Luc Berthouze

Motor Skill Acquisition Under Environmental Perturbations: On the Necessity of Alternate Freezing and Freeing of Degrees of Freedom:

In a recent study on the pendulation of a small-sized humanoid robot (Lungarella & Berthouze, 2002a, b), we provided experimental evidence that starting with fewer degrees of freedom enables a more efficient exploration of the sensorimotor space during the acquisition of a task. The study came as support for the well-established framework of Bernstein (1967), namely that of an initial freezing of the distal degrees of freedom, followed by their progressive release and the exploitation of environmental and body dynamics. In this paper, we revisit our study by introducing a nonlinear coupling between environment and system. Under otherwise unchanged experimental conditions, we show that a single phase of freezing and subsequent freeing of degrees of freedom is not sufficient to achieve optimal performance, and instead, alternate freezing and freeing of degrees of freedom is required. The interest of this result is twofold: (1) it confirms the recent observation by Newell & Vaillancourt (2001) that Bernstein’s (1967) framework may be too narrow to account for real data; (2) it suggests that perturbations that push the system outside its postural stability or increase the task complexity may be the mechanism that triggers alternate freezing and freeing of degrees of freedom.

Deferred imitation of human head movements by an active stereo vision head

Designing a mechanism that will allow a robot to imitate the actions of a human, apart from being interesting for opening the possibilities for efficient social learning through observation and imitation, is challenging since it requires the integration of information from the visual, memory and motor systems. This paper deals with the implementation of an imitation architecture on an active, stereo vision head, and describes our experiments on the deferred imitation of human head movements.

Emergence and Categorization of Coordinated Visual Behavior Through Embodied Interaction

This paper discusses the emergence of sensorimotor coordination for ESCHeR, a 4DOF redundant foveated rob ot-head, by interaction with its environment. A feedback-error-learning(FEL)-based distributed control provides the system with explorative abilities with reflexes constraining the learning space. A Kohonen network, trained at run-time, categorizes the sensorimotor patterns obtained over ESCHeRs interaction with its environment, enables the reinforcement of frequently executed actions, thus stabilizing the learning activity over time. We explain how the development of ESCHeRs visual abilities (namely gaze fixation and saccadic motion), from a context-free reflex-based control process to a context-dependent, pattern-based sensorimotor coordination can be related to the Piagetian ‘stage theory’.

On the interplay between morphological, neural, and environmental dynamics: A robotic case study

The robust and adaptive behavior exhibited by natural organisms is the result of a complex interaction between various plastic mechanisms acting at different time scales. So far, researchers have concentrated on one or another of these mechanisms, but little has been done toward integrating them into a unified framework and studying the result of their interplay in a real-world environment. In this article, we present experiments with a small humanoid robot that learns to swing. They illustrate that the exploitation of neural plasticity, entrainment to physical dynamics, and body growth (where each mechanism has a specific time scale) leads to a more efficient exploration of the sensorimotor space and eventually to a more adaptive behavior. Such a result is consistent with observations in developmental psychology.

Learning of oculo-motor control: a prelude to robotic imitation

In order to allow robot agents to adapt their behaviour, a new approach-learning by imitation-has been proposed, in which a robot learns novel behaviours through interactions with the environment and other agents. In this paper we describe how our sighted agent, ESCHeR -who is equipped with dual foveated lenses and can control head, neck and eye joints-develops fine oculo-motor control through interaction with the environment. ESCHeR successfully learns to coordinate pan, tilt and vergence such that he can track bright moving objects and saccade rapidly to new objects of interest. In developmental terms, ESCHeR can be considered to have progressed from stage 1 to stage 2 of imitation learning. A novel representation of visual scenes is then introduced, and it is discussed how ESCHeR will use this to progress to stage 3.

Modelling approaches for simple dynamic networks and applications to disease transmission models

In this paper a random link activation–deletion (RLAD) model is proposed that gives rise to a stochastically evolving network. This dynamic network is then coupled to a simple susceptible-infectious-suceptible (SIS) dynamics on the network, and the resulting spectrum of model behaviour is explored via simulation and a novel pairwise model for dynamic networks. First, the dynamic network model is systematically analysed by considering link-type independent and dependent network dynamics coupled with globally constrained link creation. This is done rigorously with some analytical results and we highlight where such analysis can be performed and how these simpler models provide a benchmark to test and validate full simulations. The pairwise model is used to study the interplay between SIS-type dynamics on the network and link-type-dependent activation–deletion. Assumptions of the pairwise model are identified and their implications interpreted in a way that complements our current understanding. Furthermore, we also discuss how the strong assumptions of the closure relations can lead to disagreement between the simulation and pairwise model. Unlike on a static network, the resulting spectrum of behaviour is more complex with the prevalence of infections exhibiting not only a single steady state, but also bistability and oscillations.

