A complex model may be more realistic yet at the same time more uncertain. This leads to the ironic situation that as we add more factors to a model, the certainty of its predictions may decrease even as our intuitive faith in the model increases. Because of the complexities inherent in natural systems, it may never be possible to say that a given model configuration is factually correct and, therefore, that its predictions will come true. In short, the “truer” the model, the more difficult it is to show that it is “true.”Oreskes is speaking about scientific models, specifically those about ecological systems. Her insights have significance for technical models used by policy-makers (e.g., NEPA or CEQA compliance, air quality conformity, or travel demand models). As opposed to purely scientific models (models built by scientists to synthesize data, make predictions, or guide future work), these ones have explicit policy ends. Expensive decisions are made on the basis of them. Surely the limits of these models are better understood?
What happens when we trade in an old modeling paradigm (four step model) for a new one (activity-based models). Is the substantial cost involved in new model development worth the investment? How do we know? Do we think that just because the new model has more behavioral realism that it's somehow better? As Oreskes suggests, adding complexity has diminishing returns (just like in larger social systems [Tainter, 1990])...
Are there simpler models/heuristics that we could employ at far lower costs? The implications of considering these questions for modeling practice are substantial.
Oreskes, N. (2003). The Role of Quantitative Models in Science. Models in Ecosystem Science. C. D. Canham, J. L. Cole and W. K. Lauenroth (Eds.). Princeton, NJ, Princeton University Press.
Tainter, J. A. (1990). The Collapse of Complex Societies. Cambridge, UK, Cambridge University Press.
