Personalising game difficulty to keep children motivated to play with a social robot: A Bayesian approach

For effective child education, playing games with a social robot should be motivating for a longer period of time. One aspect that can affect the motivation of a child is the difficulty of a game. The game should be perceived as challenging, while at the same time, the child should be confident to meet the challenge. We designed a user modelling module that adapts the difficulty of a game to the child’s skill level, in order to provide children with the optimal challenge. This module applies a Bayesian rating method that estimates the child’s skill and game item’s difficulty levels to personalise the game progress. In an experiment with 22 children (aged between 10 and 12 years old), we tested whether the personalisation leads to a higher motivation to play with the robot. Although the personalised system did not challenge the participants optimally, this study shows that the Bayesian rating system is in principle able to measure the skill and performance of children in playing a game with a robot (even without accurate estimates of the difficulty of items). We outline multiple ways in which the rating method and module can be used to further personalise and enhance the child-robot interaction, other than adapting the difficulty of games (e.g. by adapting the dialogue and feedback).     

THE POPULAR APPEAL OF APOCALYPTIC AI

The belief that computers will soon become transcendently intelligent and that human beings will “upload” their minds into machines has become ubiquitous in public discussions of robotics and artificial intelligence in Western cultures. Such beliefs are the result of pervasive Judaeo‐Christian apocalyptic beliefs, and they have rapidly spread through modern pop and technological culture, including such varied and influential sources as Rolling Stone, the IEEE Spectrum, and official United States government reports. They have gained sufficient credibility to enable the construction of Singularity University in California. While different approaches are possible (and, indeed, are common in Japan and possibly elsewhere), this particular vision of artificial intelligence and robotics has gained ground in the West through the influence of figures such as Hans Moravec and Ray Kurzweil. Because pop‐science books help frame public discussion of new sciences and technologies for individuals, corporations, and governments alike, the integration of religious and technoscientific claims made by their authors should be clear and open for public and scientific debate. As we move forward into an increasingly robotic future, we should do so aware of the ways in which a group's religious environment can help set the tone for public acceptance and use of robotic technologies.     

Computational modeling/cognitive robotics complements functional modeling/experimental psychology

This position paper explores the possible contributions to the science of psychology from insights obtained by building and experimenting with cognitive robots. First, the functional modeling characteristic of experimental psychology is discussed. Second, the computational modeling required for cognitive robotics is described, and possible experiments with them are illustrated. Next, we argue that cognitive developmental robots, robots that “live” through a development phase where they learn about their environments in several different modes, can provide additional benefits to the science of psychology. Finally, the reciprocal interactions between computational modeling/cognitive robotics and functional modeling/experimental psychology are explored. We conclude that each can contribute significantly to the other.     

Representation and learning in motor action – Bridges between experimental research and cognitive robotics

To gain a better understanding of the functionality of representation and categorization in action and interaction, it is fundamental that researchers understand how movements are represented in long-term memory. It is our position that human motor control requires that our actions be planned and represented in terms of intended perceptual effects and future task demands, and that the individual has a well-structured mental representation of the task so that the movement can be carried out successfully. Basic Action Concepts (BACs) are identified as major building blocks of cognitive representation in long-term memory, which are cognitive tools used to master the functional demands of movement tasks. In this paper, we consider relevant issues in research methodology and present an experimental method that can be used to assess action-relevant representational structures. This method permits us to observe the strong relationship between cognitive representation and performance in manual action. For example, the specific differences in the mental representations of participants are strongly related to skill level, as well as biomechanical and task constraints. We then discuss results from our learning experiments, where we have examined the development and changes in cognitive representation over time. From these experiments we have found that cognitive reference structures include task-specific spatial information, which provides the basis for action control in skilled voluntary movement. We have implemented these results on various robotic platforms. We argue that the insights gained from various experimental approaches in the field of cognitive psychology and motor control enable researchers to explore the possibilities and limitations of artificial control architectures in robot systems. Finally, we argue that this is not a unidirectional process. Researchers from the field of cognitive psychology and motor control can profit from the advances in technological systems, which enhance the understanding of human motor control in skilled voluntary action.     

