Mouvement, action et conscience : vers une physiologie de l’intention.
Colloque en l’honneur de Marc Jeannerod
27 et 28 Septembre, 2002
Institut des Sciences Cognitives
le colloque est soutenu par :
english abstract Scientific Program papers' abstracts posters Marc Jeannerod a été au cours des 30 dernières années une figure déterminante pour l’essor des neurosciences cognitives en France. Professeur de médecine et chercheur, élève de Michel Jouvet, il a dirigé durant plusieurs années un laboratoire de recherche INSERM dont l’ensemble des travaux sur le contrôle du mouvement a fait autorité dans la communauté scientifique internationale. Outre ses contributions scientifiques originales dans les domaines de la neurophysiologie et de la psychomotricité, Marc Jeannerod a publié plusieurs articles de synthèse et ouvrages destinés aux spécialistes ainsi qu’au grand public. S’intéressant tout au long de sa carrière à l’organisation de l’action au sens le plus large, depuis la genèse de l’intention consciente à la production des synergies musculaires, il a posé les questions fondamentales des relations entre cerveau et comportement, proposé de nombreuses et fécondes pistes de recherche et stimulé les échanges interdisciplinaires entre neurobiologistes, cliniciens, psychologues, philosophes, etc. C’est également cette vision ouverte qui a permis à Marc Jeannerod de mener à bien la mission, qui lui avait été confiée par le CNRS, de créer à Lyon un Institut des Sciences Cognitives qu’il dirige depuis 1998. A l’occasion de son départ à la retraite à l’automne 2002, l’Institut des Sciences Cognitives organise un colloque qui réunira des spécialistes mondiaux de « l’action » pour débattre des grands sujets de l’heure. Ce colloque sera également l’occasion de retracer le parcours scientifique de Marc Jeannerod et de prendre la mesure de son impact sur les idées et les travaux de plusieurs générations de chercheurs en France comme à l’étranger.
Movement, action and consciousness : toward a physiology of intentionality.
A symposium in honor of Marc Jeannerod27 & 28 September, 2002
Institute for Cognitive SciencesFor the past 30 years, Marc Jeannerod has been a major actor and promoter of cognitive neuroscience in France. Professor of medecine and researcher, trained under Michel Jouvet, he was for several years director of an INSERM laboratory which generated a large body of very influential work on the control of movement. In addition to his numerous scientific contributions in the field of motor physiology, Marc Jeannerod has made a number of major theoretical contributions and published several books for general audiences. Throughout his career he has been exploring the ways in which the brain creates action, from the genesis of conscious intentions to the production of motor synergies. He tackled the fundamental questions of the relation between brain and behavior, opened up new avenues for research, and understood early the importance of encouraging multidisciplinary exchanges between neurobiologists, clinicians, psychologists, philosophers, etc. This wide-ranging perspective allowed Marc Jeannerod to carry out successfully the CNRS-backed project of creating an Institute for Cognitive Science in Lyon, which he directs since 1998. As Marc Jeannerod will be retiring at the end of 2002, the Institute is organizing a symposium in his honor where the best international specialists of “action” will gather to discuss the hot topics of the moment. This symposium will also be an opportunity to pay a tribute to Marc Jeannerod, and to measure the full impact of his ideas and of his work on several generations of scientists.
Thursday, September 26th
18 :00 – 21 :00 Welcoming party at the Vinatier Hospital
Friday, September 27th
9 :00 – 9 :15
Welcoming remarks : Jean-René Duhamel9 :15 – 9 :30
Introduction : Michel Jouvet9 :30 – 12 :00 Session I
Chaired by : Yves Burnod, Denis Pélisson9 :30 – 10 :00 Hideo Sakata (Nihon University School of Medicine, Tokyo)
Toward the understanding of neural process for 3D shape perception.10 :00 – 10 :30 Francesco Lacquaniti (Università degli studi di Roma)
Cognitive and action-oriented representations of object motionCoffee Break
11 :00 – 11 :30 Paolo Viviani (University of Geneva and Università Vita-Salute San Raffaele, Milano)
Perceptuo-motor interactions11 :30 – 12 :00 Daniel Wolpert (Sensorimotor control lab, Institute of Neurology, University College London,UK)
Uncertainty in sensorimotor control12 :00 – 12 :30 Special event - chaired by Yves Paulignan
The hidden side of action, by Claude Prablanc (INSERM U534, Lyon)12: 30 – 14 :00
Lunch14 :00-16 :30 Session II
Chaired by : Yves Rossetti, Driss Boussaoud14 :00-14 :30 Hans-Joachim Freund (FZ Jülich,Düsseldorf)
Sensorimotor transformation disorders in parietal lobe disease14 :30-15 :00 A.David Milner (University of Durham,UK)
Space and action in neglect and extinctionCoffee Break
15 :30-16 :00 Alessandro Farnè (Department of Psychology, University of Bologna)
Multisensory coding of action space16 :00-16 :30 J.C. Marshall (Oxford University)
Visuo-spatial neglect: Art and insight
16 :30-17 :30 Poster presentation
17 :30 :19 :00
Public interview of : Marc Jeannerod, Giacomo Rizzolatti, Pierre Jacob and Michael Arbib, by Sylvie Coyaud19 :30 Departure for gala dinner
20 :00 Gala dinner in honor of Marc Jeannerod
Saturday, September 28th
9 :30 - 12 :00 Session III
Chaired by : Angela Sirigu, Jacques Paillard9 :30 - 10 :00 Melvin A. Goodale (University of Western Ontario, London, Ontario)
Duplex vision in the cerebral cortex: How (and why) the visual control of action differs from visual perception10 :00 - 10 :30 Patrick Haggard (Institute of Cognitive Neuroscience, University College London, UK)
Action and awareness: from visuomotor corrections to agencyCoffee Break
11 :00 - 11 :30 Jordan Grafman (Cognitive Neuroscience Section, NINDS, Bethesda)
Oh baby, let me take you where the (cognitive) action is!11 :30 - 12 :00 Marc Jeannerod (Institute for Cognitive Science, Lyon)
From hand to mind. A way to the representation of action.
12 :00-12 :30 Special Event - chaired by Jean-René Duhamel
François Michel (Lyon)
"Portrait of the scientist as a young man"Larry Parsons (San Antonio)
The Scientific Marc: Brilliance Disguised as Commonsense12: 30 – 14 :00
Lunch14 :00-16 :30 Session IV
Chaired by : Elena Daprati, Nicolas Georgieff14 :00-14 :30 Jean Decety (University of Washington Center for Mind, Brain & Learning, Seattle)
Shared neural representations between self and other: To what extent are they really common?14 :30-15 :00 Giacomo Rizzolatti (Istituto di Fisiologia Umana, Università di Parma, Parma)
Action understanding and imitation: facts and speculationsCoffee Break
15 :30-16 :00 Michael Arbib (Computer Science, Neuroscience and USC Brain Project, University of Southern California, Los Angeles)
Syntax and Schizophrenia: Extending the Mirror System Hypothesis16 :00-16 :30 Shaun Gallagher (Philosophy and Cognitive Sciences, Canisius College, Buffalo, New York)
Simulation, Imagination, and Empathy16 :30 – 17 :00 Closing remarks by Pierre Jacob (Institut Jean Nicod, Paris)
Syntax and Schizophrenia: Extending the Mirror System HypothesisMichael A. Arbib
Computer Science, Neuroscience and USC Brain Project
University of Southern California
Los Angeles, CA 90089-2520
arbib@pollux.usc.eduIn the "mirror system" for grasping in monkey – part of the F5 premotor brain region -- neurons active when the monkey executes a hand action are also active when the monkey observes someone carrying out a similar action. F5 also contains “canonical neurons” active when the monkey executes a specific hand action but not when the monkey observes a similar action. F5 is the homologue of Broca’s area, a crucial speech area in humans. The Mirror System Hypothesis (Rizzolatti and Arbib, 1998; Arbib, 2002) states that the matching of neural code for execution and observation of hand movements is the precursor of the crucial language property of parity, namely that an utterance usually carries similar meaning for speaker and hearer (using these terms neutrally for spoken and signed languages). However, this grounding hypothesis leaves many questions open, and must go beyond the basic Mirror System Hypothesis in terms of brain regions considered and mechanisms explained. The present talk will go beyond the mirror by reporting on two recent collaborations.
Work with Jean-Roger Vergnaud: We develop the hypothesis that sentences are structured as S ? NP {agent} VP {verb + patient} because the canonical system of F5 binding action to object underlies the VP binding of verb to object; while the mirror system of F5 underlies the binding of NP/subject to the VP.
Work with Nathan Mundhenk and Laurent Itti: The F5 mirror system signals does not (as far as we know) encode the agent of the action it encodes. Recent work by Jeannerod and his colleagues on the delusions of schizophrenics that other people or aliens are moving their limbs for them (Franck et al., 2001) underlies the extension of our MNS1 model of the mirror system (Oztop and Arbib, 2002) to include a preliminary account of agency.
The talk will conclude by suggesting how results of the second study may inform our work on the first study to link the linguistic and cognitive representations of the agent into the neural representations of the sentence and scene, respectively, of which it is a part.Arbib, M.A., 2002, The Mirror System, Imitation, and the Evolution of Language, in Imitation in Animals and Artifacts, (Chrystopher Nehaniv and Kerstin Dautenhahn, Editors), The MIT Press.
Franck N, Farrer C, Georgieff N, Marie-Cardine M, Dalery J, d’Amato T, Jeannerod M, 2001, Defective Recognition of One’s Own Actions in Patients with Schizophrenia, Am J Psychiatry, 158:454-9.
Oztop, E., and Arbib, M.A., 2002, Schema Design and Implementation of the Grasp-Related Mirror Neuron System, Biological Cybernetics.
Rizzolatti, G, and Arbib, M.A., 1998, Language Within Our Grasp, Trends in Neurosci, 21(5):188-194.
Shared neural representations between self and other: To what extent are they really common?
Jean Decety
University of Washington Center for Mind, Brain & Learning
Seattle - USAThere is now an impressive body of evidence to support the assumption that the same neural structures that belong to the motor system are recruited by the generation of one’s own actions, the observation of actions performed by other individuals, and the mental simulation of one’s own actions. This provides a parsimonious realization of what in evolutionary terms must have been an intertwining of perception and action functions. This also gives us a neurophysiological basis for the operation of social cognition. Indeed, the meaning of a given object, action, or social situation may be common to several individuals and thus should activate the same neural network in their respective brains. Equally important is how the self distinguishes itself from other selves and hence what the non-overlapping areas may be. Our recent work, using functional neuroimaging techniques and reciprocal imitation paradigms, demonstrates that the inferior parietal cortex (in an area adjacent to the temporoparietal junction) plays a pivotal role in distinguishing the perspective of the self from those of others.
Multisensory coding of action space
CNC, Centre of Cognitive Neuroscience
Department of Psychology, University of Bologna, Italy
Neurophysiological and neuropsychological evidence converge in showing that peripersonal space is represented through the integrated processing of multiple sensory inputs. Multisensory representations of peripersonal space have been proposed to deserve the function of localising stimuli that are approaching the subject’s body, and preparing appropriate defensive motor reactions. These multisensory mechanisms might also be involved in coding peripersonal space for purposefully interacting with objects. Here, we will report evidence of dynamic interactions between space representation and action execution in humans, as revealed through the phenomenon of visual-tactile extinction. Such interactions are unveiled by the behavioural effects that the use of a tool, as a physical extension of the reachable space, can have on the representation of the visual space surrounding the hand. In particular, it will be shown that the peri-hand area, whereby vision and touch are integrated, can be expanded and contracted depending upon the use of a tool. These dynamic changes might render the possibility of reaching and manipulating far objects as if they were near the hand.
