Christian Scheier

Sensory-motor coordination : The metaphor and beyond

Abstract Any agent in the real world has to be able to make distinctions between different types of objects, i.e. it must have the competence of categorization. In mobile agents, there is a large variation in proximal sensory stimulation originating from the same object. Therefore, categorization behavior is hard to achieve, and the successes in the past in solving this problem have been limited. In this paper it is proposed that the problem of categorization in the real world is significantly simplified if it is viewed as one of sensory—motor coordination, rather than one of information processing happening “on the input side”. A series of models are presented to illustrate the approach. It is concluded that we should consider replacing the metaphor of information processing for intelligent systems by the one of sensory-motor coordination. However, the principle of sensory-motor coordination is more than a metaphor. It offers concrete mechanisms for putting agents to work in the real world. These ideas are illustrated with a series of experiments.

Embedded neural networks: exploiting constraints

Using concepts and tools of embodied cognitive science, we investigate the implications of embedding neural networks in a physical structure, the body of a robot. Embedding a neural network in a body provides constraints that can be exploited for learning. We show that the constraints are given by the environment and object properties, the agents morphology, the agents motor system and specific ways of interacting with the objects. We argue that designing embedded neural networks implies (a) understanding these constraints, and (b) exploiting them, i.e., designing neural networks such that they-one way or other-incorporate the constraints. This in turn results in cheap and simple networks that are suited for the task environment, and have real-time responses. Moreover, this constraint-based approach provides new perspectives on two fundamental problems of cognitive science: focus-of-attention and object constancy. The main arguments are illustrated with a series of case studies with simulated and physical mobile robots that are controlled by hand-designed as well as evolved neural networks.

Dynamical analysis of schizophrenia courses

In order to assess the working hypothesis that schizophrenia may be viewed as a nonlinear dynamical disease, we examined the long-term psychoticity dynamics of 14 patients. The data consist of daily ratings of psychopathology observed for 200 or more consecutive days in each patient. We implemented nonlinear dynamical analysis methods with a potential of being applicable even to relatively short and noisy time series: two different forecasting approaches combined with surrogate methods that allow statistical testing in each single case. The resulting classification of dynamics gives evidence that eight patients show nonlinear evolutions of symptom courses. Four cases can be modeled linearly, two as random processes. Thus, a larger proportion of the schizophrenic psychoses we studied shows nonlinear time courses. In this way the validity of the concept of dynamical diseases could be supported on statistical grounds in this important area of psychopathology. The nonlinear view-a low-dimensional nonlinear system generating psychotic symptoms--may provide the foundation for a more parsimonious theory of schizophrenia compared to traditional multicausal models. In several of the nonlinear cases we also observed the qualitative fingerprint of deterministic chaos: a decay of deterministic features of the course of disorder with time.

Representation in Natural and Artificial Agents: An Embodied Cognitive Science Perspective

Abstract The goal of the present paper is to provide an embodied cognitive science view on representation. Using the fundamental task of category learning, we will demonstrate that this perspective enables us to shed new light on many pertinent issues and opens up new prospects for investigation. The main focus of this paper is on the prerequisites to acquire representations of objects in the real world. We suggest that the main prerequisite is embodiment which allows an agent - human, animal or robot - to manipulate its sensory input such that invariances are generated. These invariances, in turn, are the basis of representation formation. In other words, the paper does not focus on representations per se, but rather discusses the various processes involved in order to make learning and representation acquisition possible. The argument structure is as follows. First we introduce two new perspectives on representation, namely frame-of-reference, and complete agent. Then we elaborate the complete agent perspective and focus in particular on embodiment and situatedness. We argue that embodiment has two main aspects, a dynamic and an information theoretic one. Focusing on the latter, there are a number of implications: Representation can only be understood if the embedding of the neural substrate in the physical agent is known, which includes morphology (shape), positioning and nature of sensors. Because an autonomous mobile agent in the real world is exposed to a continuously changing high-dimensional stream of sensory stimulation, if it is to learn category distinctions, it first needs a focus of attention mechanism, and then it must have a way to reduce the dimensionality of this high-dimensional sensory stream. Learning is very hard because the invariances are typically not found in the sensory data directly - the classical problem of object constancy: it is a so-called type 2 problem. Rather than trying to improve the learning algorithms - which is the standard approach - the embodied cognitive science view suggests a different approach which focuses on the nature of the data: the agent is not passively exposed to a given data distribution, but, by exploiting its body and through the interaction with the environment, it can actually generate the data. More specifically, it can generate correlated data that has the property that it can be easily learned. This learnability is due to redundancies resulting from the appropriate interactions with the environment. Through such interactions, the former type 2 problem is transformed into a type 1 problem, thus reducing the complexity of the learning task by orders of magnitude. By observing the frame-of-reference problem we will discuss to what extent these invariances are reflected - represented - in the “neural substrate”, i.e. the internal mechanisms of the agent. It is concluded, that representation is not a concept that can be studied in the abstract, but should be elaborated in the context of concrete agent-environment interactions. These ideas are all illustrated with examples of natural agents and artificial agents. In particular, we will present a suite of experiments on simulated and real-world artificial agents instantiating the main arguments