The CORTEX cognitive robotics architecture: Use cases

CORTEX is a cognitive robotics architecture inspired by three key ideas: modularity, internal modelling and graph representations. CORTEX is also a computational framework designed to support early forms of intelligence in real world, human interacting robots, by selecting an a priori functional decomposition of the capabilities of the robot. This set of abilities was then translated to computational modules or agents, each one built as a network of software interconnected components. The nature of these agents can range from pure reactive modules connected to sensors and/or actuators, to pure deliberative ones, but they can only communicate with each other through a graph structure called Deep State Representation (DSR). DSR is a short-term dynamic representation of the space surrounding the robot, the objects and the humans in it, and the robot itself. All these entities are perceived and transformed into different levels of abstraction, ranging from geometric data to high-level symbolic relations such as “the person is talking and gazing at me”. The combination of symbolic and geometric information endows the architecture with the potential to simulate and anticipate the outcome of the actions executed by the robot. In this paper we present recent advances in the CORTEX architecture and several real-world human-robot interaction scenarios in which they have been tested. We describe our interpretation of the ideas inspiring the architecture and the reasons why this specific computational framework is a promising architecture for the social robots of tomorrow.     

Robots and the Sacred in Science and Science Fiction: Theological Implications of Artificial Intelligence

In science-fiction literature and film, human beings simultaneously feel fear and allure in the presence of intelligent machines, an experience that approximates the numinous experience as described in 1917 by Rudolph Otto. Otto believed that two chief elements characterize the numinous experience: the mysterium tremendum and the fascinans. Briefly, the mysterium tremendum is the fear of God's wholly other nature and the fascinans is the allure of God's saving grace. Science-fiction representations of robots and artificially intelligent computers follow this logic of threatening otherness and soteriological promise. Science fiction offers empirical support for Anne Foerst's claim that human beings experience fear and fascination in the presence of advanced robots from the Massachusetts Institute of Technology AI Lab. The human reaction to intelligent machines shows that human beings in many respects have elevated those machines to divine status. This machine apotheosis, an interesting cultural event for the history of religions, may—despite Foerst's rosy interpretation—threaten traditional Christian theologies.     

Planet Braitenberg: Experiments in virtual psychology

Braitenberg vehicles are simple robotic platforms, equipped with rudimentary sensor and motor components. Such vehicles have typically featured as part of thought experiments that are intended to show how complex behaviours are apt to emerge from the interaction of inner control mechanisms with aspects of bodily structure and features of the wider (extra-agential) environment. The present paper describes a framework for creating Braitenberg-like vehicles, which is built on top of a widely used and freely available game engine, namely, the Unity game engine. The framework can be used to study the behaviour of virtual vehicles within a multiplicity of virtual environments. All aspects of the vehicle’s design, as well as the wider virtual environment in which the vehicle is situated, can be modified during the design phase, as well as at runtime. The result is a general-purpose simulation capability that is intended to provide the foundation for studies in so-called computational situated cognition—a field of study whose primary objective is to support the computational modelling of cognitive processes associated with the physically-embodied, environmentally-embedded, and materially-extended mind.     

Updating the Baldwin effect : The biological levels behind Piaget's new theory

In 1964, Conrad Waddington (1905–1975) presented a paper in Geneva that led to an internal reassessment of the biological underpinnings of Jean Piaget's (1896–1980) theory. This in turn resulted in an overhaul of the theoretical framework upon which his stage theory of child development had been based, including his appeals to James Mark Baldwin's (1861–1934) “circular reaction.” In addition to leading to the emergence of what has elsewhere been called “Piaget's new theory,” this renovation also resulted in the update of the famous “Baldwin Effect.” Because aspects of the subsequent framework are of contemporary significance, this essay will review some of the work leading up to those updates. In reaching behind the translations to trace the sources of the arguments to which Piaget appealed, the resulting examination fills some of the gaps found in the secondary literature without quibbling over the “correct” English interpretation of translated French terms. We also go beyond how Piaget's writings have been understood in English and extract some useful additional ideas from his sources, including how to conceive of the social context in which development takes place. We see as a result how Waddington and his colleagues, including Paul Weiss (1898–1989), provided a constructive “existence proof” for the formal hierarchy of levels that Piaget had come to by other means.