Sensorimotor transformation disorders in parietal lobe disease
Hans-Joachim Freund, FZ Jülich, Germany
The disturbances of motor behaviour in parietal lobe disease show a broad spectrum ranging from the most elementary to the most complex functions . The maintenance of a steady force level may be equally disturbed as skilful movements. The performance of meaningless movement sequences in patients with left parietal damage is a more common finding than the various apraxias affecting expressive and purposive motor acts. Quantitative recordings reveal alterations of the spatial and temporal characteristics already at the level of simple movements. Their cumulative effects compromise the production of more complex motor behaviours.
A characteristic feature of the motor abnormalities is the deficient sensory control of movement that is further amplified by the effects of the misconceived movements on sensory input. The breakdown of the finely tuned exploratory patterns of the fingers in active touch or of the ocular scan paths during visual search reveal that the essence of these dysfunction lies in the impairment of the purposive nature of those movements necessary to shape the required sensory input. These disturbances can be highly selective and affect only unimodal sensorimotor functions leaving the remaining motor repertoire unaffected.
This is further illustrated by unimodal transformation disorders in cases with damage of the posterior parietal lobe and a selective disturbance of motor behaviour involving the use of mirrors. Performing goal-directed actions while looking in a mirror requires additional transformations of images in more than one frame of reference in order to combine what we perceive as a reflection in mirror space with what we know is present in real space. Therefore there is a distinction between passively using a mirror to perceive objects, and using mirrors to actively guide our movements using an object’s reflected image. Perturbations of the distributed network required for such complex transformations disclose their lower redundancy but leave direct visuomotor behaviour unaffected The high incidence of mirror agnosia and mirror ataxia after parietal lobe damage may therefore reflect their greater liability to disturbance. rather than the selective clustering of lesions at a ‘mirror transformation locus’.
Simulation, Imagination, and Empathy
Shaun Gallagher
Philosophy and Cognitive Sciences
Canisius College
Buffalo, New YorkSimulation approaches to the problems of social cognition and theory of mind have received important support from recent work in the neurosciences. The discovery of mirror neurons (Rizzolatti) and research that shows activation in the same pre-motor areas responsible for one's intentional action and the observation and imaginative simulation of another person's action (Jeannerod, Decety) have been used as evidence for the simulation approach (e.g., Gallese and Goldman), or more generally, for motor theories of social cognition (Wolpert). The connections between brain imaging and experimental evidence, on the one hand, and the theoretical formulations of the simulation approach, however, are not as unambiguous as they should be.
Although there is constant use of the term 'simulation' in both the experimental and the theoretical literatures, the use is not consistent in regard to its conceptual meaning. In some cases 'simulation' appears to refer to processes that are taking place in the motor system itself, on a subpersonal and non-conscious level. In other cases, simulation is understood as equivalent to conscious use of the imagination. I want to ask several questions about simulation (in its sub-personal sense) and its relation to imagination and empathy.
(1) Is there a clear distinction between simulation and intersubjective perception (that is, between simulation of the other and perception of the other)? Does simulation involve neuronal processes over and above those responsible for intersubjective perception, or vice versa?
(2) Does simulation generate its own phenomenological aspects – is there a phenomenological correlate to the motor reverberation?
(3) If we grant the distinction between simulation and conscious imagination, can we say that intersubjective imagining depends in all cases on motor simulation?
(4) Closely related to the previous question, are simulative processes active in both first-person egocentric imagining, and third-person allocentric imagining?
(5) Finally, is simulation and/or intersubjective imagining sufficient to account for empathy? Does empathy involve something more than simulation/imagination?I suggest that we will be able to clarify the role of simulation in social cognition and theory of mind only when the distinctions between simulation, imagination, and empathy have been clarified, and the various questions above answered.
Duplex vision in the cerebral cortex: How (and why) the visual control of action differs from visual perception
M.A. Goodale (London, Ontario)
Visual systems first evolved not to enable animals to see, but to provide distal sensory control of their movements. Vision as 'sight' is a relative newcomer on the evolutionary landscape, but its emergence has enabled animals to carry out complex cognitive operations on perceptual representations of the world. Recent behavioral and fMRI studies from our laboratory will be presented that provide evidence of separate processing for visuomotor control and perceptual representations. Converging lines of research from a number of laboratories, including Marc Jeannerod's, suggest that these two functions of vision are mediated by separate cortical visual pathways arising from area V1 in the human cerebral cortex: a dorsal ‘action’ pathway projecting to the posterior parietal cortex and a ventral ‘perception’ pathway projecting to inferior temporal cortex. As our behavioral and fMRI studies indicate, these cortical pathways (which appear to be homologous with corresponding pathways in the monkey) make use of different metrics and different frames of reference in carrying out their computations
"Oh baby, let me take you where the (cognitive) action is!"
Jordan Grafman, Ph.D.
Abstract
I hypothesize that structured event complexes (SECs) are stored in the human prefrontal cortex. I propose that the key representational components of the SEC include: (1) thematic macroknowledge that must be abstracted across events, (2) thematic macroknowledge that directly represents the cognitive features of specific events, (3) specific grammatical properties of multi-event knowledge structures, (4) sequential and temporal codes specific to the SEC (e.g., long time durations containing multiple complex events - e.g., eating at a restaurant), and (5) category-specific domains of storage (e.g., social versus non-social). When activated in concert (controlled primarily by hippocampal binding mechanisms) with knowledge stored in other cortical regions (e.g., objects, faces, words, motor actions), the SEC participates in the formation of various kinds of memory episodes. We have studied the SEC in behavioral experiments with normal subjects, patients with damage to the prefrontal cortex, and during functional neuroimaging. The results of these studies indicate that SECs are sensitive to frequency and can be subdivided by knowledge domain. They also suggest that as humans develop, they go from representing rigidly ordered SECs to being able to rapidly adapt new events to previously existing SECs in order to modify them or to form new SECs. Functional neuroimaging studies suggest that adaptive SEC processing requires anterior prefrontal cortex whereas rigidly formed SECs are processed in medial prefrontal cortex. Studies with patients indicate that SECs can be selectively impaired suggesting a representational
deficit. Some evidence suggests that across-event knowledge is stored in the right prefrontal cortex whereas within-event knowledge is stored in the left prefrontal cortex. So, what does this evidence have to do with action? First of all, I argue that the SEC is a form of "semantic memory or knowledge representation that controls actions" that has similar storage constraints to words and objects. Second of all, I argue that when the
hippocampus provides a binding process to link different cortical representations into an episode, the SEC provides the contextual knowledge for the episode placing objects, words, and faces into a certain time and
place and action. By virtue of the claim that SEC representations must encode and store forms of information that can only be processed across time, it has been mistakenly confused for a sort of "working memory system."
Instead, I argue that it is merely an activated form of LTM. The SEC should play a prominent role in mentally-derived action. For example, if you wanted to initiate a meaningful actopm or understand the symbolic aspects of an action, you would need to distinguish it from the many other actions you have previously encoded. The SEC provides the temporal and symbolic knowledge structures that are unique to time and place and therefore constrains retrieval of a specific action increasing the likelihood that an action will be retrieved. Patients with damage to the prefrontal cortex should have difficulty retrieving some old actions because of their inability to retrieve a specific context in which the action occurs. There are methodological advantages to using the SEC framework, and I will argue how this framework can be distinguished from other popular views of the role of the prefrontal cortex in the execution of intentional behavior.
Action and awareness: from visuomotor corrections to agency
Patrick Haggard, Institute of Cognitive Neuroscience, University College London
Intentional action is one of the most essential attributes of human beings, yet its scientific basis has only recently been investigated. An important element in these investigations involves comparing the neural activity and motor patterns of actions which are "intentional" with those that are not. Beginning with the early work on visuomotor corrections by Marc Jeannerod and colleagues, such studies have shown that intentional action is associated both with characteristic conscious experience, and also with a specific neural circuits linking the frontal and parietal lobes. I will describe recent experiments in which we directly compare motor performance and awareness of visuomotor adjustments in pointing and anti-pointing tasks. The results suggest that action awareness has two components. The first involves effortful intention. The second is based on the extent to which the actual movement matches the intention, and may be computed by an internal forward model.
Cognitive and action-oriented representations of object motion
Francesco Lacquaniti and Myrka Zago
Scientific Institute Santa Lucia and University of Rome “Tor Vergata”Abstract:
People interact daily with moving objects. How accurate are our predictions and what sources of information do we use to predict objects’ motions? These questions have received wide attention from a variety of different viewpoints. On one end of the spectrum are the ecological approaches that assume all the information about the visual environment is present in the optic array, with no need to postulate conscious or unconscious representations. On the other end of the spectrum are the constructivist approaches that assume that a more or less accurate representation of the external world is built in the brain using explicit or implicit knowledge or memory besides sensory inputs. Representations can be related to naïve physics or to externalisation of our own body movements, or to the construction of internal copies of environmental invariants. We address the issue of prediction of objects’ motion in different contexts. Cognitive understanding and perceptual judgment of simple Newtonian mechanics can be surprisingly inaccurate. By contrast, motor interactions with moving objects are often very accurate. The pragmatic action-oriented behaviour and the perception-oriented behaviour may use different modes of operation and different levels of representation.
Visuo-spatial neglect: Art and insight
John C. Marshall, Hilary Osborne and Jennifer M. Gurd:
Professional artists are clearly as likely as anyone else to suffer brain damage to right hemisphere sites, including parietal cortex. This talk investigates the effect of such lesions on their life and work: Do visual artists manifest left visuo-spatial neglect that is indistinguishable from that shown by any other patient? Or does their pre-morbid competence in any way "protect" them against showing severe neglect? How does neglect affect their drawing and painting? Can style-change always be unambiguously distinguished from neglect? Do artists have good insight into their neglect? Answers to some of these questions will be given.
Perception and action in neglect and extinction
A D Milner and R D McIntosh
(University of Durham, UK)Summary
The classical line bisection task has been used extensively in neglect research, and the effects of manipulating the length and spatial location of the stimulus lines have been well studied. We have taken a new perspective on the task by treating the location of the left and right endpoints of the stimulus line as the independent variables of interest. First we have asked how the responses of neglect patients change according to whether the right or the left endpoint location varies – in other words what ‘weights’ are attached to the two endpoints in determining behaviour. We have found a characteristic asymmetry amongst neglect patients, such that the weight attached to the right endpoint exceeds that given to the left. The data suggest that these weights may provide independent information from that offered by directional bisection errors. This alternative description of neglect bisection performance subsumes the traditional analyses based on line length and location.
In a second study, we have applied the same methods of analysis in different behavioural contexts. Neglect patients were required explicitly to bisect the space between two cylinders or to reach between the two cylinders en route to a more distant target. As expected, the right hand cylinder was more influential than the left in determining the bisection responses. In contrast, no such asymmetry was observed in the reaching task, with the spatial path of the hand being influenced equally by the cylinders on the left and right sides. This finding suggests that neglect need not extend to visually guided navigation with respect to obstacles within the array.
In a third study, we have tested a single patient who suffers from left-sided visual extinction, using a similar reaching task. This time, obstacles to reaching were presented bilaterally or unilaterally (on the left or right). The duration of the patient’s view of the array was adjusted in order to ensure extinction on approximately half of the bilateral trials. On each trial, he was required to reach to a target situated beyond the obstacles, and to report verbally which obstacles he had seen. His reaching trajectories took full account of the left obstacle regardless of whether he reported seeing it or not. This remained the case whether made his reach before making his verbal report, or vice-versa. We conclude that visual information does not need to be conscious in order for it to provide effective guidance around obstacles.
