Connectionism and dynamic systems: are they really different?

Autor: Elizabeth Bates, Esther Thelen
Rok vydání: 2003
Předmět:
Zdroj: Developmental Science. 6:378-391
ISSN: 1467-7687
1363-755X
DOI: 10.1111/1467-7687.00294
Popis: We propose that connectionism and dynamic systems theory are strong contenders for a general theory of development that holds true whatever the content domain. We illustrate, through our own career narratives, the origins of these theories in motor and language development. We situate connectionism and dynamic systems among other classic and contemporary theories and conclude that, although there are meaningful differences, these differences pale in relation to the shared assumptions about the fundamental processes and mechanisms of change. Once upon a time, thanks to Jean Piaget, the field of cognitive development had a coherent, interesting, testable, and widely accepted theory. Now, alas, we are back in the preparadigmatic boat with our colleagues in the rest of psychology, with theory fragments, almost-theories, and pseudotheories bobbing about around us. . . . I will try to differentiate several distinct possibilities for future theories. . . . Most probably each of these theories will turn out to be true of different aspects of development. . . . For any particular developmental phenomenon, one theory or another will be true, and we want to know which one it is. (Gopnik, 1996, p. 221) According to Alison Gopnik’s vision of the postPiagetian era, the search for a grand developmental theory is futile. Piaget was wrong: broad, general explanations seem ‘increasingly implausible’ (p. 221) and efforts are better spent working out the details, domain-by-domain. The traditional big issues of developmental theory ‐ nature vs. nurture, continuity and discontinuity, modularity vs. distributed processes, and so on ‐ should be cast aside in favor of specific theories about content. While we agree about the importance of the details, we also believe that Gopnik, and others who espouse the domain-specific view, have scuttled the theoretical boat somewhat prematurely. In this special issue of Developmental Science we present a discussion of two contemporary contenders for the ‘big picture’: connectionism and dynamic systems theory. The articles are the result of a symposium held at the 2001 meeting of the Society for Research in Child Development. The large audience who attended the symposium presumably did not share Gopnik’s pessimism over the future of developmental theory (although it is possible they came to witness the duel-to-the death!). Rather, we sensed a growing interest and excitement, especially as these new theories are implemented in mathematical models, and thus, make contact with similar theoretical enterprises in cognitive science, biology and neuroscience. Here we focus specifically on the question of whether we are dealing with one grand theory or two: how alike or how different are connectionism and dynamic systems? To begin, we state forthrightly that we do believe that there are general principles of development: mechanisms and processes that hold true whatever the content domain. Moreover, we assert that researchers make theoretical assumptions about these processes and mechanisms whenever they design and conduct a study. Often these assumptions are implicit, but they influence every decision about the choice of variables, design, research population, sampling intervals and statistical technique. Good theory makes these assumptions explicit, and subjects them to empirical test. Furthermore, the role of formal models is to make these underlying assumptions extremely precise. The specific form of the model is thus less important than the general principles of development on which it is based. We devote our introduction, therefore, to the developmental principles ‐ the qualitative theory and assumptions ‐ that form the bases for the two related approaches of connectionism and dynamic systems. We compare them to the assumptions of other general theories and discuss the strengths and weaknesses of each. Note that there is a not-so-subtle irony here. Neither author, Bates nor Thelen, has ever programmed a computer, written an original equation, or run a simulation. How did it come
Databáze: OpenAIRE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje