University of Alberta Dictionary of Cognitive Science: Bonini's Paradox

Bonini’s paradox is the name given to the problem that emerges when a model of a phenomenon is just as hard to understand as the phenomenon that it is supposed to explain. Dutton and Briggs (1971) point out that if the computer simulation researcher “hopes to understand complex behavior, he must construct complex models, but the more complex the model, the harder it is to understand.”

Bonini’s paradox is particularly relevant to connectionist cognitive science (Dawson, 1998), because while it is easy to create PDP models of phenomena of interest, it is much harder to understand how these networks actually function. Mozer and Smolensky (1989) have noted “one thing that connectionist networks have in common with brains is that if you open them up and peer inside, all you can see is a big pile of goo”. To contribute theories to cognitive science, connectionist researchers have to find some way out of Bonini’s paradox.

References

Dawson, M. R. W. (1998). Understanding Cognitive Science. Oxford, UK: Blackwell
Dutton, J. M., & Briggs, W. G. (1971). Simulation model construction. In J. M. Dutton & W. H. Starbuck (Eds.), Computer Simulation Of Human Behavior (pp. 103-126). New York, NY: John Wiley & Sons.
Mozer, M. C., & Smolensky, P. (1989). Using relevance to reduce network size automatically. Connection Science, 1, 3-16.

(Added October 2009)