Rectification of EMG in low force contractions improves detection of motor unit coherence in the beta-frequency band

Rectification of surface EMG before spectral analysis is a well-established preprocessing method used in the detection of motor unit firing patterns. A number of recent studies have called into question the need for rectification before spectral analysis, pointing out that there is no supporting experimental evidence to justify rectification. We present an analysis of 190 records from 13 subjects consisting of simultaneous recordings of paired single motor units and surface EMG from the extensor digitorum longus muscle during middle finger extension against gravity (unloaded condition) and against gravity plus inertial loading (loaded condition). We directly examine the hypothesis that rectified surface EMG is a better predictor of the frequency components of motor unit synchronization than the unrectified (or raw) EMG in the beta-frequency band (15-32 Hz). We use multivariate analysis and estimate the partial coherence between the paired single units using both rectified and unrectified surface EMG as a predictor. We use a residual partial correlation measure to quantify the difference between raw and rectified EMG as predictor and analyze unloaded and loaded conditions separately. The residual correlation for the unloaded condition is 22% with raw EMG and 3.5% with rectified EMG and for the loaded condition it is 5.2% with raw EMG and 1.4% with rectified EMG. We interpret these results as strong supporting experimental evidence in favor of using the preprocessing step of surface EMG rectification before spectral analysis.

Epigenetic robotics: Modelling cognitive development in robotic systems

According to Zlatev and Balkenius (2001), the goal of Epigenetic robotics is to understand, and model, the role of development in the emergence of increasingly complex cognitive structures from physical and social interaction. As such, Epigenetic Robotics is an interdisciplinary effort, combining developmental psychology, neuroscience, and robotics. This still recent field is being driven by two main, somewhat parallel, motivations: (a) to understand the brain by constructing embodied systems – the so-called synthetic approach, and (b) to build better systems by learning from human studies. While this two-pronged approach has led to promising results (see (Lungarella, Metta, Pfeifer, & Sandini, 2003) for a comprehensive review), these editors believe that the field will benefit from amore rigorous coupling between both components. Pro-

Human EEG shows long-range temporal correlations of oscillation amplitude in Theta, Alpha and Beta bands across a wide age range

OBJECTIVE Long-range temporal correlations (LRTC) of EEG amplitude fluctuations in adults reveal power-law statistics and have been interpreted within the framework of self-organized criticality (SOC). In physical systems states of self-organized criticality showing power-law statistics take time to develop. In this paper we have sought evidence for the idea that brain development tends towards SOC through examining the hypothesis that during normal human development a power law behaviour of EEG oscillations is approached with increasing chronological age. METHODS We examined EEGs from central and parietal electrodes in 36 subjects aged between 0 and 660months during performance of a steady wrist extension task with their dominant hand and applied spectral and detrended fluctuation analysis in 36 subjects to assess long-range temporal correlations of oscillation amplitude in the Theta, Alpha and Beta frequency bands. RESULTS Our data indicate that at all subject ages power-law statistics dominate the records at Alpha, Beta and Theta frequencies. Small consistent effects of chronological age were detected for amplitude fluctuations at Theta and Beta frequencies. CONCLUSIONS The data suggest that the scale-free nature of EEG LRTCs is a feature from early childhood through to maturity but that there are changes in the magnitude of these effects with age. SIGNIFICANCE This study is the first to have explored long-range temporal correlations over a wide range of chronological age.

Neural learning of embodied interaction dynamics

This paper presents our approach towards realizing a robot which can bootstrap itself towards higher complexity through embodied interaction dynamics with the environment including other agents. First, the elements of interaction dynamics are extracted from conceptual analysis of embodied interaction and its emergence, especially of behavioral imitation. Then three case studies are made, presenting our neural architecture and the robotic experiments on some of the important elements discussed above: self-exploration and entrainment, emergent coordination, and categorizing self-behavior. Finally, we propose that integrating all these elements will be an important step towards realizing the bootstrapping agent envisaged above.