Read DGNR04-MirrorTraining-neu.PDF text version

The neural mechanisms of mirror training

C. Dohle1,2 , K.M. Stephan2, R. Kleiser2, J. T. Valvoda1,2,4 , T. Kuhlen4, R.J. Seitz2, H.-J. Freund 2,3


Godeshöhe Neurological Rehabilitation Center, Waldstraße 2-10, 53177 Bonn, Germany, [email protected] 2 Department of Neurology, University Clinics Düsseldorf 3 Institute of Medicine, Research Center Jülich 4 Center for Computing and Communication, Aachen University, Germ any


It has been proposed that rehabilitation of hemiparesis after stroke can be improved by use of the so-called "mirror training ". During this procedure, movements of the affected limb are viewed as movements of the affected limb (figure 1). The mirror training is well known in the field of neurorehabilitation, even as its underying neural mechanisms are still unclear.





Figure 1: Principal setup of the mirror training.

We describe two imaging studies in normal subjects (reported previously) that explored the cerebral activation during mirroring of human movements. In the first study, hand movements are viewed while subjects were lying in a fMR scanner (Dohle et al. 2004). In the second study, subjects viewed movements of a 3 -D-computergraphic display of a human arm moving synchronuously to their own arm 's movements while lying in a PET scanner (Dohle et al., 2002, manuscript in preparation).

Figure 2: Results of the two activation studies . Left part:: Experiment 1 (fMR study - video image of subjects ` hand), right part: Experiment 2 (PET study -computergraphic display of human arm). A: Experimental set -up: Upper row : Performed action, lower row : visual feedback B: Cerebral activation plotted on coronar slices. In both experiments , activations are found in the hemisphere contralateral to the seen limb (Note the differences in laterality of the slices due to the different analysis programs used).


Experiment 1 (fMR, video image of human hand)

Subjects - 6 normal, right -handed subjects, mean age: 29.0 years - written informed consent, study approved by the local ethics committee Experimental procedure - subjects required to move either their right or left hand, observed via a PC-based video chain (Leutron demonstration software) - Hand is either static of moving, viewed either normal or inverted - Task: Motor sequence, scaling finger-thump opposition distance three different distances fMR scanning - pseudo-randomized protocol for either hand, lasting 10 minutes each - Main factor analysis (moving vs. static, inverted vs. normal) with BrainVoyager, normalized to the subjects' individual MRs, p < 0.01 (corrected)


(1) In normal subjects, mirroring of hand / arm movements lead to an activation of the hemisphere contralateral to the seen limb. (2) For the video image of a human hand, this activation is restricted to the primary and higher order visual areas. For the computergraphic display of a human arm, this activation is confined to the precuneal region. (3) The first result might reflect the process of drawing attention to the encoding of a visually presented body scheme. The latter one seems to correspond to the task of translating a desired coordinate in space into a


(1) The observed activation of the contralateral hemisphere by movement mirroring might explain the beneficial effect of mirror training. It needs to be confirmed, however, if the mechanism is the same in hemiparetic patients. (2) For maximum activation of the contralateral

hemisphere, the attention has to focussed on decoding the visual configuration of the human body. (3) Experiment 2 implies that lateralized activations are not only elicited by real images, but also by computer graphic displays of a human limb. This might open a new use of virtual reality techniques in the field of neuro-rehabilitation.

Experiment 2 (PET, computergrapic image of human arm)

Subjects - 10 normal, right -handed subjects, mean age: 24.4 years - written informed consent, study approved by the local ethics committee Experimental procedure - subjects being presented with a `virtual arm' (figure 2A), a computergraphic representation of a human arm - difference between right and left arm by: position of the shoulder, flexing direction of the fingers, mirroring of movement direction - virtual arm animated by means of a custom-made C++ program (Liebmann 1988), based on the C++ library WorldToolKit, running on a Dual-Head Pentium II Intergraph PC with two Wildcat graphic cards - position of subjects' own arm monitored by means of a Polhemus Fastrak electromagnetic tracking device, sampling rate at least 30 Hz - virtual arm displayed by means of a head-mounted display (V8, Virtual Research), providing a stereoscopic view of the arm - subjects always required to move their own right arm, visual display either right (RVA) or left (LVA) virtual arm - Task: Draw ,, imaginary" circles as accurate as possible PET scanning - introductory + training sessions prior to the actual experiment - scanning session: 2 x 3 sessions with 555 MBq O15-Butanol each, lasting two minutes each, spaced by 8 minute intervals - analysis with SPM 99, normalized to the subjects' individual MRs, p < 0.05 (corrected)

corresponding limb configuration.


Altschuler EL WS, Stone L, Foster C, Galasko D, Llewellyn DM, Ramachandran VS. (1999) Rehabilitation of hemiparesis after stroke with a mirror. Lancet 353:2035-2036. Dohle C, Kleiser R, Seitz RJ, Freund H-J (2004). Body scheme gates visual processing. J Neurophysiol 91: 2376-9. Dohle C, Stephan KM, Kleiser R, Valvoda JT, Tellmann L, Hefter H, et al. (2002). Movements of a right or left virtual arm synchronous to one's own arm movements cause lateralized precuneal activations . Society for Neuroscience. Orlando. Liebmann M (1998) Entwicklung, Validierung und neurophysiologisc he Interpretation der Ansteuerung einer virtuellen Darstellung eines menschlichen Armes. Diploma thesis, FH Aachen, Abt. Jülich.


Supported by grants of the Deutsche Forschungsgemeinschaft (SFB 194, A9 & A13) and the `Forschungskomission der Heinrich-HeineUniversität D üsseldorf'. Special thanks to M. Liebmann for the development of the prototype of the program as well as to O. Hosseiny and L. Tellmann for invaluable technical support during the performance of the study. The PET study was performed at the institute of medicine at the research centre J ülich (director: Prof. Dr. K. Zilles)



1 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

jsc200152 1242..1248
Exo_paper - ICORR 2007 v18.dvi
Educing Information