Comparisons of developmental modeling frameworks and levels of analysis in cognition: connectionist and dynamic systems theories deserve attention,but don’t yet explain attention

In this article, an investigator who is outside of the connectionist / dynamic systems (CD) tradition, but is sympathetic to many of its theoretical stances, considers whether these are separate approaches and what the two approaches mean for the development of information processing. It is argued that these approaches have not been defined in the target articles of this special issue narrowly enough to make them clearly separate approaches. Rather, they appear to emphasize different aspects of a common approach. It is also suggested that they do not form an overarching theory of development but do form an important theoretical approach at one level of analysis. In particular, one may sometimes, but not always, have to get into nonlinear interactions between faculties of the mind to predict and explain behavior. Finally, an attempt is made to challenge the CD approach to become more global by confronting the issue of how attention and conscious awareness should be represented in these models, and how the development of them may occur. Evidence of changes in automatic and attention-related parameters of processing may provide a scaffolding upon which better CD models can be constructed.