Information Technology Reference
In-Depth Information
None of the models translated their outputs into familiar economic terms beyond
the SKIN model's use of market pricing. To provide insights into the role played by
innovation in a whole economy or society, the models surveyed here need more con-
nection between the results of innovation on the one hand, and the future ability of
agents to innovate on the other. Knowledge and technologies serve purposes, while
both human agents and firms have needs. Failure to meet their basic needs (food,
capital) will make it harder to continue to engage in practices such as experimenta-
tion, learning from others and trading, thus slowing down the rate at which innova-
tions are generated and diffuse. So far, innovation models have omitted the needs of
their human components.
A key feature of technologies and practices is their ability to support more than one
attribution of functionality. Technologies and routines developed for one purpose may
be reinterpreted and developed to serve a new purpose, one not foreseen during earlier
development and present seemingly only by accident. This is the process known in
biology as exaptation. Villani, Bonacini, Ferrari, Serra and Lane [36] hold exaptive
bootstrapping to be responsible for the continual growth in the number of types of
thing, and present an early attempt at a model in which agents' cognitive attributions
of functionality to artefacts enables the representation of this process. Several of the
models surveyed above can simulate an innovation in the function or value of one
thing, due to the appearance or disappearance of another, but they do not distinguish
these changes in actual functionality from changes in agents' awareness of functional-
ity. While the models are good at representing innovation as recombination, they are
less clear at representing innovation as reinterpretation.
Introducing cognition raises the roles of imagination, analogy and metaphor in
generating innovations. The range of logically possible combinations of all our cur-
rent technologies is vast, and most of them seem nonsensical or unviable. We only
have time to try out a tiny fraction of the combinations, so how do we manage to find
any useful ones? Models that treat component technologies or beliefs as indivisible
base units will lack a basis for their agents to anticipate how well two never-before-
combined base technologies are likely to fare when combined for the first time.
Gavetti, Levinthal and Rivkin [37] simulate the use of analogical reasoning in gener-
ating new strategies. Attention to reasoning processes and the circumstances in which
they work well will be issues for modellers of other forms of innovation as well.
Acknowledgements.
We acknowledge the support of the SIMIAN project
(www.simian.ac.uk) funded by the UK Economic and Social Research Council's
National Centre for Research Methods. A more detailed version will be included in a
forthcoming topic [24].
References
1.
Lazer, D., Friedman, A.: The network structure of exploration and exploitation. Adm. Sci.
Q 52, 667-694 (2007)
2.
Silverberg, G., Verspagen, B.: Self-organization of R&D search in complex technology
spaces. J. Econ. Interac. Coord. 2, 195-210 (2007)
