Abstracts of Poster Presentations / Clinical Neurophysiology 125, Supplement 1 (2014) S1–S339 S61 whereas direct responses and action potentials remained unchanged (Fig. 2). These changes were largely reversible. Conclusions: Our results indicate that in cultured neuronal networks, partial hypoxia during three or six hours causes isolated disturbances of synaptic connectivity. We speculate that in chronic mild cerebral ischemia, synaptic failure plays a role in the emergence of cognitive impairment. Reference: [1] Hofmeijer J, Stroke 2012. P44 Functional connectivity network breakdown and restoration in blindness M. Bola 1 , C. Gall 1 , C. Moewes 2 , A. Fedorov 1 , H. Hinrichs 3,4,5 , B. Sabel 1 1 University of Magdeburg, Institute of Medical Psychology, Magdeburg, Germany; 2 University of Magdeburg, Department of Computer Science, Magdeburg, Germany; 3 University of Magdeburg, Department of Neurology, Magdeburg, Germany; 4 Leibniz Institute for Neurobiology, Department of Behavioral Neurology, Magdeburg, Germany; 5 German Center for Neurodegenerative Diseases (DZNE), Magdeburg, Germany Question: Loss of vision after brain damage is thought to be caused mainly by the primary tissue loss. Little is known how the damage affects interactions in widely distributed brain networks and how this, in turn, contributes to vision loss. We now studied functional connectivity in the brain of partially blind subjects to delineate the role of wide range neuronal networks in blindness. Methods: Resting state eyes-closed EEG activity was recorded in patients with partial optic nerve damage (n=15) and uninjured controls (n=13). Power density and functional connectivity (coherence, Granger Causality) were analyzed, the latter as (i) between-areal coupling strength and (ii) individually thresholded binary graphs. Functional connectivity was then modulated by non-invasive repetitive transorbital alternating current stim- ulation for 10-days (rtACS for 40 min daily; n=7; sham, n=8) to studied how this would affect connectivity networks and perception. Results: Blind patients had lower EEG spectral power (p=0.005) and decreased short- (p=0.015) and long-range (p=0.033) coherence in the high-alpha EEG band (11-14Hz) and less densely clustered coherence net- Figure 1. Coherence networks topology. The first two columns show coherence net- works graphs with black dots representing standard EEG electrode positions. All graphs contain the same number of edges (connections). The difference graphs (3rd column) display edges present in controls but missing in patients (blue) and edges missing in controls but present in patients (red). The most pronounced between groups differences occur in the alpha II band. works (p=0.025). rtACS strengthened short- and long-range coherence again which correlated with recovery of detection ability (r=0.57, p=0.035) and processing speed (r=0.56, p=0.049). Conclusions: Peripheral damage to the visual system permanently ham- pers spontaneous cortical synchronization and functional networks topol- ogy. Synchronization breakdown accompanies loss of perceptual abilities, therefore we argue that cortical activity might either hamper or facil- itate perception in patients. Based on these findings cortical functional connectivity emerges as a therapeutic target in vision restoration. P45 Event related desynchronisation/synchronisation (ERD/ERS) in the actual and imagined reaching movements M. Rakusa 1,2 , A. Belic 3 , P.P. Battaglini 4 , I. Zidar 1 , J. Zidar 1 1 Institute of Clinical Neurophysiology, Division of Neurology, University Medical Centre Ljubljana, Ljubljana, Slovenia; 2 University Medical Centre Maribor, Department of Neurology, Maribor, Slovenia; 3 University of Ljubljana, Faculty of Electrical Engineering, Ljubljana, Slovenia; 4 Università di Trieste, Dipartimento di Scienze della Vita, Trieste, Italy Question: Despite growing number of published papers, the ERD/ERS po- tentials, which give real time information on the ongoing neural activity, is still poorly understood when comparing actual vs. imagined reaching movements. Methods: Eleven subjects were reaching and imagining reaching with right hand toward the target in the right upper quadrant of touch screen. Reaction and movement times were measured for actual reaching. EEG was recordedby 64 electrodes. After off-line artefact rejection, ERD/ERS for each subject were measured in alpha and beta frequency bands for frontal, central, centro-parietal and occipital electrodes. Differences between both tasks were compared using boothstrap method. Results: Reaction and movement times were 369±65 ms and 383±78 ms respectively. In actual reaching, alpha ERD and beta ERD were significantly stronger 0.5 s and 2.5 s, respectively, before target onset. In imagined reach- ing, ERD was stronger in all electrodes around reaction time. Thereafter differences in alpha ERD become non significant for 0.5 s to 0.7 s in left frontal and central to parietal electrodes respectively. On the right side, we found significant ERD differences approximately 0.2 s after actual reaching onset. ERS rebound was stronger in both bands approximately between 1 and 2 s after actual reaching onset. Conclusion: Actual and imagined reaching activated similar cortical areas including the primary motor cortex. Stronger alpha and beta ERD across the entire bands in imagined reaching around movement onset may be due to undetermined strategy or the need to refrain the actual movement. Longer ERD during reaching and stronger ERS rebound after it may reflect muscle activation and proprioceptive feedback. LP3 Weighted time-varying network differences between cognitively intact and impaired MS patients using correlation as connectivity measure J. Van Schependom 1 , J. Gielen 1 , J. Laton 1 , M.B. D’hooghe 1,2 , J. De Keyser 1 , G. Nagels 1,2,3 1 Vrije Universiteit Brussel, Center For Neurosciences, Brussel, Belgium; 2 National MS Center Melsbroek, Neurology, Melsbroek, Belgium; 3 Université de Mons, Faculté de Psychologie et des Sciences de l’Education, Mons, Belgium Question: Multiple Sclerosis is the most frequently encountered neurode- generative disease in young adults (Inglese, 2006). Although approximately half of the MS patients suffer deteriorated cognitive abilities, the underpin- nings this cognitive deterioration are poorly understood and no sufficient biomarkers have yet been developed (Schoonheim et al., 2013). Methods: We have included 305 MS patients, 162 of which were denoted as cognitively impaired based on extensive neuropsychological testing. All patients underwent an EEG oddball experiment. Based on the aver- aged response to target stimuli, we constructed weighted networks using correlation as connectivity measure. Results: For most edges, weights were significantly different between both patient groups, with the most significant edges in centro-parietal region. Next, we applied a sliding time window approach to track the origins of these differences to a particular time window of the cognitive process. We