Skip to main content Accessibility help
  • Print publication year: 2010
  • Online publication date: July 2011

6 - Evolution, (sequential) learning and generalisation in modular and nonmodular visual neural networks

from Part II - The use of artificial neural networks to elucidate the nature of perceptual processes in animals



In general terms, modular systems are systems that can be decomposed in functional and/or structural independent parts. In cognitive science, modularity of mind (Fodor, 1983) is the controversial cognitivist view according to which human mind is made up of specialised innate modules. In contrast, the connectionist view tends to conceive of mind as a more homogeneous system that results from development and learning during life (see Karmiloff-Smith, 2000).

What is a module? Are modules innate? What is the relationship between modularity, robustness and evolvability? And what is the role of nonmodularity? These are only a few of the open central questions in the nature–nurture debate.

In a paper published in the journal Cognition, Gary Marcus (2006) deals with the vexata quaestio of modularity of mind from an enlightening point of view. He does not offer a detailed definition of modularity, nor an answer to the controversial issue of what is innate and what is learned during life. More simply, he identifies and contrasts two competing ‘hypothetical conceptions of modularity’, which would represent distinct perspectives about modularity of mind, implicitly present in the scientific literature and different in their implications: a ‘sui generis modularity’ and a ‘descent with modification modularity’. According to the former conception, ‘each cognitive (or neural) domain would be an entity entirely unto itself’; according to the latter conception, ‘current cognitive (or neural) modules are to be understood as being the product of evolutionary changes from ancestral cognitive (or neural) modules’.

