Pitfalls in the Modeling of Developmental Systems

Abstract

Computational models of development aim to describe the mechanisms that underlie the acquisition of new skills or the emergence of new capabilities. The strength of a model is judged by both its ability to explain the phenomena in question as well as its ability to generate new hypotheses, generalize to new situations, and provide a unifying conceptual framework. Although often constructed using traditional engineering methodologies, evaluating the performance of a computational model of development in terms of traditional perspectives is a flawed approach. This paper addresses the fundamental issues that confound quantitative analysis of computational models of developmental systems. In particular we focus on the following recommendations: 1) don't equate the success of a developmental model with its peak performance at some task; 2) don't employ purely subjective or vague measures of model fitness; and 3) don't hide or reject variation as found in the computational model. Along the way, we discuss the aspects of computational models of development that lead to the requirements for specialized methods of analysis.

Modelling Concept Prototype Competencies using a Developmental Memory Model

“…[23]. This widening of scope underlines the importance of considering ontogeny: exploring the developmental substrate of such integrated functionality provides insight not only into how such cognitive competence arises, but also indicates a means to achieve autonomous operation for robotic agents [31], with the changes in perspective on system design and evaluation that this entails [26]. In this paper, the question of how to account for concept utility embedded in wider cognitive processing within a developmental framework is explored.…”

Can a Robot Smile? Wittgenstein on Facial Expression

“…The goal of developmental roboticists is that their robots, from an initial state of mere 'motor babbling' and aided by scaffolding from humans, learn early human social behaviours in a way that is biologically plausible. The researchers' vision is of 'a machine that can learn incrementally, directly from human observers, in the same ways that humans learn from each other' (Scassellati et al, 2006, p. 41; see Shic and Scassellati, 2007). It is in this sense that Kismet is a 'child-machine'.…”