Predictive learning: its key role in early cognitive development

Abstract

What is a fundamental ability for cognitive development? Although many researchers have been addressing this question, no shared understanding has been acquired yet. We propose that predictive learning of sensorimotor signals plays a key role in early cognitive development. The human brain is known to represent sensorimotor signals in a predictive manner, i.e. it attempts to minimize prediction error between incoming sensory signals and top–down prediction. We extend this view and suggest that two mechanisms for minimizing prediction error lead to the development of cognitive abilities during early infancy. The first mechanism is to update an immature predictor. The predictor must be trained through sensorimotor experiences because it does not inherently have prediction ability. The second mechanism is to execute an action anticipated by the predictor. Interacting with other individuals often increases prediction error, which can be minimized by executing one's own action corresponding to others’ action. Our experiments using robotic systems replicated developmental dynamics observed in infants. The capabilities of self–other cognition and goal-directed action were acquired based on the first mechanism, whereas imitation and prosocial behaviours emerged based on the second mechanism. Our theory further provides a potential mechanism for autism spectrum condition. Atypical tolerance for prediction error is hypothesized to be a cause of perceptual and social difficulties. This article is part of the theme issue ‘From social brains to social robots: applying neurocognitive insights to human–robot interaction’.

Discovering the Neuroanatomical Correlates of Music with Machine Learning In Eduardo Reck Miranda (Eds.). Handbook of Artificial Intelligence for Music. Part 1. Springer

“…Within the predictive coding framework, a recent study on human-robot interaction (Nagai, 2019) likewise indicated that interacting with other individuals often increases prediction error, which can be minimized by executing one's own action corresponding to others' actions.…”

A Modular Dynamic Sensorimotor Model for Affordances Learning, Sequences Planning, and Tool-Use

“…It is argued that predictive learning of sensorimotor information plays a key role in cognitive development, and may be able to produce a continuous development between different levels of cognitive abilities [15]. Rolf and Asada also point out the interesting role of goals in developmental robotics and autonomous mental development [41].…”

“…The internal dynamics of the agent and the dynamics of the environment, coupled during the acquisition of sensorimotor skills, may allow both low-and high-level skills to share a common metric space [14]. This advocates that a continuous development between different levels of cognitive abilities can lead to complex behaviours such as the use of tools [15], [16]. In a developmental framework, low-level behaviours can clearly form atomic components of high-level learning [11], [17].…”

Sensorimotor contingencies as a key drive of development: from babies to robots

“…Ultimately, it is assumed, sensitivity to sensorimotor contingencies will contribute to the establishment of socio-cognitive and motor skills (Gergely and Watson, 1996;Tarabulsy et al, 1996;Nagai and Asada, 2015;Verschoor and Hommel, 2017) like language (e.g., Goldstein and Schwade, 2008;Schwartz et al, 2012), reaching and grasping (e.g., Corbetta, and Snapp-Childs, 2009) and walking (e.g., Adolph et al, 2000).…”