Action understanding and imitation: facts and speculations
Giacomo Rizzolatti, Istituto di Fisiologia Umana, Università di ParmaWe live in a word full of objects, sounds and movements. Among all these stimuli, the movements of other living creatures –especially of our conspecifics- are particularly important for us. How we understand the actions of others? What are the neurophysiological bases of this ability? It is possible that this capacity involves merely a visual analysis of the action. There is growing evidence, however, stemming from neurophysiological experiments in monkeys and brain imaging studies in humans, that the mechanism underlying this capacity is based on mapping observed action visual information onto internal motor representation of the same action. Here I will discuss evidence of such a system (“mirror system”), review its properties, and explore its implication for a cognitive capacity fully developed only in humans: action imitation.
Toward understanding the neural mechanisms of 3D shape perception
Hideo Sakata 1,2, Masato Taira 1, Ken-ichiro Tsutsui 1
1 Nihon University School of Medicine, Department of Physiology
2 Seitoku Junior College of Nutrition, Laboratory of Anatomy and Physiology,Marc Jeannerod's discovery of "preshaping" during goal-directed hand movement was the origin of our ideas leading to the study of neural mechanisms of 3D shape perception in the parietal cortex. Michael Arbib proposed his theory of coordinated control program of visually guided hand movement on the basis of the observation of "preshaping". That was a good working hypothesis of our initial study of hand manipulation related neurons(Taira et al. 1990). The most clear cut selectivity was found in the axis-orientation. However, it was obvious that "preshaping" is much more elaborate than the adjustment of hand orientation. Therefore, we introduced 3D objects of simple geometric shapes(cube, sphere, cylinder, cone, ring and square plate) as the targets for grasping. Surprisingly, many of the hand manipulation related neurons were highly selective preferring one out of six different shapes(Murata et al. 2000). Since the study of Giacomo Rizzolatti's group suggest that the motor command signals of various hand movements are generated in the ventral premotor cortex and their efferent copy is sent to area AIP, we started to search for a visual area in the parietal cortex that processes visual signals relevant to the 3D shape perception. Soon we found in the caudal intraparietal(CIP or c-IPS) area on the lateral bank of the intraparietal sulcus(IPS) purely visual neurons that responded to the stereoscopic stimuli with binocular disparity. Two classes of neurons, axis-orientation- selective(AOS) neurons and surface-orientation selective(SOS) neurons, were identified in area CIP. We studied the visual properties of SOS neurons in detail, because the representation of surface is the main step toward the representation of 3D shape, as suggested by David Marr(1982). The first significant results that support Marr's theory was the finding of SOS neurons that responded to the 3D surface in the random dot stereogram. This fits very well to the Marr's hypothesis that the surface orientation is computed from the gradient of binocular disparity. More recently we found many SOS neurons responded to the 3D orientation of the surface defined by texture gradient, which was found to be an important cue of depth by Gibson(1950). We also found the monocular cue of linear perspective for the perception of depth was effective to enhance the response of many SOS neurons to 3D surface. We also have preliminary results to show the CIP neurons discriminating cylinder from angular column. Area CIP is neighboring to area V3A that contains many disparity-sensitive neurons and retinotopically organized. These results strongly support the view that the visual information processing for the perception of 3D shape is going on in and around area CIP.
Daniel Wolpert
Uncertainty in sensorimotor controlSensory and motor uncertainty form a fundamental constraint on human sensorimotor control. We describe how signal-dependent noise arising at the neural level places constraints on performance. Given these constraints we show how goal-directed arm movements can be considered within a general optimal control framework called Task Optimisation in the Presence of Signal-dependent noise (TOPS). We then describe how uncertainty about the state of our body and the state of the world, that is the movement context, can be minimized. We show that to determine the current context the central nervous system can use information from both prior knowledge of how the context might evolve over time and from the comparison of predicted and actual sensory feedback. We show that these two sources of information may be modelled within the central nervous system and combined to derive an accurate estimate of the context that adjusts motor command selection. Together these studies provide a probabilistic framework for sensorimotor control.
How the Sign frame (SiF) is embedding the Speech frame (SpF)?
(an ontogenetic puzzle if not a puzzling phylogenetic scenario for language)C. Abry, J.-L. Schwartz and A. Vilain
ICP, UMR CNRS 5009, Stendhal-INPG (abry@icp.inp.fr)
We start from a non a priori predictable rendez-vous, discovered quite recently: namely the co-emergence of a basic speech control, coarticulation, with the first word(s) at the end of one year. Consequently we ask three empirical questions: (i) what is needed in speech control to get coarticulation? (ii) what is needed in sign control to get a word? (iii) what is the metric which can allow cross-talk and integration between sign and sound?
(i) At about half a year, canonical babbling emerges suddenly: [bababa…] or [dadada…] typical behaviors are clearly neither controlled (in access consciousness) as syllables, vowels or consonants. What MacNeilage has dubbed a premotor, i.e. SMA frame (hence SpF) is from the point of view of motor control, a control of the initiation a cycle and of its self pacing, using the mandibular system as a carrier. The carried and contact articulators are controlled later. Typically in [bababa…] movement patterns the upper lip is less and less compliant to the pushing up of the lower lip carried by the jaw, hence showing finally evidence of an autonomous contact control. At one year the tongue can resist to the elevation of the jaw carrying the lower lip toward the upper lip, using a depressor tongue muscle such as hyoglossus: consequently during the very closure phase for [b], the configuration of the [a] vowel can be fairly maintained, which is properly coarticulation control. Why does this type of control emerge together with the first words?
(ii) Prerequisites for sign control include probably a fair number of devices. One can assume that objecthood theories can offer a basis for the grounding of word learning. Event-, object-, agent-indexing —i.e. the skill to track attentionally (with vivid conciousness) from one to three targets— is first achieved by eye, long before index pointing (9-12 months), after some steps in the shared attention mechanism have been completed. The multimodal "where/there" localisation system is recruited long before the "what/that" object properties system (the latter at about 10-12 months, with object shape —affordance— before object colour), and both are prempted by the event "when/then" system. The wh- and th- stems will prove as powerful "root-tools" for grammar in the development of the argument-predicate structure.
(iii) Recent movement data on rhythmic behaviour of hearing babies provided a 3 Hz mode; and the same 3 Hz mode, plus a half 1.5 Hz mode, for babies imitating (with empathy or mirror neurons) signing parents. The 3 Hz mode has otherwise proven to include the rhythmic behavior of the jaw as well. Our hypothesis is that the 3 Hz mode is used by the speech frame and that the 1.5 Hz is used by the sign frame. In nonsigning babies the pointing discrete gesture (SiF) has proven to be of this order of duration. Moreover it is fairly synchronized with toddlers' speech to the point to be included in the Mean Length Utterance index. The metric of the SpF is the syllable and the metric of the first words is the foot, the prosody of the first words being typically one-two syllables, i.e. a foot.
Finally when the word gets the foot as a template, coarticulation can be used for enhancing word harmony: no change in control of local contacts ("consonants") or global postures ("vowels") is licensed throughout the foot. This delivers a non ambiguous parsable word form in memory, perfectly phased papa or tata types (daddy in French and Serbo-croatian) throughout the articulatory loop.
Reaching and grasping with whole body movements : the problem of co-ordination
N. Allami, Y. Paulignan and D. Boussaoud
Institut des Sciences Cognitives CNRS/UCBL UMR5015, 67 bd Pinel, F69675 Bron Cedex
In everyday life, we are reaching for and grasping many different objects located at various distances and with variable sizes and shapes, by use of many body segments (hand, arm, trunk…). Our aim is to find out how the co-ordination of the different segments is achieved during the movement in the prehension space.
A kinematic analysis of parameters of the two principal components of prehension movements in humans (reaching and grasping) was performed. Subjects were instructed to grasp objects from four different starting positions. Three factors were tested : the distance between the subject and the object, the subject's orientation and the object's size. The recorded parameters were the total movement time, amplitude and latency of wrist velocity peak, amplitude and latency of grip size peak, finger closure time and total grasping time.
The initial position and posture of the subjects do not affect fundamentally the classical description of prehension movements. Transport of the hand with one single peak on the wrist velocity curve and grasping with one single peak on the grip size curve remain valid. Moreover, an important de-coupling between the components was present as shown by the difference in time between the onset of each components. This is a very strong argument in favour of the existence of distinct channels. We have also found that the maximal amplitude of the grasp is invariant according to the distance. However, the existence of an invariance of the time of the finger closure across conditions underline the necessary co-ordination between the channels. At last, the existence of a constant phase of the grasp closure constitute an important indication of the phase of movement where we must look for the mechanisms of the co-ordination.
Vestibular control of arm movement
J. Blouin, J.-P. Bresciani, G.M. Gauthier, C. Bourdin and J.-L. Vercher
UMR Mouvement et Perception, Marseille
Recently, vestibular signals have been found to contribute to the online control of goal-directed arm movements (Bresciani et al. 2002). We further explored the vestibular control of arm movement testing the possibility to keep the unseen arm stationary in space while subjects are submitted to passive whole-body rotations. Subjects (N=6) sat at the center of a rotating platform in the dark fixating a chair-fixed light located straight-ahead. Their arm was extended 10° to their right. Subjects successfully kept the hand-in-space still during rotations. On average, their hand was at 10.12°, 9.60° and 9.31° after 20°, 30° and 40° rotations, respectively. In a second experiment, we tested subjects' possibility to adapt to a new arm-to-body relationship through vestibular signals. Subjects were asked to keep the hand above a target located 10° to their right on an earth-fixed table. Position of the target was controlled by the chair position signal. Subjects fixated a chair-fixed light and a LED was fixed on the finger. Rotations were the same as in the first experiment. During rotations, the target was moved with a gain of 0.25 with respect to the chair, in the opposite direction. 40 trials were performed per rotation magnitude. Subjects perceived the moving target as being stationary. After the adaptative procedure, visual feedback of the index was removed. Subjects showed a strong post effect as they moved the hand with a gain of 0.20, 0.14 and 0.12 with respect to the chair rotation. These results highlight the possibility to process vestibular signals to control arm movements online during body-in-space motion.
References
Bresciani et al. (2002). Galvanic vestibular stimulation produces online arm movement deviations when reaching towards memorized visual targets. Neuroscience Letters 318: 34-38.
How intention-in-action is realized within the motor system: a TMS study
M. Bonnard, M. Camus, J. de Graaf and J. Pailhous
« Mouvement et Perception », UMR 6152, Université de la Méditerranée, CNRS, Faculté des
Sciences du Sport, 163 avenue de Luminy, CP 910, 13288 Marseille. Cedex 9, France.
Several studies have demonstrated that corticospinal (CS) excitability is task-dependant. But, these experiments do not allow understanding whether these task-related modulations of CS excitability are due to differences in the states of the motor system while in action or to differences in intentions. In order to dissociate these two aspects of the performance, we performed an experiment in which the subject was engaged in the same motor task (rhythmical flexion/extension movement at the wrist) but with two different intentions with respect to potential perturbations (in the direction of an increased flexion), induced by applying TMS on the hand area in M1. In one condition (NINT: non-intervention), subjects were asked to prepare to let the movement be perturbed so that they can see their movement deviate from the initial trajectory. In the other condition (COMP), they were asked to prepare themselves to compensate for the potential TMS-induced effects such that no movement deviation at the wrist joint can be seen. Throughout the experiment, a continuous visual-feedback of the wrist angular displacement showing the movement during the preceeding 5 seconds was given on a computer screen, and two horizontal lines indicated the maximal flexion and extension angles. The results showed that, in the COMP condition, the movement did not deviate or deviated much less from its initial spatial trajectory than in the NINT condition. Moreover, when comparing MEPs observed for successful trials in the two conditions, MEPs for the wrist flexor were found to be smaller for NINT than for COMP conditions. In conclusion, these results suggest that modification of the CS excitability is the mechanism allowing the intention to be realized. Moreover, it appears that, to a certain extent, a subject is able to intentionally prepare him/herself to compensate for the effect of a TMS pulse which has important methodological implications.
