Developing Hierarchical Schemas and Building Schema Chains Through Practice Play Behavior

Abstract

Examining the different stages of learning through play in humans during early life has been a topic of interest for various scholars. Play evolves from practice to symbolic and then later to play with rules. During practice play, infants go through a process of developing knowledge while they interact with the surrounding objects, facilitating the creation of new knowledge about objects and object related behaviors. Such knowledge is used to form schemas in which the manifestation of sensorimotor experiences is captured. Through subsequent play, certain schemas are further combined to generate chains able to achieve behaviors that require multiple steps. The chains of schemas demonstrate the formation of higher level actions in a hierarchical structure. In this work we present a schema-based play generator for artificial agents, termed Dev-PSchema. With the help of experiments in a simulated environment and with the iCub robot, we demonstrate the ability of our system to create schemas of sensorimotor experiences from playful interaction with the environment. We show the creation of schema chains consisting of a sequence of actions that allow an agent to autonomously perform complex tasks. In addition to demonstrating the ability to learn through playful behavior, we demonstrate the capability of Dev-PSchema to simulate different infants with different preferences toward novel vs. familiar objects.

Discovering Schema-Based Action Sequences Through Play in Situated Humanoid Robots

“…The aforementioned works deal with the development of high-level action sequences that allow a robotic agent to solve complex manipulation problems. Here, we describe a novel chaining mechanisms of Dev-PSchema; a open-ended learning tool based on PSchema (which was firstly introduced in [25] and [26]) to extend its problem-solving capability. It is designed to discover a set of possible solutions capable of achieving the desired target state of a given problem through artificial play.…”

“…It includes the number of times an object has appeared in the environment (similarity S), the number of times it has been used in schemas (novelty N ), and the success rate and frequency of use of those schemas in memory (habituation H). These concepts are fully described in [25] and modelled as shown in the supplementary material.…”

“…Having the excitation mechanism in its core, the intrinsic exploratory behaviour modelled above allows the robot to create, combine and evaluate schema-based sequences. Expanding the previous system [25], we designed a problemsolving mechanism capable of producing multiple alternatives, while taking advantage of the extended chaining possibilities and thus the agents learning capacity. Figure 4 depicts the interactions between the user and the autonomous agent.…”

Interaction in Assistive Robotics: A Radical Constructivist Design Framework

“…Primary and secondary circular reactions clearly encompass the above-mentioned intuitive notion of (simple, basic, stereotyped) “sensorimotor associations.” They are subsequently hierarchically organized in action schemes thanks to assimilation, accommodation and organisation. Action schemes corresponding to more complex, high-level, goal-directed actions can be decomposed into finer-grained action schemes, and, in the end, into their constituent circular reactions (this idea already appears in, e.g., von Glasersfeld, 1995 ; Kumar et al, 2018 ). For instance, “reaching for and grasping a cup of tea” can be decomposed into simpler actions schemes, e.g., “focus on the cup,” “stretch the arm,” “pre-shape the hand,” etc.…”

Cognitive System of a Virtual Robot Based on Perception, Memory, and Hypothesis Models for Calligraphy Writing Task

“…By combining cognitive psychology and robotics, [7][8] revealed a psychological structure with schema and chain for robots to learn simple tasks. These studies showed good effect on the application of higher-level cognitive strategies for robots.…”