| Nonlinear Dynamics, Psychology, and Life Sciences, Vol. 3, Iss. 1, Jan, 1999, pp. 93-111 @1999 Society for Chaos Theory in Psychology & Life Sciences Modeling Historical Development: Fitting a Competing Practices System to Coded Archival Data Abstract: This project used curve fitting to refine an ecological model of historical development. Dirlam (1972, 1980, and 1996) constructed multidimensional classifiers for coding sociocultural practices by using theories of children's drawing, students' writing, and developmental researchers' methods. The last involved an eleven-dimensional classifier based mostly on Danziger's (1990) insights. An NDS analysis began with Van Geert's (1991) variant of the Lotka-Volterra two-species model, which was generalized by Dirlam (1997) to many competing species, each embodying an evolutionary strategy. Excellent fits to codings of research strategies in 599 articles from Child Development and Developmental Psychology, 1969-1992, revealed chaotic growth unless suppressed by new strategies. In this paper, coding was extended to 313 new articles published from 1930-1968. A refined model using Levins' (1969) "logistic weed" produced more meaningful parameter values and suggested dynamic differences between evolutionary strategies and sociocultural practices. Statistically adequate solutions with both low growth and high growth were found. To differentiate solutions, we proposed independent experimental testing and examining the scientific meaning of parameter values. The analysis identified two novel dynamic entities: default and polarized practices. Removing the person-practice link and coding many dimensions at once extends dynamic modeling to a greatly enriched variety of cultural and historical processes. Keywords: developmental models, historical models, cultural practices, Lotka-Volterra models, developmental methodology |