CC BY
34
Volga Neuroscience School 2016 Astroglial control of rhythm genesis in the brain
Is There a Connection between BCI Performance and the Neurophysiological Effects of Motor Imagery?
OM&P
L.V. Yakovlev1,2 *,A.N. Vasilyev1,3, S.P. Liburkina1,3 and A.Y. Kaplan1'2,3
1 Lomonosov Moscow State University, Moscow, Russian Federation;
2 Lobachevsky State University of Nizhny Novgorod, Nizhny Novgorod, Russian Federation;
3 Pirogov National Russian Medical University, Moscow, Russian Federation. * Presenting e-mail: [email protected]
Abstract. Motor imagery (MI) is a mental rehearsal of person's own body parts movement and thiscan be helpful as a training technique for neurorehabilitation [2, 3]. MI is known to promote patterns of the event-related desynchronization (ERD) of mu-rhythm found in EEG over the sensorimotor areas of the human cortex. Brain-computer interfaces (BCI) introduce ways to decode those specific EEG patterns into the control signal for external devices providing direct communication channel between the brain and the outer world [4]. Initially it seems that BCI approach promises additional benefit for the motor imagery practice by adding a feedback and therefore helping the subject to monitor the imagery quality. On the other hand, BCI-training favors subjects with a stronger mu-rhythm response which is not considered to be an indicator of the motor imagery effort quality. Due to the weak or absent EEG response during MI, a substantial portion of the population is characterized as «BCI-illiterate» or «inefficient» indicating poor performance in a brain-computer interface circuit [1], and therefore those people are being eliminated from such activity. The aim of our research is to clarify the connection between user's BCI performance and the neurophysiological effects of motor imagery.
In the present study, we used EEG and TMS (transcranial magnetic stimulation) to quantify and compare two physiological responses during MI in subjects with different levels of BCI performance (ranging from good to poor). The amplitude of TMS-induced motor evoked potential (MEP) in a resting muscle was used as a measurement of the excitability level of M1 cortex (fig. 1A), and 64-channel EEG was used to measure anERD during MI. Both measurements were related to the referential «visual attention» state. All participants underwent at least five sessions of MI-training containing both feedback and non-feedback runs. TMS assessment was conducted once during the last experimental session. Subjects were trained to perform motor imagery of sequential finger flexion and extension. During TMS the electromyogram from two forearm muscles (EDC and FDS) muscles was recorded.
We have found that increased M1 cortex excitability during MI was quantitatively uncorrelated with mu-ERDlevel (Spearman's r=0.23) and BCI performance (Spearman's r=0.28). Our results could be explained by a contribution of two reasons. At first, mu-rhythm power decreases during MI and thereby its modulation range is limited by resting-state power value which varies both in the general population and within subjects on different experimental days. That is why the subjects whose resting mu-rhythm power is low generally demonstrate poor BCI-performance. On the contrary, M1 excitability level increases during motor imagery (fig 1A,1B) and therefore has greater measurable range.
Fig.1. A - Example of MEPs from EDC and FDS muscles during MI and "visual task". B -Amplitude increase of MEPs in FDC during finger flexion imagery for subjects of three BCI performance groups.
Secondly, mu-rhythm appears to be an indicator of the general inhibitory input into the vast cortical areas, whereas M1 excitability reflects the state of the local neuron group corresponding to a discrete muscle [5]. Based on prior knowledge, MI should promote excitability of local cortical pathways involved in imagined movement, but not necessarily alter general inhibitory output of thalamocortical circuits.
Our results suggest that if MI practice is considered to be beneficial in regard of training discrete motor cortex pathways, poor BCI performance should not discourage users from mental exercises. EEG control should be accompanied by other cortex excitability measurements such as TMS to provide a more comprehensive picture.
Volga Neuroscience School 2016 Astroglial control of rhythm genesis in the brain Acknowledgements
This study was partially supported by funding from the Skolkovo Foundation (project 1110034) and from the Russian Science Foundation (15-19-20053).
References
1. B. Allison andC. Neuper, Brain-computer interfaces, 2010, 35-54.
2. P. Jackson, M. Lafleur, F. Malouin, C. Richards, and J. Doyon, Arch. Phys. Med. Rehabil., 2001,82(8), 1133-1141.
3. S. Page, P. Levine, and A. Leonard, Stroke, 2007, 38(4), 1293-1297.
4. J. Wolpaw, N. Birbaumer, D.Mcfarland, G. Pfurtscheller, and T. Vaughan, Clin. Neurophysiol., 2002, 113(6), 767-791.
5. L. Fadiga, G. Buccino, L. Craighero, L. Fogassi, V.Gallese and G. Pavesi, Neuropsychologia, 1998, 37(2), 147-158.
Computational Model of Neural-Glial-Ecm Interactions
S.V. Stasenko1 *, LA. Lazarevich1, V.B. Kazantsev1 and A.E. Dityatev2
1 Lobachevsky State University, Nizhny Novgorod, Russia;
2 German Center for Neurodegenerative Diseases, Magdeburg, Germany. * Presenting e-mail: [email protected]
Abstract. The study of principles and mechanisms of information processing in the brain is major issue of the modern neuroscience. Recent works have uncovered a participation glial cells (astrocytes) and ECM in information processing [1-4]. Astrocyte influences neural activity by releasing gliatransmitters: glutamate, D-serine and others. It has been discovered that ECM-mediated regulation mechanisms are involved in homeostatic modulation of neuronal activity [1,2]. Besides the well-known interactions, ECM-astrocyte interactions also may exist. Actrocyte acitivity leads to production of ECM molecules [1]. The influence of ECM molecules on astrocytes is associated with the change in the number and the properties of glial cells [5].
We present a model of neuronal activity regulation including ECM-glial-neuronal interactions. The neural network is modeled by a modification of the mean-field Wilson-Cowan-type model. The neural network consist of excitatory and inhibitory populations. The synaptic dynamics and astrocyte dynamics are modeled by using the mean-filed approach. The ECM is described by three activity-dependent variables (ECM molecules, ectoproteases and ECM receptors), which are involved in the following feedback loops: 1) decreasing the neuron excitation threshold due to ECM production; 2) increasing the excitation threshold due to ECM cleavage by ectoproteases; 3) changing effective strength of synaptic inputs due to signaling via the ECM receptors [2]. The ECM provides the regulation of the average firing rate, preventing hypo- or hyper-excitation of neurons due to the ECM-mediated feedbacks.
It was shown that interaction between ECM, astrocytes and the neuronal network leads to spontaneous activity oscillations on extended timescales. The interaction parameters determine the oscillation period (hours to days) and their existence and switching to bistable regimes.
Acknowledgements
The research was supported by the Russian Science Foundation (Agreement 14-11-00693). References
1. A. Dityatev and D. A. Rusakov, Curr. Opin. Neurobiol., 2011, 21, 353.
2. V. Kazantsev, S. Gordleeva, S. Stasenko, and A. Dityatev, PLoS One, 2012, 7, e41646.
3. A. Semyanov and D. M. Kullmann, Nat. Neurosci., 2001, 4, 718.
4. M. Min, Z. Melyan, and D. Kullmann, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 9932.
5. D. M. Kullmann, Neuron, 2001, 32, 561.
OM&P
