From Babies to Robots: The Contribution of Developmental Robotics to Developmental Psychology

The latest developments in AI and machine learning, and the parallel advances in robotics, have very recently contributed to a shift in the approach to modeling human intelligence. These innovations, accompanied by the new emphasis on embodied and grounded cognition in AI and psychology, have led to the establishment of the field of Developmental Robotics. This is the interdisciplinary approach, built on the close collaboration of the disciplines of cognitive robotics and child psychology, to the autonomous design of behavioral and cognitive capabilities in artificial cognitive agents, such as robots, which takes direct inspiration from the developmental principles and mechanisms observed in children. We illustrate the benefits of this approach by presenting a detailed baby robot case study of the role of embodiment during early word learning, as well as an overview of several developmental robotics model of perceptual, social and language development. Introduction Computational models of cognition have significantly contributed to the definition, testing, and validation of psychology and neuroscience theories, including developmental psychology. Such computational approaches, ranging from symbolic rule-based systems, connectionist neural networks, and Bayesian models, have typically resulted from scientific and technological developments in artificial intelligence (AI) and its attempt to reproduce and simulate the uniqueness and complexity of human-like adult symbolic intelligence. However, since the origins of AI, there have been proposals to study the full spectrum of child development, rather than adult-like intelligence. This is for example what Alan Turing proposed in 1950: “Instead of trying to produce a programme to simulate the adult mind, why not rather try to produce one which simulates the child’s? If this were then subjected to an appropriate course of education one would obtain the adult brain.” Turing (1, page 440). The latest developments in AI and machine learning, and the parallel advances in robotics, have contributed to a shift in the approach to modeling human intelligence. These innovations have been accompanied by an increased emphasis on embodied and grounded cognition in AI (2) and psychology (3,4). This has permitted the first attempts to realize Turing’s vision, i.e. the idea that an embodied agent (e.g. robot), using a set of intrinsic motivation principles regulating the real-time interaction between its body, brain and environment, can autonomously acquire and develop an increasingly This is the author's accepted manuscript. The final published version of this work is published by John Wiley in Child Development Perspectives available at DOI 10.1111/cdep.12282. This work is made available online in accordance with the publisher's policies. Please refer to any applicable terms of use of the publisher. complex set of sensorimotor and mental capabilities, grounded in the interaction with its