The dynamics of permanence
Jun Luo, Michael Gasser, Linda B. Smith
- Year
- 2004
- Citations
- 2
Abstract
This dissertation is a simulated-robotics study of the development of the A-not-B behavior and object permanence capacity in infancy. Drawing from recent empirical, theoretical, and modeling work as well as earlier work of Poincare and Piaget, I argue, through a series of theoretical discussions and computer simulations, (i) that the body rotation of infants, which no extant model of the A-not- B behavior explicitly takes into account, shall be taken into account because it causes the occluded target to change egocentric location without direct visual feedback, and can be taken into account in a simulated robotic model that extends Thelen and colleagues' dynamic systems model via introducing kinesthesis-based updating of spatial working memory that compensates for body rotation; (ii) that the notion of attractor stability underwriting analysis of behavior from the dynamic systems perspective, needs to be extended to a notion of “relational stability” so as to account for infants' compensatory memory activity in relation to relative target shifting, an extension that suggests a way of explicating the representational character of cognition; (iii) that underwriting representation of location is a developmentally more primary achievement of a capacity for localization, a point that is corroborated by findings from researchers such as Acredolo and Newcombe and in agreement with challenges raised by Newcombe, Smith and Scheier to models, such as that from Munakata, in which locations are modeled by units labeled with location names such as ‘A’ or ‘B’; (iv) that the spatial working memory dynamics and the associated kinesthesis-based compensation mechanism can be learned by neuroscience-inspired unsupervised learning, which provides a way to concretize the well-documented correlation between the development of object permanence and the development of locomotion; and (v) that both successful tracking of an occluded target and the development of such tracking mechanism in the presence of incessant bodily movements require that the metric detail of relative target displacements be available.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002