Sudden awareness of low level sensorimotor behavior
M. Bonnard, E. Varraine and J. Pailhous
UMR 6152 « Mouvement et Perception », Université de la Méditerranée, CNRS, Faculté des
Sciences du Sport, 163 avenue de Luminy, CP 910, 13288 Marseille. Cedex 9, France.
The motor control can be achieved in the absence of awareness, even when performed intentionally. The aim of this study was to understand the mechanisms of the sudden awareness of our own movement. This was studied in locomotion because it is an automatic behavior which can be intentionally modulated. Subjects walked continuously with the instruction to maintain either a constant walking speed (COMPENSATION condition) or constant propulsive forces (NO-INTERVENTION CONDITION); they were sometimes faced to slow variations in resistance that they had to detect. The results show that 1°) the subject remains unaware of his force increase (in COMPENSATION) or his walking velocity decrease (in NO-INTERVENTION) for a long time, although these modifications go largely beyond the variability range in which he is able to intentionally control his force (in NO-INTERVENTION) or his velocity (in COMPENSATION) and 2°) the detection of the resistance increase occurs at the same time in both conditions. We conclude that the sudden awareness of a movement pattern produced at a low level emerges from the interaction between a top down mechanism where the intentional control of goal feedback delays the aware perception of the other sensory sources and a bottom up mechanism where high level mechanisms of sensori-motor integration come into play beyond a discrepancy threshold between different sensory information.
Autobiographical memory and conscious experience: A preliminary event related fMRI study
A. Botzung1, L. Manning1, C. Scheiber2 and C. Paulos3
1. Laboratoire de Neurosciences Comportementales et Cognitives (UMR 7521)
2. Institut de Physique Biologique (UMR7004)
3. Laboratoire des Systèmes BioMécaniques et Cognitifs (IMFS, UMR 7507)
Within the memory research, event-related fMRI is used to carry out post-hoc selection of memory performance. Recognition memory for words (RMW) appears to be one of the domains most frequently studied. Among several other researchers, Henson et al. (2000) analysed RMW and degree of confidence. However, autobiographical memory and event-related fMRI, was reported only once, to our knowledge, by Maguire et al (2001) whose aim was to analyse the content and the age of recollections.
Our aim in the present study was to obtain a pattern of brain activations during an autobiographical memory task, in normal subjects, and to investigate the likelihood of different loci of activations related to distinct conscious experiences, i.e. Tulving’s Remember / Know paradigm. A series of preliminary steps were necessary in order to achieve our goal. Thus, we had to ensure the methodological work allowing the use of event-relate fMRI procedure, which involved the comparison with brain activations observed through the box-car design. Such a comparison was carried out for three normal subjects during their performance on the control task. A reaction time study was also performed to guarantee the adequacy of the control task to its use during brain scanning. The pool of autobiographical recollections was obtained by means of the Crovitz Test (Hodges’ modified version), which allowed getting 50 to 80 recollections equally distributed into four lifetime periods. Every souvenir was then coded using two words. The experimental task, autobiographical memory, was studied in two subjects through event-related fMRI. Moreover, post-hoc selection was carried out using their responses R/K to every scanned recollection. Data analysis showed that the control task was appropriate, more importantly, brain activations are in accordance with the reports on autobiographical memory studied in PET scan and fMRI block design. Data analysis on the R/K model in currently carried out.
Cognitive versus sensorimotor maps in the vestibular control of goal-directed movements
J.-P. Bresciani, J. Blouin, G.M. Gauthier, F. Sarlegna, C. Bourdin, and J.-L. Vercher
UMR Mouvement et Perception, CNRS et Université de la Méditerranée, Marseille
Using passive whole-body rotations, we recently evidenced that vestibular signals allow to preserve the reaching accuracy of an ongoing arm movement when body orientation is modified after movement onset. The present experiment was aimed at differentiating cognitive from more sensorimotor aspects of the processes underlying such “vestibular control” of goal-directed arm movements. Seated subjects deprived of visual cue were instructed to reach for an earth-fixed memorized visual target, body orientation being changed either before (update condition) or during (online condition) reaching movement execution. The respective effects of an adaptive modification of (a) vestibular input/arm motor output matching (sensorimotor adaptation) and (b) vestibular input/perceptive map of visual space matching (perceptive adaptation) on each of the two reaching conditions were compared. Reaching movements performed after rotation completion (update condition) were slightly affected by the sensorimotor adaptation phase (F(1,5) = 24.93, p<.005) and strongly affected by the perceptive adaptation phase (F(1,5) = 251.30, p<.00001), the effect of perceptive adaptation being significantly greater than the effect of sensorimotor adaptation (F(1,5) = 8.85, p<.05). On the other hand, reaching movements performed in the online condition were strongly affected by the sensorimotor adaptation phase (F(1,5) = 143.04, p<.0001) but remained statistically unaffected by the perceptive adaptation phase, the effect of sensorimotor adaptation being significantly greater than the effect of perceptive adaptation (F(1,5) = 9.15, p<.05). Our results strongly suggest that whereas updating the egocentric position of surrounding objects after passive body displacements in darkness mainly implies vestibular signals processing by the structures involved in the cognitive mapping of visual space, vestibular control of ongoing goal-directed movements relies on more sensorimotor processes by which subjects merely counteract sensed body displacement to continuously cancel motion-induced effects on motor output.
Development of Movement Imagery Vividness Through a Structured Intervention
C. Calmels and C. Berthoumieux
Institut National du Sport et de l’Education Physique, Paris, France
Individual differences in imagery ability and more especially vividness can influence the learning and the performance of motor and cognitive skills (e.g., Denis, 1982; Isaac & Marks, 1994). In particular participants classified as “good imagers” according to their self-rated imagery vividness performed better than their counterparts classified as “poor imagers.” The development of imagery vividness through structured interventions could thus be viewed as an asset as it may lead to subsequent enhancements of performance. Studies in therapeutic context have shown that vividness could be developed through verbal reports training of an imagined behavioral experiences (e.g., Lang, 1979). The interpretation of these results to sport settings is questionable. Thus the purpose of the present study was to examine the influence of a structured imagery intervention on self-reported vividness of movement imagery of four national softball players. A multiple-baseline design across individuals (Kazdin, 1982) was employed. Between nine to 12 data collection points were gathered over time. Data was collected via the measures of the Vividness of Movement Imagery Questionnaire (VMIQ; Isaac, Marks, & Russel, 1986), which assesses vividness from an external and internal imagery perspective. The treatment in the present study consisted of an audiotaped imagery program composed of 28 sessions, which required the participants to spend ten minutes a day practicing. Imagery scripts were based on stimulus and response propositions (Lang, 1979). Results stemming from a visual inspection (magnitude of changes) showed general improvements in vividness scores on external and internal imagery perspectives with increases of 5 % to 32 % between baseline and intervention for all the participants. Statistical analyses supported mainly the results of the visual analysis. Significant changes were observed except for (a) participant 3 on her external and internal imagery perspective, and (b) participant 1 on her external imagery perspective. Findings were discussed with regard to individual cases and Lang’s (1979) bioinformational theory.
References
Isaac, A.R., & Marks, D.F. (1994). Individual differences in mental imagery experience: Developmental changes and specialization. British Journal of Psychology, 85, 479-500.
Isaac, A., Marks, D.F., & Russel, D.G. (1986). An instrument for assessing imagery of movement: The vividness of movement imagery questionnaire (VMIQ). Journal of Mental Imagery, 10, 23-30.
Denis, M. (1982). Imaging while reading text: A study of individual differences. Memory and Cognition, 10, 540-544.
Kazdin, A.E. (1982). Single-case research designs. Methods for clinical and applied settings. Oxford: Oxford University Press.
Lang, P.J. (1979). A bio-informational theory of emotional imagery. Psychophysiology, 16, 495-512.
Duration of Physical and Mental Execution of Gymnastic Routines
C. Calmels and J. Fournier
Institut National du Sport et de l’Education Physique, Paris, France
In light of previous research in the cognitive psychology and neuroscience literature (e.g., Decety & Jeannerod, 1996; Decety, Jeannerod, & Prablanc, 1989), this study aims to examine the temporal organization of routines in artistic gymnastics executed under mental and physical conditions. The participants were twelve national female gymnasts. They were asked to perform their floor routines mentally, and then perform the same routines physically. During the imagery routine execution, the gymnasts seated at a table were invited with the use of a smaller ruler to signal the six different stages of their routines by tapping on the tabletop. This procedure allowed for the collection of total time for the routine and separate times for each of the six aforementioned six stages. Relative durations for the six stages were thus calculated for each type of rehearsal. For each participant, performances were filmed, and the durations of the mental and actual routines were timed. The results showed that mental movement times were shorter than physical routine times. It was concluded that the speed of visualization depends upon the situation in which the gymnasts visualize, as well as upon the function, which the athlete attributes to the use of imagery. We also observed a trend when comparing the different stages of the relative duration of mental and actual routines. If confirmed, we hypothesized that the lengthening of the relative duration of certain stages under mental conditions could be linked to the perceived difficulty of the gymnastics elements.
References
Decety, J., & Jeannerod, M. (1996). Mentally simulated movements in virtual reality: Does Fitts's law hold in motor imagery? Behavioral Brain Research, 72, 127-134.
Decety, J., Jeannerod, M., & Prablanc, C. (1989). The timing of mentally represented actions. Behavioral Brain Research, 34, 35-42.
The appropriate processing of hypothetical predictions and necessary deductions by the readers
N. Campion
Limsi, Université Paris XI
According to current models of text comprehension, the readers make inferences to build a coherent representation of the entities and situations that are described in the text. A main type of inferences that the reader of a narrative can make corresponds to probable consequences of events that are described in the text. However, like many inferences, these predictions are not always verified, because they are mainly based on general knowledge of what should probably occur in the type of situation that is described. Thus, many theorists consider that the predictions are not made by the readers. In a former study (Campion & Rossi, 2001), our results suggested that readers activate the predictions when the reading of the text also activates knowledge related to these predictions. Similar results were provided in an other study which also indicated that an explicit event was more accessible than a prediction in the representation of the readers (Cook, Limber & O'Brien, 2001) . Accordingly, the readers would only infer partial and imprecise expectations. The present study examined an other hypothesis: when activating a prediction, the reader would categorize it as an hypothetical fact that waits further confirmation. Therefore the prediction would not be processed like is an explicit event or a logical deduction, both being necessary facts according to the text. In 3 experiments, we compared probable predictive inferences to necessary logical deductions. These inferences had to be drawn from short narrative texts that were successively read by students on a computer screen. In experiment 1, the data of a lexical decision task indicated that both types of inferences were activated during reading. In experiment 2, the data of a judgment task confirmed that the predictions were considered hypothetical and the deductions necessary by the participants: after the reading of a text, a sentence expressing that an event "may have happened" was quickly judged true when it probed a prediction but not when it probed a deduction. In experiment 3, the data indicated a slow down when reading a critical sentence from a text when that sentence was presupposing that an inference corresponding to an earlier portion of the text was false instead of true. In addition, the amplitude of the slowing down was smaller for the predictions than for the deductions. Accordingly, the predictions are inferences that are activated but that are considered as hypothetical facts. Thus, the readers distinguish them from necessary deductions or explicit facts.
