked into particular structures and shapes and spatial relations at
first. The ambiguity fosters unintended discoveries as well as flexibility of
rearrangement. In fact, the protocol of one experienced designer showed clearly
that after he perceived a new arrangement of parts, he was more likely to make an
unintended discovery, and after an unintended discovery, he was more likely to
regroup the parts.
This positive cycle, of unintended discoveries generating reinterpretations and
reinterpretations generated new unintended discoveries seems general and produc-
tive. In a partner study, students were asked to generate as many interpretations as
possible for a set of ambiguous, suggestive sketches. Those who spontaneously
adopted a strategy of mentally reorganizing the parts of the sketch generated
significantly more new interpretations than those who didn't. Other strategies
were not as effective. This positive cycle also suggests simple ways to foster design
creativity, or possibly creativity in other domains: to encourage and practice
rearranging, reconfiguring parts of a whole. More abstractly, this would translate
to changing the relationships among the components, and then interpreting them.
Generating new interpretations is an essential part of creativity, the divergent
part. But especially in design, creativity needs also to be focused to an end or an
outcome. That requires convergent thinking. Further research investigated that,
again using the paradigm in which students are asked to generate new interpreta-
tions for ambiguous sketches (Suwa and Tversky 2003 ) . In that study, students,
both designers and non-designers, were first given a remote associates test, requir-
ing then to find a relationship between distant word concepts. This test requires
focusing on the aspects of the concepts that are abstract or aspects that connect the
concepts. Students who scored high on the remote associates test also produced
more interpretations for the ambiguous figures. These two predictors of generating
new ideas, reinterpretation and remote associations reflect the two sides of creative
thinking: reorganization and connection, divergence and convergence. In the ser-
vice of design, the interaction of this dual set of processes was termed Constructive