Simulation Modifies Prehension: Evidence for a Conjoined Representation of the Graspable Features of an Object and the Action of Grasping It

et al (2007) Simulation Modifies Prehension: Evidence for a Conjoined Representation of the Graspable Features of an Object and the Action of Grasping It. PLoS ONE 2(3): e311. doi:10.1371/journal.pone.0000311 Simulation Modifies Prehension: Evidence for a Conjoined Representation of the Graspable Features of an Object and the Action of Grasping It Victor Frak 0 1 Isabelle Croteau 0 1 Daniel Bourbonnais 0 1 Christian Duval 0 1 Cyril Duclos 0 1 Henri Cohen 0 1 0 Academic Editor: Chris Miall, University of Birmingham , United Kingdom 1 1 De partement de kinanthropologie, Universite du Que bec a` Montre al , Montreal, Quebec , Canada , 2 Institut de Re adaptation de Montre al, Centre de recherche interdisciplinaire en re adaptation du Montre al me tropolitain, Universite de Montre al , Montreal, Quebec , Canada , 3 Psychology and Cognitive Neuroscience Laboratory (CNRS-Paris Descartes), Quebec Memory and Motor Skill Disorders Research Centre, Clinique Sainte-Anne , Que bec City, Quebec , Canada Movement formulas, engrams, kinesthetic images and internal models of the body in action are notions derived mostly from clinical observations of brain-damaged subjects. They also suggest that the prehensile geometry of an object is integrated in the neural circuits and includes the object's graspable characteristics as well as its semantic properties. In order to determine whether there is a conjoined representation of the graspable characteristics of an object in relation to the actual grasping, it is necessary to separate the graspable (low-level) from the semantic (high-level) properties of the object. Right-handed subjects were asked to grasp and lift a smooth 300-g cylinder with one hand, before and after judging the level of difficulty of a ''grasping for pouring'' action, involving a smaller cylinder and using the opposite hand. The results showed that simulated grasps with the right hand exert a direct influence on actual motor acts with the left hand. These observations add to the evidence that there is a conjoined representation of the graspable characteristics of the object and the biomechanical constraints of the arm. - INTRODUCTION Grasping is a kind of active contact with objects in the environment. This action requires a number of visuomotor transformations to code for the intrinsic and extrinsic properties of the objects to be grasped (i.e., shape and location, respectively). The geometric attributes of the object will trigger the finger grasp, while its semantic properties will determine functional interactions. These attributes and properties are not mutually exclusive and the pragmatic and semantic modes of operation with objects interact. The distinction between these properties has been made evident in lesion studies in humans [1]. These observations suggest that actions have a central origin and that kinesthetic images formed from sensory clues are stored in the motor cortex [2][][][5]. While the superior parietal lobule is involved in the automatic control of visually guided actions, the inferior parietal lobule (particularly on the left side) is concerned with the planning of actions and involves the retrieval of complex representations thought to be produced in that structure [6]. The parietal areas, together with the premotor cortex [7], account for so-called pragmatic representations. The relevance of the pragmatic aspects of prehension in normal behavior is an important question. Interestingly, Sakata et al. [8] have described neuronal discharges in the monkey in response to both passive and active observation of objects with graspable shapes (e.g., cylinder, cube). This neuronal activity has also been taken as evidence of the anatomofunctional substrate of a pragmatic theory of grasping in humans [9]. It remains to be determined whether this relationship between the internal representations of the graspable characteristics of the object and of grasping that object is a property seen in normal subjects. Orientation grasping [10] is the appropriate paradigm to investigate this question. In an earlier study, we showed that precision grasping with one hand influences grasping orientation, in an anticlockwise manner, with the other hand [11]. It is likely that this result reflects the existence of a functional engram combining the graspable characteristics of the cylinder and the grasp orientation. Grasp orientation, defined by the opposition axis (OA), represents the effector of the movement and is a main parameter to control for when completing a grasp [12]. Paulignan et al. [13] analyzed the OA during the grasping of cylinders of various sizes and weights placed in different locations. They found that the OA orientation was constant with respect to an egocentric frame of reference for all conditions. Thus, it is reasonable to think that in these circumstances the OA was computed from representational grasping coordinates. Nevertheless, in view of the fact that perception and action go hand in hand in motor activities [14], it is difficult to dissociate somesthetic afferences from grasp representation when healthy subjects produce a real grasp. In order to determine whether the graspable characteristics of an object and the associated grasping orientation have a conjoined representation, it is necessary to separate the objects geometric properties from its semantic properties. In this study, we asked subjects to grasp and lift a 300-g smooth cylinder with one hand, before and after judging the feasibility of a grasping for pouring action with a considerably smaller cylinder, using the other hand. What the actual grasp and the simulated grasp share in common is the graspable characteristics of the object. METHODS Participants Twenty-one healthy right-handed volunteers participated in the experiment (age range between 21 and 52 years, mean = 26.6 years; 5 women, 16 men). Handedness was assessed using the Edinburgh Handedness Inventory [15]. Only subjects scoring a laterality quotient of 100 were included in the study. All participants were recruited and tested in accordance with the ethical considerations set out by the Centre for Interdisciplinary Research in Rehabilitation of Montreals ethics committee. Subjects were initially instructed about the methods used in the study; the purpose of the study was revealed once the experiment was over. Procedure Participants were seated in front of a table; the initial position of the left hand was 13 cm left of the sagittal axis, while the right hand was 13 cm to the right. Participants were asked to perform 10 consecutive real grasps with one hand before and after 400 simulated grasps with the other hand. For the real task condition, they were asked to reach for, grasp, lift and return to its original position a smooth 300-g resin cylinder (6 cm in diameter, 10 cm high) placed in the center of the table at a distance of 32 cm from the body plane, using a precision grip formed by the thumb and (...truncated)