Vision for perception and action in the fronto-parallel and sagittal plane
Y. Coello1, S. Richaud1, Y. Rossetti2, M. Boucart3 and J.-C. Hache3
1: URECA - UPRES EA 1059, Université Charles de Gaulle, F.59653 Villeneuve d'Ascq
2: INSERM Unité 534. F.69500 Bron.
3: CHRU – Lille. Service Exploration Fonctionnelle de la Vision. F. 59037 Lille.
Among supports for distinct visual sub-systems responsible for perception and action, one of the main psychophysical evidences in healthy subjects came from studies about visual illusion such as the induced Roelofs effect (IRE, Bridgeman 1991). With the IRE, the location of a visual target surrounded by an off-centre frame is misperceived when evaluated verbally, but not with a reaching response. This dissociated effect was interpreted within the framework of two independent representations of visual space devoted respectively to categorisation and egocentric localisation. These "cognitive" and "sensorimotor" representations have been assumed to be produced through specific anatomical pathways stemming from the primary visual cortex (respectively the ventral and dorsal streams, Milner & Goodale 1995, Rossetti 1998). To account for the dissociation found with the IRE, it was suggested that only the cognitive system is sensitive to contextual information (Bridgeman, 2000). However this view has been challenged by recent psychophysical studies demonstrating the influence of environmental cues on distance perception and the guiding of movement (Coello & Magne, 2000, 2002). In particular, distance performance in pointing task is dependent on the structure of the action space, with little awareness of the variation in performance. Furthermore, by replicating the IRE but dissociating the near-far and right-left dimensions we showed that the IRE in the right-left dimension gives rise to a perceptual misperception of target position with no effect on motor performance. Conversely, when the IRE was induced in the near-far dimension a misperception of the target position affected both perceptual and motor responses (Coello et al., 2002). This dissociation indicated that the spatial constraints of the task, and not only the nature of the response, interfere with sensitivity to contextual information leading to visual illusions. The contribution of the ventral and dorsal pathways to distance coding is discussed in the light of patients performance when submitted to the IRE, in particular patients suffering of brain lesion affecting selectively the dorsal stream.
Neglect affects intention and execution of a movement sequence ending to the left
S. Courtois-Jacquin 1, G. Rode 1,2, Y. Rossetti 2, B. Ferraton 1 and D. Boisson 1
1 Service de Rééducation Neurologique, Hôpital Henry Gabrielle, Hospices Civils de Lyon
2 INSERM U534, Espace et Action, Bron
Introduction. Unilateral neglect provides an opportunity to study the mechanisms underlying space related behavior. This complex syndrom may affect different modalities of sensory domain, as visual, auditory, somatosensory or olfactive and motor domain. In the motor domain, patients with unilateral neglect may exhibit motor neglect for movements of the limb contralateral to the lesion side, but also specific impairments of directed movements of the ipsilesional limb in the side opposite to peripersonal space. These impairments concern the initiation (directional hypokinesia) and the execution (directional bradykinesia) of movement. An interesting question is to assess whether unilateral neglect may also affect the preparation or anticipation of action, i.e. intention.
Material and methods. We have investigated whether a deficit would be observed in the first reaching movement of a grasping action when the subsequent movement is directed to the left. Patients had to reach and grasp a tennis ball placed on their sagittal axis and then throw it in a left or right basket. Movements of 15 brain damaged patients (8 RBD with unilateral neglect, 3 RBD without unilateral neglect and 4 LBD) were recorded in 3D with a Vicon system (Oxford Metricsâ). Several parameters were studied: Reaction Time (RT), Movement Time (MT), wrist Peak Velocity (PV) and Time to Peak Velocity (TPV), for the first (reach to grasp) movement. Trajectories of wrist, index and thumb markers were also analyzed in the XY plane.
Results and discussion. A significant interaction between patient groups and side of reach was observed for RT, such that neglect patients were slower to initiate their movement when ending to the left as compared right. No comparable effect was found in the control groups. This result suggests that the preparation of a movement sequence ending to the left is subjected to unilateral neglect. Anticipation of the ultimate goal of an action leds to a retrograde transfer of neglect onto action elements that have no right-left component by themselves.
Fictive execution of a forthcoming movement. A fMRI study
J. de Graaf, C. Galléa, J. Pailhous, J.-L. Anton1, M. Roth1and M. Bonnard
UMR 6152 Mouvement et Perception,
Université de la Méditerranée Faculté des Sciences du Sport,
163 avenue de Luminy, CP 910, 13288 Marseille CEDEX 09, France
1Centre IRMf, CHU Timone, 264 Rue Saint Pierre 13385 MARSEILLE Cedex 05, France
Since external forces vary, no one-to-one relation exists between the motor commands send to the effector system and the movement that they produce. We studied how the brain organises constant motor output with changing external forces. Subjects were instructed to make rhythmical movements at the wrist with a self-chosen frequency. A manipulandum allowed us to modify the resistance to movement by increasing the internal friction. In one condition, the subjects performed the so-called “reference movement” with low internal friction of the manipulandum and self-chosen amplitude. In two other conditions, the so-called “task conditions”, the internal friction was increased and the subjects were instructed either to maintain the force applied to the manipulandum, either to maintain the movement amplitude. Before each task condition, the subjects repeated a condition in which they executed the reference movements. We expected to find differences in cerebral activity between the two task conditions. However, we found differences between the reference conditions, that we considered as being pure control tasks, in cerebral structures implicated in motor imagination, memorisation and attentional processes. This shows the involvement of processes preparing to the subsequent task condition and was interpreted as revealing a fictive execution of the forthcoming task. Furthermore, the force reproduction task clearly involved more cerebral activation during the preceding reference condition, maybe because it is an abstract task. In conclusion, our study of cerebral processes underlying motor behaviour revealed a real simulation of action that is not observable at the behavioural level. This strongly suggests that whereas behaviour is constrained by the actual task, the brain is free to execute a forthcoming task while in the meantime executing the actual behaviour.
The cerebral supervision of cerebellar plasticity does not require error signals
M. Dufossé1, A.A. Frolov2, A. Kaladjian3
1- INSERM U483, Université Pierre et Marie Curie, Paris, France.
2- Institute of Higher Nervous Activities, Academy of Science, Moscow Russia.
3- Faculty of Medicine, Hospital La Timone, Marseille, France.
Our goal was to understand how each cerebellar Purkinje cell receives from the inferior olive the supervising signal suited to modify its parallel fiber contacts.
We developed a neural network model which links a visual target to the finger's tip displacement via two pathways, a direct cortico-cortical path and an indirect side-path via the cerebellum.
The central part of our model consists of a set of cerebral modules, which perform locally several operations. In the upper layers, a first learning rule is based on the match between the motor command and the visual perception of the resulting finger's tip displacement. A second learning rule is based on the mismatch between target and final position. In the intermediate layer, a third learning rule adjust the gain of the plastic cerebello-thalamo-cortical pathway. In the lower layers, the signals from the two pathways are added into an output motor command, which produce a change of the muscle stretch reflex thresholds.
In parallel, the motor command is sent to the inferior olive in order to supervise the cerebellar learning. A fourth learning rule schematizes the cerebellar LTD process.The analytical formulation shows that a random matrix of transformation between the vector of motor commands and the vector of climbing fiber activities is enough to supervise the cerebellar learning. The adaptive processes that take place in the different sites of the cerebral cortex and the cerebellum do not interfere but complement each other. Although other signals from the periphery may also be used, the cerebral origin signals that supervise the cerebellar learning probably do not code any error in movement performance.
Action identification in patients with schizophrenia
N.Franck (1), C. Farrer (1), N. Georgieff (1) (2), and M. Jeannerod (1)
(1) Institut des Sciences Cognitives, Bron, France.
(2) ITTAC, Lyon, France.The study of the clinical as well as of the cognitive patients’ phenotype is a major attempt in research on schizophrenia. It is essential to define the trouble we want to cure or we want to prevent. The present experiment aimed at identifying the mechanisms of what Kurt Schneider considered as the core of schizophrenia: the feeling of influence.
Now this symptom, such as other first-rank symptoms, seems to be relevant to a dysfunction of the awareness of one's own action and to a dysfunction of recognition of actions performed by others (Frith, 1992). It raises the point of which cues are normally used for producing conscious judgements about the origin of an action (self vs. others). A change in the cortical pattern of activation in relation to action might represent a mechanism for the schizophrenic pathological interpretation of reality (Daprati et al, 1997). The present experiment aimed at understanding which factors are involved in the determination of the origin of an action.
Twenty-four schizophrenic patients and 29 normal subjects were required to perform movements with their right hand holding a joystick and to monitor their execution on a computer monitor presenting pictures of a virtual hand. Biases were introduced in some trials, modifying the parameters of the movements. Participants had to judge whether presented movements was their own or not. When subjects did not recognize their own movement or consider a modified movement as their own, errors were recorded.
In this experiment, patients with delusion of influence made significantly more over-attribution errors than controls, and more than non-influenced patients. These results provide arguments in favor of the existence of a cognitive mechanism underlying delusional influence, a key feature of schizophrenia.References
Daprati E, Franck N, Georgieff N, et al: Looking for the agent: an investigation into consciousness of action and self-consciousness in schizophrenic patients. Cognition 65, 71-86, 1997.
Frith CD: The cognitive neuropsychology of schizophrenia, Lawrence Erlbaum Associates, Hove, 1992.
Visual motion integration, temporal coding and mind-brain dissociation in autism: Neuropsychological and epistemological perspectives
B. Gepner
Service de psychiatrie infanto-juvénile, Centre Hospitalier Montperrin & Laboratoire Parole et Langage, UMR CNRS 6057, Université de Provence, Aix-en-Provence.
Perception and integration of visual motion provide crucial experiences for processing the physical and human environment as continuous and coherent, and thus play a major role in the consciousness of unity, of self and others. In the same way, perceiving and integrating the speech dynamics is essential for coding the basic properties of language.
Since the original work by Gepner et al.(1995), a growing evidence demonstrates that subjects with autism have difficulties in perceiving and integrating visual motion, especially fast visual speeds. Besides, our group has recently showed that autistic children over-categorize inter-categorial phonems, and that their categorial perception is normalized when phonems are displayed slowly. Finally, our group has also shown that the same autistic children have a deviant motor anticipation in a load-lifting bimanual task, by using preferentially feedback rather than feedforward, which leads to a delayed stabilization. Accordingly, children with autism have specific deficiencies to process visual, auditive and sensorimotor stimuli on line, especially when they are displayed rapidly. Consequently, autistic children would experience dissynchronization between sensory inputs from various modalities, as well as sensorimotor detuning. Thus, their simultaneous, continuous and coherent perception and comprehension of the world and action on the world would be severely affected.
A common source of these anomalies may plausibly be a deficit of temporal coding, involving sensori-cerebellar pathways devoted to speed and temporal coding of sensory inputs, and cerebello-motor-premotor pathways as well as basal ganglia that contribute to real-time fine-tuning of motor outputs.
Reciprocally, according to a dualist-interactionist approach of the mind-brain relations, such anomalies of temporal coding would disturb the transformation of intention into action, and thus contribute to a dissociation between mind and brain-body within the autistic person (Gepner, submitted). This view may bring new guidelines for the reeducation of this population.References
Gepner, B. & Mestre, D. (2002). Brief Report: Postural reactivity to fast visual motion differentiates autistic from children with Asperger syndrome. Journal of Autism and Developmental Disorders, 32, 231-238.
Gepner, B. (2001). « Malvoyance » du mouvement dans l’autisme infantile ? Une nouvelle approche neuropsychopathologique développementale. La Psychiatrie de l’Enfant, XLIV, 1, 77-126.