Classification as sensory-motor coordination

In psychology classification is studied as a separate cognitive capacity. In the field of autonomous agents the robots are equipped with perceptual mechanisms for classifying objects in the environment, either by preprogramming or by some sorts of learning mechanisms. One of the well-known hard and fundamental problems is the one of perceptual aliasing, i.e. that the sensory stimulation caused by one and the same object varies enormously depending on distance from object, orientation, lighting conditions, etc. Efforts to solve this problem, say in classical computer vision, have only had limited success. In this paper we argue that classification cannot be viewed as a separate perceptual capacity of an agent but should be seen as a sensory-motor coordination which comes about through a self-organizing process. This implies that the whole organism is involved, not only sensors and neural circuitry. In this perspective, “action selection” becomes an integral part of classification. These ideas are illustrated with a case study of a robot that learns to distinguish between graspable and non-graspable pegs

Information Theoretic Implications of Embodiment for Neural Network Learning

The traditional view of neural networks is algorithmic. The general learning problems are typically hard and powerful networks and learning algorithms such as MLPs with BP must be used. It is a well-known fact that the power of the learning algorithm required to solve a problem depends on the statistical distribution of the input data. If the distribution is known, a “taylored” network can be used. We argue that in the real-world the- distributions are not given, but can be generated in the process of sensory-motor coordination as the embodied autonomous agent interacts with its environment. It is shown that sensory-motor coordination can lead to dramatic reduction of learning comlexity in the information theoretic sense. The ideas discussed in this paper tie in with a set of design principles for autonomous agents that we have established over the last few years.

Der interaktionelle Ansatz in der Kognitionswissenschaft

Zusammenfassung. Seit der kognitiven Wende besteht eine enge Anlehnung der Psychologie an die Computertechnologie und die Kunstliche-Intelligenz-Forschung. Der “computationalistische“ Ansatz geht von einer (meist sequenziell gedachten) Abfolge symbolischer Verarbeitungsschritte aus. Zwischen Wahrnehmung und Handlung werden “hohere“ Prozesse wie Kategorisierung, Gedachtnis und Planung angenommen. Solche Prozesse sind als eigenstandige Module ohne direkte Schnittstelle zur Umwelt gedacht. Dieser Ansatz wird in der vorliegenden Positionsarbeit wegen seiner inharenten fundamentalen Probleme kritisiert. Alternative Forschungsprogramme fordern eine Hinwendung zu Situiertheit und “embodiment“ von Kognition, die nicht unabhangig von Umwelt und Verhalten verstanden werden kann. Eine neue Methode dieser interaktionellen Sicht stellen autonome Agenten (Roboter) dar. Konsequenzen fur die Psychologie werden herausgearbeitet, zunachst eine theoretische Konzeptualisierung autonomer Agenten basierend auf der Theorie dyna...

Embodied Cognitive Science: Concepts, Methods and Implications for Psychology

Since the “cognitive shift” of psychology, a close association between psychology and the advances in computer technology and artificial intelligence research has evolved. According to the ‘computational’ symbol processing approach, cognition consists of a series of sequentially ordered processing stages. Between perception and action, input is processed by higher cognitive functions, such as categorization, memory, and planning. These cognitive functions are conceived as independent modules lacking a direct interface with the environment. This approach is criticized due to its inherent fundamental problems. Alternative research programs, such as embodied cognitive science, primarily address the issues of embodied cognition, i. e., cognition is viewed as originating from the interaction of body and environment. The methods of the corresponding “new AI” encompass robotics and the use of autonomous agents. It is investigated here which implications for psychology may arise. A theoretical conceptualization of autonomous agents based on dynamical systems theory and synergetics is outlined. Within this context, the cognitive system is conceived as a complex system comprising numerous sensorimotor loops; coherent and adaptive perception-action processes emerge from the influence of affordances. Examples cited from the field of applied psychology indicate that these perspectives lead to the formulation of new research questions and reinterpretation of empirical findings.

Feature Binding Through Temporally Correlated Neural Activity in a Robot Model of Visual Perception

An agent performing a task in an environment must be able to selectively attend to visual stimuli. This ability is of critical importance for adaptive behavior in (vision-based) biological and artificial agents. In this paper we present a connectionist model of how visual attention can serve an agent to perform its task. The model is embedded in a mobile robot. Visual stimuli are segregated by means of synchronization of spiking neurons. They then enter a selection process, the result of which determines what region of the visual field the robot will attend and consequently react to. Results from the behavior of the robot as well as the underlying neuronal dynamics are presented, and limitations as well as future extensions of the model are discussed.

Implications of embodiment for robot learning

The work is based on that of Brooks (1986), who argued that intelligence requires a body and therefore suggested that robots be used to study principles of intelligence. We show in more detail why some of the problems in intelligent behavior like category learning are simplified if the embodiment is exploited appropriately. We will also demonstrate that embodied systems can learn in an unstructured and unlabelled environment. Moreover, the seemingly intractable problems of behaving and learning in the real world become manageable. We will substantiate our argument with an information theoretic analysis. The work presented in this paper is theoretically motivated. It has been derived from a number of design principles of autonomous agents. They will be briefly outlined to provide the general context in which this research is situated.