Allman, J. M. & Kaas, J. H. 1971. A representation of the visual field in the caudal third of the middle temporal gyrus of the owl monkey (Aotus trivirgatus). Brain Res 31, 85–105.
Ancel, L. W. & Fontana, W. 2000. Plasticity, evolvability, and modularity in RNA. J Exp Zool B: Molec Dev Evol 288(3), 242–283.
Calabretta, R. 2002. Connessionismo evolutivo e origine della modularità. In Scienze della Mente (ed. Borghi, A. & Iachini, T.), pp. 47–63. Il Mulino.
Calabretta, R. 2007. Genetic interference reduces the evolvability of modular and nonmodular visual neural networks. Phil Trans R Soc B 362(1479), 403–410.
Calabretta, R. & Parisi, D. 2005. Evolutionary connectionism and mind/brain modularity. In Modularity. Understanding the Development and Evolution of Complex Natural Systems (ed. Callebaut, W. & Rasskin-Gutman, D.), pp. 309–330. MIT Press.
Calabretta, R. & Wagner, G. P. 1997. Evoluzione della modularità in reti neurali. In Atti del Congresso Nazionale della Sezione di Psicologia Sperimentale (ed. Nigro, G.), pp. 35–36. Associazione Italiana di Psicologia (AIP).
Calabretta, R., Wagner, G. P., Nolfi, S. & Parisi, D. 1997. Evolutionary Mechanisms for the Origin of Modular Design in Artificial Neural Networks. Technical Report # 51, Yale Center for Computational Ecology, Yale University. Also in Abstract Book of the First International Conference on Complex Systems (ed. Y. Bar-Yam). Nashua (NH), 23 September, New England Complex Systems Institute.
Calabretta, R., Nolfi, S., Parisi, D. & Wagner, G. P. 1998a. A case study of the evolution of modularity: towards a bridge between evolutionary biology, artificial life, neuro- and cognitive science. In Artificial Life VIAdami, (ed. C., Belew, R., Kitano, H. & Taylor, C.), pp. 275–284. MIT Press.
Calabretta, R., Nolfi, S., Parisi, D. & Wagner, G. P. 1998b. Emergence of functional modularity in robots. In From Animals to Animats 5 (ed. Pfeifer, R., Blumberg, B., Meyer, J.-A. & Wilson, S.W.), pp. 497–504. MIT Press.
Calabretta, R., Nolfi, S., Parisi, D. & Wagner, G. P. 2000. Duplication of modules facilitates the evolution of functional specialization. Artificial Life6, 69–84. Also Technical Report # 59, Yale Center for Computational Ecology, Yale University.
Calabretta, R., Di Ferdinando, A., Keil, F. C. & Parisi, D. 2002. Sequential learning in nonmodular neural networks. Abstract in Proceedings of the Special Workshop on Multidisciplinary Aspects of Learning, European Society for the study of Cognitive Systems, Clichy, 17–19 January.
Calabretta, R., Di Ferdinando, A., Wagner, G. P. & Parisi, D. 2003. What does it take to evolve behaviorally complex organisms? BioSystems 69, 245–262.
Calabretta, R., Di Ferdinando, A. & Parisi, D. 2004. Ecological neural networks for object recognition and generalization. Neural Processing Letters 19, 37–48.
Calabretta, R., Di Ferdinando, A., Parisi, D. & Keil, F. C. 2008. How to learn multiple tasks. Biol Theory3, 1 (MIT Press). Also Technical Report 29 July 2003. Institute of Cognitive Science and Technologies, Italian National Research Council (CNR).
Cosmides, L. & Tooby, J. 1994. The evolution of domain specificity: the evolution of functional organization. In Mapping the Mind: Domain Specificity in Cognition and Culture (ed. Hirschfeld, L. A. & Gelman, S. A.). MIT Press.
Di Ferdinando, A. & Calabretta, R. 2000. Evoluzione di reti neurali modulari per compiti di “what” e “where”. In Atti del Congresso Nazionale della Sezione di Psicologia Sperimentale, Associazione Italiana di Psicologia (AIP) (ed. B. Pinna), pp. 90–92. Carlo Delfino Editore.
Di Ferdinando, A., Calabretta, R. & Parisi, D. 2000. Evolution of modularity in a vision task. Technical Report # IP/CNR SNVA 00–2. Institute of Psychology, Italian National Research Council (CNR), Rome.
Di Ferdinando, A., Calabretta, R. & Parisi, D. 2001. Evolving modular architectures for neural networks. In Connectionist Models of Learning, Development and Evolution (ed. French, R. M. & Sougné, J. P.), pp. 253–262. Springer-Verlag.
Elman, J. L., Bates, E. A., Johnson, M. al. 1996. Rethinking Innateness. A Connectionist Perspective on Development. MIT Press.
Fodor, J. A. 1983. The Modularity of Mind. MIT Press.
French, R. M. 1999. Catastrophic forgetting in connectionist networks. Trends Cogn Sci 3(4), 128–135.
Ghirlanda, S. & Enquist, M. 2003. One century of generalization. Anim Behav 66, 15–36.
Glover, S. 2004. Separate visual representations in the planning and control of action. Behav Brain Sci 27, 3–78.
Gould, S. J. 1997. Evolution: the pleasures of pluralism. N Y Rev Books, 26 June.
Gould, S. J. & Vrba, E. S. 1982. Exaptation – a missing term in the science of form. Paleobiology 8(1), 4–15.
Holland, J. H. 1992. Adaptation in Natural and Artificial Systems: an Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. MIT Press.
Hummel, J. E. 1995. Object recognition. In Handbook of Brain Theory and Neural Networks (ed. M. A. Arbib), pp. 658–660. MIT Press.
Kaas, J. H. 1984. Duplication of brain maps in evolution. Behav Brain Sci 7(3), 342–343.
Karmiloff-Smith, A. 2000. Why babies' brains are not Swiss army knives. In Alas, poor Darwin (ed. Rose, H. & Rose, S.), pp. 144–156. Jonathan Cape.
Keil, F. C. 1999. Nativism. In The MIT Encyclopedia of the Cognitive Sciences (ed. Wilson, R. A. & Keil, F. C.), pp. 583–586. MIT Press.
Jacob, P. & Jeannerod, M. 2007. Précis of ways of seeing, the scope and limits of visual cognition. PSYCHE13(2), April.
Marcus, G. 2006. Cognitive architecture and descent with modification. Cognition 101(2), 443–465.
Maynard Smith, J. 1978. The Evolution of Sex. Cambridge University Press.
Milner, A. D. & Goodale, M. A. 1995. The Visual Brain in Action. Oxford University Press.
Milner, A. D. & Goodale, M. A. 2006. The Visual Brain in Action. 2nd edn. Oxford University Press.
Milner, A. D. & Goodale, M. A. 2008. Two visual systems re-viewed. Neuropsychologia 46(3), 774–785.
Mitchell, M. 1996. An Introduction to Genetic Algorithms. MIT Press.
Ohno, S. 1970. Evolution by Gene Duplication. Springer Verlag.
Parisi, D., Cecconi, F. & Nolfi, S. 1990. Econets: neural networks that learn in an environment. Network 1, 149–168.
Pereira-Leal, J. B. & Teichmann, S. A. 2005. Novel specificities emerge by stepwise duplication of functional modules. Genome Research 15, 552–559.
Pereira-Leal, J. B., Levy, E. D. & Teichmann, S. A. 2006. The evolution and origins of modularity: lessons from protein complexes. Phil Trans R Soc B 361(1467), 507–517.
Pinker, S. & Prince, A. 1988. On language and connectionism: Analysis of a parallel distributed processing model of language acquisition. Cognition 28, 73–193.
Plaut, D. C. & Hinton, G. E. 1987. Learning sets of filters using backpropagation. Comp Speech Lang 2, 35–61.
Rauschecker, J. P. 1998. Cortical processing of complex sounds. Curr Opin Neurobiol 8(4), 516–521.
Rueckl, J. G., Cave, K. R. & Kosslyn, S. M. 1989. Why are “what” and “where” processed by separate cortical visual systems? A computational investigation. J Cogn Neurosci 1, 171–186.
Rumelhart, D. & McClelland, J. 1986. Parallel Distributed Processing: Explorations in the Microstructure of Cognition. MIT Press.
Schwartz, R. 1999. Rationalism and empiricism. In The MIT Encyclopedia of the Cognitive Sciences (ed. Wilson, R. A. & Keil, F. C.), pp. 703–705. MIT Press.
Striedter, G. 2005. Principles of Brain Evolution. Sinauer.
Ungerleider, L. G. & Mishkin, M. 1982. Two cortical visual systems. In The Analysis of Visual Behavior (ed. Ingle, D. J., Goodale, M. A. & Mansfield, R. J. W.), pp. 549–586. MIT Press.
Wagner, G. P., Pavlicev, M. & Cheverud, J. M. 2007. The road to modularity. Nat Rev Gen 8, 921–931.
Wagner, G. P., Mezey, J. & Calabretta, R. 2005. Natural selection and the origin of modules. In Modularity. Understanding the Development and Evolution of Complex Natural Systems (ed. Callebaut, W. & Rasskin-Gutman, D.), pp. 33–49. MIT Press.
Werbos, P. J. 1974. Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences. PhD thesis. Harvard University.