Gepner, B. (soumis). Relations psychisme-cerveau, dualisme interactionniste et gradients de matérialité. In J-L. Petit Repenser le corprs, l’action et la cognition avec les neurosciences, Intellectica.
Gepner, B., Deruelle, C. & Grynfeltt, S. (2001). Motion and emotion : a novel approach to the study of face processing in young autistic children. Journal of Autism and Developmental Disorders, 31, 37-45.
Gepner, B., Mestre, D., Masson, G. & de Schonen, S. (1995). Postural effects of motion vision in young autistic children. NeuroReport, 6, 1211-1214.
Gepner, B., Rondan, C., Masson, G., Castet, E., Deruelle, C., Tardif, C. & Mestre, D. (2002). Optokinetic nystagmus is markedly impaired in children with autism: another argument for a visual motion processing deficit. Poster accepted at 15th Congress of International Association for Child and Adolescent Psychiatry and Allied Professions (IACAPAP), New Delhi, 29th October-2nd November.
Tardif, C., Gepner, B. & Rey, V. (à paraître). Evaluation du système phonologique explicite chez des enfants autistes. Parole
Schmitz, C., Assaïante, C. & Gepner, B. (à paraître). Modulation de la réponse anticipée dans une tâche de délestage bimanuel : étude chez l’enfant sain et l’enfant autiste. TIPA.
Situational Awareness and visual representation: A way of perceiving and acting.
A. Koustanai
Institut National de REcherche sur les Transports et leur Sécurité, Salon-de-Provence
A considerable amount of road accidents results from failure to detect hazards. Indeed, in most cases drivers claim they do not to see anything before the collision.
This inability to see events or elements is closely related to phenomena known as "change blindness" or "inattentional blindness". Previous research reports that the implication of "alertness-attention" and expectation played a particular role in conscious explicit representation of the outside world.
The description and analysis of more than 500 road accidents in a detailed database allowed classifying several scenarios that produce failure in the detection of hazards. Several thousands of observations have been compiled to determine what type of attentional conditions are more likely to yield such "blindness" in driving activity in terms of "situational awareness", i.e. action (goal reach at the moment of the accident), knowledge (experience of the road) and environment (road type and configurations)
Results from database analyses show that for all types of accident the main factors remain the same (e.g. in/out intersection, in/out built-up areas), whereas the components of available information are quite different. In most cases accidents are due to the general road difficulties (e.g. environmental complexity). However, failure to detect hazards can also be due to misperception of visual attributes of road users (e.g. two-wheeled vehicles or pedestrians) and to misinterpretation of the situational components configuration (e.g. manoeuvre feasibility or road traffic conditions).
Based on these results, we decided to investigate change blindness experiments in order to link situational awareness with attentional processes. First, we will compare static (pictures), and dynamic displays (films). Second, we will compare passive and active observers. We hypothesize that explicit representation is a function of current task, and level of driving experience.
Goal-directed action representation in autism
N. Labruyere (1), N. Georgieff (1) (2), and T. Zalla (1)
(1) Institut des Sciences Cognitives, Bron, France.
(2) ITTAC, Lyon, France.
Autism is a developmental disorder characterised by severe difficulties in communication, social interaction and executive functions. According to Leslie (1991) and Frith (1989), an impairment of the “Theory of Mind”, an innate functionally specialised system which allows the acquisition of mental concepts and the construction of metarepresentation, would be the core deficit of autism. For instance, Baron-Cohen et al., (1986) have shown that autistic children are impaired in understanding intentional stories despite a preserved ability to understand stories depicting behavioural and mechanical-physical events.
The aim of the present study is to investigate the ability to represent goal-directed actions in subjects with autism. Goal-oriented actions differ from both mechanical and intentional events in that they are represented in memory as sequences of structured events sharing the same goal. According to our hypothesis, the cognitive processes subserving goal-directed action representation would be impaired in autistic individuals.
In the present experiment, 16 autistic subjects were presented with series of five pictures depicting component events of four types of sequences: (a) simple goal-directed actions (b) complex goal-directed actions, (c) social actions and (d) physical stories. They were requested to reconstruct the correct event sequences under two conditions. In the mixed condition, children ability to sort the relevant events was challenged by presenting concurrently actions belonging to four different familiar activities. In the non-mixed condition, the same sequences were presented separately in order to minimise working memory demands. Their performance was compared with those of mentally retarded subjects (n=14) and normal subjects (n=16). Autistic children were impaired in organising goal-directed events (both short-term and long-term sequences) in their correct temporal order in both conditions (mixed and non-mixed), while they had no difficulty in ordering the mechanical sequences.
The present study indicates that an impairment in the ability to represent human actions as structured sequences of goal-oriented events might contribute to explain social and executive dysfunctions in autism.
Effects of perceived visual coherence on visuo-manual coordination
C. Lalanneab*, M.Zamania and J. Lorenceaub
a France Telecom R&D, DIH/UCE/RCE, 38-40 rue du Général Leclerc, Issy-les-Moulineaux, 92131 Cedex, France
b Unité de Neurosciences Intégratives et Computationnelles, UPR 2191 CNRS, 1 Av. de la Terrasse, Gif-sur-Yvette, 91198 Cedex, France
In order to study the interaction between on-line sensory processing and control of goal-oriented manual action, we design a dual task of visuo-manual coordination. Our experimental design implies driving and tracking computer generated visual shapes of varying perceptual coherence with a joystick.
In two alternate and successive blocks of trials, 10 subjects had to drive an outlined rigid geometrical stimulus along cued conic trajectories (driving task), and to track the centre of the same stimulus describing fixed uncued conic trajectories (tracking task). Stimuli were diamond, cross and chevron, and movement could be either clockwise or counter-clockwise. No visual feedback of the active hand was given during the task. Visual coherence was manipulated by presenting the stimuli behind static contrasted apertures, such that object motion resulted from the visual integration of the ambiguous motion of their contours. At the end of each 7 seconds trial, the subjects were asked to evaluate the perceptual coherence of the visual stimulus on a numeric scale independently of their motor performance. Coherence responses and the spatial positions of the joystick were recorded.
Results showed that (1) active movements did not facilitate a coherent interpretation of visual shapes’ motion, (2) motor performance were degraded at a low coherence level, especially in the tracking task, where we observed important spatial phase lags and hypermetric mean amplitude ratio, and finally (3) on-line adjustments were more variable in the tracking task. This pattern of results is consistent with the proposal that on-line visual processing of shape and motion differently interacts with the planning of voluntary motor actions, according to the goal of action. Implications within the framework of sensorimotor integration and perception-action coupling are discussed.
The agent of intentional action
D. Legrand
CEPERC - Département de Philosophie, Université de Provence, Aix-en-Provence
An intentional action is classically defined as an action done for a reason. In such an account, the agent is defined as the one who entertains the right sort of mental states and who has a first-personal experience of his intentional action. However, I argue that this view is too reductive. Here, I propose to specify and enlarge it in three main points. First, a closer analysis of the kind of consciousness associated with intentional agency leads to the conclusion that being an agent is more than accessing one's mental states. It is also accessing the world according to one's goal, in function of the current action. Moreover, several types of action awareness has to be dissociated to account more specifically for the experience of being the agent of an intentional action. It includes not only the consciousness of one's intention, but also of oneself as being actively engaged in the execution of the action, of the goal of the action, and of the sensorimotor properties of the action. Second, on the top of "mental intention", one can describe "bodily intention". Indeed, the body presents an activity oriented towards the world according to a "body schema". By this, the body is more that blind machinery and testifies to intentionality. The agent can thus not be reduced to mental states: he is and has a body, embedded in a world. Third, intention is not the only causal source of intentional action. External forces has also to be taken into account. Intentional action is also passivity. The agent is also receptivity. Finally, we end up with a new definition of the agent. He can not be reduced to the one entertaining mental intention who turns his eyes inward. Rather, he is also and first of all a body acting in a world.
Motor principles related to predicted and feed-back driven modes of control are not equally emergent during the visual perception of action
D. Méary and J.P. Orliaguet
Laboratoire de Psychologie Expérimentale, CNRS UMR 5105 Université Pierre Mendès-France, 38040 Grenoble Cedex 9, France.
Recent results from neuroimaging studies indicate that cortical areas, part of the motor system, are activated during the visual perception of action (Grossman et al, 2000). Along this line, it is assumed that the involvement of motor areas may well underline action imagination, recognition, and imitation (Fadiga et al 2000, Jeannerod 2001). Likewise, results from behavioural studies demonstrate that knowledge about motor principles emerges when observers perceive or judge dynamic stimuli representing human motion (Viviani, 2002).
In line with this experimental evidence, it was hypothesized that participants’ judgement about end-point velocity during the visual perception of target directed movement (i.e. pointing) and expressive movement (i.e. writing) would reveal a perceptual tendency toward Fitts law (pointing) and isochrony (writing). Seemingly at odds with this hypothesis, we expected a difference between perceived and measured movement times in the case of pointing movement toward targets of different size. The rationale for this prediction is that different modes of control relying on prediction and sensory feedback to different extents are involved during pointing movement. Indeed, visual feedback crucially determines the performance during the final phase of target directed motion when target size is small. Supposedly, it is of lower relevance when implementing pointing movement toward large target size, or writing movement.
The results confirm the presence of a perceptual isochrony for writing and of a linear relation between movement time and pointing distance. As expected, the participants' choice of correct velocity was independent of target width. Measure of coherence in the participants' answers also showed significant results. It is suggested that only predicted components of action could emerge during visual perception. Aspects of control involving the use of visual feedback would fail to influence perceptual judgement. These results are discussed with reference to recent hypotheses concerning the common mechanisms involved in perception and action.References
Fadiga L, Fogassi L, Gallese V, Rizzolatti G, 2000 “Visuomotor neurons : ambiguity of the discharge or "motor" perception ?” International Journal of Psychophysiology 35 165-177
Grossman E, Donnelly M, Price R, Pickens D, Morgan V, Neighbor G, Blake R, 2000 “Brain areas involved in perception of biological motion” Journal of Cognitive Neuroscience 12 711-720
Jeannerod M, 2001 “Neural simulation of action: a unifying mechanism for motor cognition” Neuroimage 14 S103-9.
Viviani P, 2002 “Motor competence in the perception of dynamic events: a tutorial” in Attention and performance XIX: Common mechanisms in perception and action Eds W Prinz, B Hommel (NY: Oxford University Press) pp. 406-442
People and object drawings activate different visual areas
C. Paulos1, C.Scheiber2, L. Manning3 and C. Bonnet1
1 Laboratoire des Systèmes BioMécaniques et Cognitifs (IMFS, UMR7507)
2 Institut de Physique Biologique (UMR 7004)
3Laboratoire de Neurosciences Comportementales et Cognitives (UMR 7521) Université Louis Pasteur, Strasbourg
Neuropsychological data reported discrete cortical category-specific representations in the posterior ventral occipital and temporal cortex. Focal brain-damaged patients illustrated the functional dissociations between representations of man-made objects and living items. Neuroimaging studies demonstrated that different categories of meaningful visual stimuli (chairs, buildings, faces, animals etc.) activate different areas in the visual brain.
The present study investigates whether there are different brain activation areas when visually processing man-made-objects and human silhouettes. The work was conducted through a functional magnetic resonance imaging (fMRI, box-car design) in 10 healthy human subjects. Stimuli were line drawings belonging to four categories: people in movement, people at rest, objects suggesting an action (e.g a ball) and objects not suggesting an action (e.g. a clock). Ten exemplars of each category were randomly presented two times in each block. Subjects (n=10) were made aware of the characteristics of each category. Passive viewing was used.
