Postersession 14. Movement-related cortical potentials Sl19 expiratory RRP at breaking of the maximum breath-holding ses- sions at the total lung capacity level. Using dipole-tracing it was estimated that the source generator of the subtracted potential was located in the prefrontal area. These results suggest that the prefrontal cortex may act as the center of the desire to breathe. PS-14-7 1 Movement-related cortical potentials upon forward stepping -- difference by varied trigger points Kouichi Saitou 1, Yukihiko Washimi z, Nobuo Sakurai ], Yasuo Koike 1, Gen Sobue 1. 1Department of Neurology, Nagoya University School of Medicine; ~Department of Neurology, Nagoya Ekisaikai Hospital It is not well known which trigger points are suitable for obtain- ing movement-related cortical potentials (MRCPs) steadily upon forward stepping, because this paradigm consists of various move- ments. To make this issue clear, we recorded MRCPs in 10 healthy subjects when they stepped the right leg forward volitionally every 15 second from the standing position, and investigated the changes of MRCPs in terms of various trigger points. In this paradigm, we first make a weight-shift, then supporting the weight with the left leg, simultaneously elevating the right leg, and lastly make a right foot contact. Although the latency of the initial weight-shift was ex- tensively variable among the trials and/or the subjects, those of the following trigger points were less variable; the EMG onsets of the left leg, left lower leg, right leg, and right tibialis anterior muscle and the EMG increasing points of left gastrocnemius and soleus muscle owing to forward locomotion. Each MRCP obtained here shifted each other according to the latency of each trigger point. The magnitude of the gradients of each negative slope (NS') was, however, very constant against variable trigger points, We were able to record MRCP clearly and steadily, triggering at the onset of the right tibialis anterior EMG burst. We were able to evaluate the gradient of NS p as a constant value of this paradigm, but not the latency nor the amplitude measured from the trigger point. cies and may represent stages of gradual transition to high fre- quency coherence oscillations required for optimal sensorimotor integration. I PS-14-9 of cortical when Changes potentials performing visually-guided saccades T.S. Constantinidis, I. Evdokimidis, D. Liakopoulos, C. Papageorgiou. ENG Laboratory of the Neurological Clinic of Athens National University, Aeginition Hospital, Greece It is already known that cortical slow potentials prior to visually- guided saccades consist, mainly, of an increasing negativity ob- tained over Cz, which turns to positivity at about 300 ms prior to the saccade onset. These potentials represent the overall cortical activity over a quite long period of time and reflect the average cor- tical D.C. level changes just prior to the saccade onset. Although several studies have already deeled with prosaccade potentials, little is known about the EEG changes which, probably, occurs during the execution of a saccade task. We recorded the EEG (t.c. = 1-sec) from 15 healthy subjects which performed horizon- tal saccades towards randomly appearing target to the left or to the right. The time performance lasted 30 minutes and we col- lected and averaged (triggering on the saccade onset) the EEG in two different time periods. The first 10 minutes (the early po- tentials, E-) and the last ones (the late potentials, L-). Comparing the cortical potentials of the two groups, we found several differ- ences. First, the E-potentials present three areas of activation, i.e. over CFz, ipsi- and contralateral P, while the L-potentials shows a rather stable area of activation over Cz. Second, these differences occur over the time period of 100-300 ms prior to the saccade onset. Our findings suggest that at the beginning of the task the performance of the saccades requires the activation of both ante- rior and posterior-parietal cortical areas, while with the time pass the cortical activity is reduced, not in amplitude but in topography, and constricted over the central midline areas. I PS-14-8 I of functional cortical Modulation interconnections during self-paced movements as revealed by event-related coherence analysis Letizia Leocani, Camilo Toro, Paolo Manganotti, Ping Zhuang, Mark Hallett. Human Motor Control Section, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, USA The sensorimotor activation during movements was studied using EEG Event-Related Coherence (ERC) and Event-Related Desyn- chronization/Synchronization (ERD/ERS), These techniques pro- vide information on cortical activity and functional cortical con- nections with a high degree of temporal resolution compared to other imaging techniques. Ten subjects performed self-paced right index finger abductions every 10 sec. ERD/ERS and ERC within the alpha and beta bands were computed for 29 EEG channels from -2.5 to 2.5 sec after movement onset. Both bands showed ERC increase between central and frontal scalp regions, within and across hemispheres. Alpha band coherence increased earlier and with longer duration than the beta band. In both bands, the ERC increased when ERD began and decreased when ERS be- gan. Sensorimotor activation during self-paced movements is as- sociated with a decrease in signal power and a coherence increase over sensorimotor areas. These events progress in a frequency- dependent orderly manner, developing earlier for slower frequen- I PS-1 I potentials preceding voluntary 4-1 0 Cortical elbow movement in recovered hemiparesis Jun-ichi Kitamura 1, Hiroshi Shibasaki 2. 1 Department of Rehabilitation, The Second Branch Hospital of Nippon Medical School, Kawasaki, Kanagawa 211, Japan; 2Department of Brain Pathophysiology, Kyoto University School of Medicine, Shogoin, Sakyo, Kyoto 606, Japan Scalp-recorded movement-related cortical potentials (MRCPs) preceding voluntary elbow flexion were studied in two hemiparetic patients 8 months after the stroke. Both cases had unilateral lentic- ular and internal capsular vascular lesion. Their motor recovery stage of the paretic arm was in "synergistic" movement. NS r (Neg- ative Slope) preceding voluntary unilateral elbow flexion of the intact arm increased its amplitude at the contralateral precentral area, while NS r preceding that of the paretic arm increased its am- plitude equally at bilateral precentral areas. It is postulated that the unilateral synergistic elbow flexion of the paretic arm is associ- ated with an increased activation of bilateral brachial areas of the sensorimotor cortex (SI-MIs), while the isolated elbow flexion of the intact arm is associated with an increased activation of mainly the contralateral SI-MI.