Data were analyzed with SPM99. The four categories were regrouped in two (object and people). Each category was contrasted with the hemodynamic response for the rest field (fixation cross). Thereafter, brain maps were obtained comparing the regions activated by both categories, uniquely by images of objects or uniquely activated by images of people. In spite of individual differences, regions of common activation are found bilaterally in the more occipital regions, region of pure activations by objects were bilateral and centred in the medial fusiform gyrus, while pure activations by people were also bilateral but centred in the lateral fusiform gyrus. These results are in agreement with those obtained by Ishai et al. (2000). Pure bilateral activations by people were also obtained in the motion area V5 (hMT+).
On-line and off-line visuomotor guidance of the hand: Insights from bilateral dorsal stream lesions in humans
L. Pisella3, H. Gréa3, A. Vighetto3, H.C. Dijkerman2, R.D. McIntosh1, M. Desmurget3, C. Michel3, C. Tilikete3, A.D. Milner1, C. Prablanc3 and Y. Rossetti3
1Department of Psychology, University of Durham, England.
2Helmholtz Institute, University of Utrecht, The Netherlands.
3Espace et Action INSERM Unit 534, Bron, France.
Following damage to the superior posterior parietal cortex, optic ataxic patients make gross spatial errors when pointing towards targets and fail to scale their grip aperture to the size of objects when reaching out to grasp them.
A recent study (Milner et al. 1999) has shown that when the bilateral optic ataxic patient AT was asked to delay her pointing response, her accuracy improved considerably. It was hypothesised that the delay enabled her to form a cognitive perceptual representation of the object that could be used for action. Consistently, other investigations have distinguished between fast-automatic and slow-intentional visuomotor guidance (Pisella et al. 2000). The bilateral optic ataxic patient IG showed a lack of on-line visuomotor guidance, suggesting that the “automatic pilot” of the hand relies on the superior posterior parietal cortex (Pisella et al. 2000, Gréa et al. 2002). The behaviour of IG and AT was further assessed in delayed pointing and grasping tasks, where the object was presented once, then hidden for 5s before being presented again to be grasped. During the delay, the first object could be replaced by a second object of different size or at a different location. In these conflict situations between the current and the represented objects, the results clearly confirmed that visuomotor guidance in optic ataxia is not elaborated in real time but instead based on a cognitive representation of the first object evoked off-line (Milner et al. 2001).
Finally, Patient I.G. was tested for prism adaptation, another scale of off-line visuo-motor modification. She showed normal aftereffects and high inter-manual transfer. This result suggests that on-line visuo-motor guidance and visuo-motor adaptation are dissociated processes (Pisella et al. Submitted).
Impaired frontal scalp activation during a reaction time task in schizophrenia, an ERP study.
A. Posada and N. Franck
Institut des Sciences Cognitives, Bron, France.
Several studies have shown that schizophrenia is characterised by cognitive impairments associated with frontal lobe dysfunction such as executive functions, strategic reasoning or anticipation. Using event-related potentials (ERPs), we assessed the brain's electrical activity in a sample of patients with schizophrenia (n=12) and a control group (n=14) during a reaction time task requiring the use of a rule in order to give correct response. ERP waves were compared with those elicited in a similar task based on a direct sensory association. In the control group, ERP analysis showed a negative wave moving from the posterior to the anterior regions of the scalp in the latency range of 250 to 400 ms before setting up at frontal scalp region. This negative wave was sensitive to the task and showed a higher negativity during the rule operation. In schizophrenic patients, this wave was totally absent in the rule and the control tasks. Scalp current density analysis showed that the negative wave had a near scalp generator which suggests that ERP abnormalities observed in schizophrenic patients could be attributed to a dorsolateral prefrontal deficit. Our experimental results support the hypothesis of a posterior-frontal disconnection in schizophrenia patients.
Hierarchical processing of sensory informations : effect of the optokinetic stimulations.
P. Revol and Y. Rossetti
INSERM U534, Espace et Action, 16 avenue Doyen Lépine, 69500 BRON
Sensory inputs are integrated and merged permanently in order to perform accurate oriented behaviours. A moving background is know to elicit a reflexive optokinetic nystagmus. However, when the fixation of a visual stimulus is imposed, the nystagmus was suppressed and a perception of motion of the fixated stimulus occurred in the opposite direction of the optokinetic stimulation. Moreover, apparent motion of a sound source can be induced by such stimulation suggesting a cross-modal interactions between the visual and auditory systems. Is there such a processing between visual and proprioceptive systems?
In this study, subjects estimated speed and direction of the apparent movement of a target under optokinetic stimulation. This visual stimulation was performed with an homogenous field moving at high or low speed even on the right or on the left.
Subjects performed three different experimental conditions. In the first one the target was centred on a shelf (visual target) which hid subject’s limb and in the other two it was fixed on subject’s right index finger (finger target). In these two last conditions, subjects saw their finger through a window centred on the shelf. The index position was maintained either passively or actively.
When the optokinetic stimulation occurred, each subject perceived a clear-cut illusion of movement of the target in the opposite direction of the moving background. An ANOVA reveals that only main effect of background speed is significant while the other main effects and interactions are not significant. This result shows that a type of target (visual target or finger target) is not effective against the magnitude of induced movement and it is depended only on the speed of background. Besides, its suggests a hierarchical processing for sensory information, with a leadership for visual information.
Hyperschematia a disorder due to right brain damage
G. Rode1, G. Vallar 2, C. Michel1, Y. Rossetti 1, B. Schott and D. Boisson 1
1Hospices Civils de Lyon and I.n.s.e.r.m. U534 Espace et Action.
2Dipartimento di Psicologia, Università di Milano-Bicocca, Milan, Italy.
Patients with unilateral brain damage may show a distortion of size perception, underestimating the horizontal extent of contralesional objects (hemimicropsia or horizontal dysmetropsia). The deficit is independent of visual field deficits and spatial unilateral neglect, resulting from a specific disorder in object perception (Cohen, Gray, Meyrignac, Dehaene, & Degos, 1994; Ferber & Karnath, 2001; Frassinetti, Nichelli, & di Pellegrino, 1999; Milner & Harvey, 1995).
We report two right-brain-damaged patients with damage extended to basal ganglia, parietal and frontal white matter sparing the striate occipital visual cortex. Both patients showed a distortion of horizontal size perception, with an underestimation of left-sided objects, associated to a size distortion in drawing, with an enlargement of the left side of objects, in copying and drawing from memory tasks. Both patients showed neither spatial unilateral neglect nor visual field deficits.
Methods
Subjects. Two right-handed right-brain damaged patients, and six matched healthy control-subjects were examined.
Drawing. Patients were asked to draw a daisy in different conditions: i) drawing from memory in a baseline standard-view condition; ii) copying; iii) drawing from memory with eyes closed; iv) drawing from memory while preventing direct view by an epidiascope, in order to disentangle the putative perceptual vs. premotor factors underlying the left-sided distortion in drawing.
Perceptual matching task. Twenty-five pairs of black rectangles were presented in a pseudorandom series, one in the left and one in the right side of space. Patients were required to choose the longer rectangle.
Reproduction of horizontal line extent (Bisiach, Pizzamiglio, Nico, & Antonucci, 1996). The subjects’ task was to reproduce the horizontal extent of black lines (4,6,8 cm), either rightwards or leftwards.
Results
Patients i) showed a systematical enlargement of the left half of objects drawn from memory or copied; this left-sided distortion was also observed when they draw with eyes closed, or with a mirror vision of the display; ii) in a perceptual matching task, they underestimated the horizontal extent of left-sided objects; iii) in a motor matching task, patients produced a line longer than the model on the left side, but not on the right. The performance of control subjects was symmetrical in all three tasks.
Discussion
Two right-brain-damaged patients exhibited a distortion of left visual space in both the perceptual and the representational domains. This deficit can not be explained in terms of a general spatial unilateral neglect disorder, reflecting instead a specific misrepresentation of extra-personal space, possibly involving a contralesional relaxation of the spatial medium. The existence of such a spatial disorder was first suggested by Pierre Bonnier (1905; review in Vallar & Papagno, in press), with reference to the spatial representation of the body, and termed ‘hyperschematia’. The relationships of the present findings with the syndrome of spatial unilateral neglect (Bisiach & Vallar, 2000) are discussed.References
Bisiach, E., Pizzamiglio, L., Nico, D., & Antonucci, G. (1996). Beyond unilateral neglect. Brain, 119, 851-857.
Bisiach, E., & Vallar, G. (2000). Unilateral neglect in humans. In F. Boller, J. Grafman & G. Rizzolatti (Eds.), Handbook of neuropsychology (2 ed., Vol. 1, pp. 459-502). Amsterdam: Elsevier Science, B.V.
Bonnier, P. (1905). L'aschématie. Revue Neurologique (Paris), 13, 605-609.
Cohen, L., Gray, F., Meyrignac, C., Dehaene, S., & Degos, J. D. (1994). Selective deficit of visual size perception: two cases of hemimicropsia. Journal of Neurology, Neurosurgery, and Psychiatry, 57, 73-78.
Ferber, S., & Karnath, H.-O. (2001). Size perception in hemianopia and neglect. Brain, 124, 527-536.
Frassinetti, F., Nichelli, P., & di Pellegrino, G. (1999). Selective horizontal dysmetropsia following prestriate lesion. Brain, 122, 339-350.
Milner, A. D., & Harvey, M. (1995). Distortion of size perception in visuospatial neglect. Current Biology, 5, 85-89.
Vallar, G., & Papagno, C. (in press). Pierre Bonnier’s (1905) cases of bodily “aschématie”. In C. Code, C.-W. Wallesch, Y. Joanette & A. R. Lecours (Eds.), Classic cases in neuropsychology (Vol. 2). Hove, East Sussex: Psychology Press.
Control of the head in the frontal plane during gravitoinertial changes:
Role of peripheral visual informationF. Sarès, J.-M. Prieur, C. Bourdin, G. M. Gauthier, J. Blouin and J.-L. Vercher
UMR 6152 Mouvement & Perception, CNRS & Université de la Méditerranée, Faculté des Sciences du Sport, Marseille
Accurate control of head position is a prerequisite for visual target localization and reaching. In a previous experiment, we showed that during gravitoinertial changes in darkness, presenting subjects with a peripheral visual reference frame induced non-perceived modifications in head orientation. Other studies showed that proprioceptive feedback and/or force production control are altered during gravitoinertial changes (Bock & Cheung, 1998).
The aim of the present study was to determine the accuracy with which human subjects control their head position and how this control is influenced by peripheral visual references.
Seated subjects, firmly attached to a chair with the head free to move were either in total darkness or facing a chair-fixed peripheral visual reference frame. After a learning period, subjects were instructed to reproduce learned head angles in the frontal plane before, during, and after off-vertical axis rotation (OVAR). The same test was carried out on a tilted chair without rotation. The angle of the gravitoinertial vector during centrifugation and the tilt angle of the immobile chair were identical (17.5°). Head orientation relative to the fixed or the rotating chair was recorded by means of an electromagnetic sensor. Requested head position angles in the different conditions were compared.
On the rotating chair during OVAR as on the immobile tilted chair, the accuracy of head position control decreased as compared to that before and after centrifugation. This may result from the misalignment between body longitudinal axis and resulting gravitoinertial force. Moreover, when the chair-fixed visual reference frame was available the head position error was not reduced but the within-subject variability was. Self-motion-induced retinal information may contribute to this reduction of variability whereas the position error may result from an ocular counter-torsion induced by head tilt or gravitoinertial changes.Different control of the amplitude and the direction of ongoing reaching movements after unperceived changes of visual feedback on hand position
F. Sarlegna, J. Blouin, J.-P. Bresciani, C. Bourdin, J.-L. Vercher and G. M. Gauthier
UMR Mouvement & Perception, Marseille, France
We tested how human subjects (N=6) use visual feedback of hand position when reaching for visual targets. In a first experiment, when the target (36cm from hand starting position) was lit in the dark, participants produced simultaneous eye and arm movements toward it. In some trials, the pointer LED indicating hand position was displaced 4.5cm either leftward (Fig. 1) or rightward at peak velocity of the saccade (i.e. LED displacements were not consciously detected).
If subjects control their ongoing movements processing visual feedback, they would have to increase/decrease movement amplitude by 4.5cm to bring the LED onto the target. Subjects increased movement amplitude by only 1.3cm when the LED was displaced leftward. First significant corrections appeared 470ms after movement onset (mean MT=503ms). When the LED was displaced rightward, subjects brought their actual hand to the target rather than the pointer LED that overshot the target by 4.5cm.
In a second experiment, subjects (N=6) had to control only movement direction. In some trials, pointer LED was shifted 4cm (4° at target level) either rightward or leftward (Fig. 2) during the saccade. Subjects brought the LED under the target rather than their actual hand. When the pointer LED was displaced 4.0° leftward, movement was deviated by 3.5° rightward. When the LED was displaced 4° rightward, movement was deviated by 2.8° leftward. Modifications of movement direction appeared earlier than movement amplitude modifications observed in the previous experiment. Then visual feedback of the hand was processed more efficiently in the direction task than in the amplitude task, supporting the hypothesis of separate control of direction and amplitude of reachings.
A comparison of visual and tactile space exploration in unilateral neglect
I. Schindler1, S. Clavagnier², H.-O. Karnath3, L. Derex4 and M.-T. Perenin²
1Department of Psychology, University of Durham, South Road, Durham DH13LE, UK
²INSERM U 371"Cerveau et Vision", Bron, France
3Department of Cognitive Neurology, University of Tübingen, Tübingen, Germany
4 Hôpital Neurologique, Bron, France
Studies investigating independently visual and tactile exploration found for each modality a pronounced rightward shift of the distribution of eye/hand movements in neglect but a symmetrically distributed search pattern of control patients and normal subjects along the body midsagittal plane. These findings have been interpreted in favour of an ipsilesionally deviated supramodal reference system in neglect.
Hence the aim of the present study was to determine within an identical subject sample/search space if a) patients showing visual neglect show also neglect in the tactile modality and if b) the severity of neglect in one modality can predict neglect in the other.
Thirty subjects divided in three groups of ten (neglect, RBD controls, normal subjects) participated in the study. They were presented with geometrical objects mounted on a large semicircular table and which served as targets for both modalities. Subjects were instructed to name or touch and identify as many objects as possible. Hand and trunk movements were not restricted. Based on video camera recordings the number of named/touched objectswas analysed.
Between group analyses revealed dramatic rightward shifts of the exploration distributions in neglect patients including more restricted distribution ranges as compared to both control groups regardless of the modality. For the neglect groupthe deviation of the median of the distribution was even more pronounced in the visual as compared to the tactile condition. Additionally significant correlations between the visual and tactile condition could be found for the neglect group but not for the control groups.
The findings of a generalized rightward shift of orientational behaviour across the visual and tactile modalities and their systematic relationship indicates that the critical lesion leading to spatial neglect affects the processing within a supramodal reference system.
Anticipatory postural adjustments during a bimanual load-lifting task: kinematic and electromyographic evidence of impairment in autism.
C. Schmitz (1), C. Assaiante (1), J. Martineau (2) and C. Barthélémy (2)
(1) CNRS, INPC, MARSEILLE; (2) INSERM, U316, TOURS
Anticipation is one of the motor functions that enable children as they grow up to exercise an action that is integrated in their sensorial and relational environment. This implies that information given by the environment are properly translated into internal representations. Our aim was to investigate the building processes of anticipatory function in autistic children. We used a bimanual load-lifting task to assess the use of anticipatory postural adjustments (APA) aiming at stabilizing the postural forearm when a load placed on it is voluntary lifted by the subject.
8 autistic children aged 6 to 10.6, and 16 control children aged 4.8 to 8 participated in this experiment. We compared two situations of unloading: one imposed to the subject by the release of a load suspended bellow the child wrist by the experimenter, and the other one consisting on the voluntary lifting of the load placed upon the wrist by the subject himself. A gauge constrain indicated the onset and the duration of unloading. We used elbow angle measurement to assess for postural stabilisation, and surface EMG.
Temporal parameters resulting from the measurement of the latencies for the kinematic, as well as the latencies for the muscular events underlying postural stabilisation, indicated that the mode of control preferentially used by those children is retroactive and based on the use of proprioceptive information during movement. This suggests that anticipatory mode of control is affected in autistic children. Nevertheless, by slowing down unloading, retroactive control substituted for proactive control ensures the performance to be maintained during voluntary unloading. Whether the building or the correct use of internal representations is affected in autistic children because of their disorders in the integration of environmental constraints is still an open question.
Cortical activity in a precision grip-lift task with predictable versus unpredictable weight changes
C. Schmitz, P. Jenmalm, H. H. Ehrsson and H. Forssberg
Neuropediatric, Karolinska Inst., Stockholm, Sweden
To properly lift an object, we program in advance the fingertip forces applied to the object, on the basis of a memory representation of its weight. However, it happens that the program is erroneous: for instance when we lift a package of milk which is lighter than predicted. Thus, one critical aspect of motor control in dexterous manipulation is the use of somato-sensory information to update the memory representation of the object. Here we aimed at characterizing the active areas involved in the updating when subjects are lifting an object with either unpredictable weight changes, or predictive ones. Functional MRI was performed on a 1.5 T scanner. The task was to lift a load using a precision grip. The weight of the load was either 200g (light weight), or 600g (heavy weight). In the target condition, changes in weight were made in an unpredictable way. Three conditions were used as controls. During the two first ones, subjects were asked to do repetitive lifts with a constant weight, either light, or heavy. During the third condition, regular changes occurred between every lift. Images were analysed using SPM99. A pilot study was performed with 6 right-handed subjects. When comparing brain activity related to unpredictable changes versus repetitive lifts with constant weight, inferior posterior parietal areas and the inferior frontal gyrus were bilaterally activated. When comparing the unpredictable changes in weight with the regular changes condition, a clear bilateral activation appeared on the inferior frontal gyrus corresponding to Broca's area, and its right homologue. These results suggest that the updating of the weight of an object involves the bilateral non-primary fronto-parietal areas. These areas could participate in the use of somato-sensory information for the on-line update of the object representation, up-dating process of the program in itself, or related to the erroneous correction of the grip forces.
The influence of hand posture on corticospinal excitability during motor imagery: a TMS study.
C.D. Vargas 1,2; E. Olivier 1,3, L. Craighero 4; L. Fadiga 4, J.R. Duhamel 1 and A. Sirigu 1
1ISC, CNRS, France; 2IBCCfO, UFRJ, Brazil; 3University of Louvain, Belgium and 4University of Ferrara, Italy.
We investigated how congruent and incongruent postural signals influence corticospinal excitability during motor imagery. Six healthy subjects were asked either to imagine a hand movement (joining the tips of the thumb and little finger) or to simply rest, while maintaining one of the two following hand postures: P1) little finger, index and thumb extended, the remaining fingers flexed; P2) index and thumb extended, other fingers flexed. Motor potentials (MEPs) evoked by focal (figure-of-eight coil) transcranial magnetic stimulation (TMS) of the left motor cortex were recorded from the right opponens pollicis muscle during both motor imagery (MI) and rest (R) conditions. MEPs area were computed from rectified EMG recordings and used to calculate an excitability index (MI-R/ MI+R) for each of the postures. Subjects then rated on a 0-5 scale how easily they imagined the movement while maintaining P1 and P2. We also recorded subjects' imagined and actual movement time to quantify how accurately motor imagery predicted movement execution.
All subjects rated the imagined finger opposition movement as easier to perform with the hand adopting posture P1 than P2 (p<0.01) and the correlation between motor imagery and movement execution was higher for P1 than for P2 (p<0.01). TMS data analysis showed that MEP excitability index was also higher for P1 than for P2 (p<0.001). Basal EMG excitability index obtained at the temporal window of 200 ms before the TMS pulse was however similar for both postures (P<0,98). Thus, we conclude that corticospinal excitability is highest when the actual hand posture and the corresponding imagined movement are coincident.
Planning and action knowledge in schizophrenic patients: Comparisons with patients with frontal lobe damage.
T. Zalla, A. Posada, N. Franck, N. Georgieff, and A. Sirigu
Institut des Sciences Cognitives, BRON, France
A line of research has hypothesised that a frontal cortical dysfunction is implicated in the pathophysiology of schizophrenia and documented that schizophrenic patients are impaired in executive functions. By using a set of multicomponent tasks which segregated various information processes used in action planning, we assessed the ability to generate action knowledge and elaborate a plan of action in a group of seventeen schizophrenic patients. Their performance was compared with those of a sample of ten patients with frontal lobe lesions and of sixteen normal control subjects. The study consisted in two experiments: an action generation task and an action ordering task. Schizophrenic patients were found to perform significantly worse than normal subjects and frontal lobe patients in the semantic retrieval of action knowledge. Similarly to frontal lobe lesions patients, they encountered difficulties in action sequencing and in setting priorities among the action events with regard to the stated goal. Interestingly, schizophrenic patients' impairment in organising a plan of action increased when distractors or novel situations were presented. These findings suggest that schizophrenic patients are unable to apply an optimal planning strategy.
Study of the dynamical acuity through a sensory substitution device in blindfolded subjects
M. Ziat, O. Gapenne, C. Lenay and J. Stewart
UTC – Costech
The main goal of our research is to understand the perception as a constitutive process in reference to the action strategies produced by a subject. In the present study, our aim is to measure the dynamic perceptive acuity of blindfolded subjects using a sensory substitution device. The dynamic acuity is defined as the discriminative ability during rapid or normal displacements of the observed object or of the observing subject (as in our case).
The experimental device couples the pen of a graphics tablet with tactile sensory stimulators. A virtual receptor field linked by means of an equipped computer to the pen, is responsible for the stimulator activation. These sensors are square matrices composed of 16 elementary receptive fields, each field may contain 1 pixel (M1), 4 pixels (M2), 9 pixels (M3) or 16 pixels (M4). The matrix’s orientation is constant i.e. independent of the orientation of the pen. When the subject touches a virtual form, she/he is informed by a tactile feedback. The tactile stimulation is a static mechanical one (non electrotactile or vibrotactile) produced by a dynamic control of 2 electronic Braille cells (2X8 points of stimulation). The mechanical components stimulate the pad of the left index tip (all the subjects are right-handed). The finger is kept on the Braille cells during the experiment. The “perceptive” trajectory produced by the subjects with the stylus is recorded.
Sixteen adult subjects (aged between 21 and 32 years) were tested with the tumbling “E” test with seven different sizes (8 trials by size). After each trial (duration 60s), they had to indicate the orientation of the E (one among four possibilities). We divided the subjects into four groups basing on the type of the matrix.
Results showed that subjects had adopted a relevant spatio-temporal strategy, which assure a good performance in the recognition of the orientation (vertical or horizontal) of the three parallel bars of the tumbling “E” even for very small sizes of “E”. Moreover, we noticed that this performance fell down rapidly (swiftly) when the size of an elementary receptive field exceeded the size of a bar of the E. Evidently this point is considered as the threshold of the performance.
* This research was supported by A.C.I Cognitique
33 (0)4 37 91 12 12 
33 (0)4 37 91 12 10 
web@isc.cnrs.fr
