Articles in PresS. J Neurophysiol (March 25, 2015). doi:10.1152/jn.00467.

2013

1 Brain oscillatory signatures of motor tasks

2
3 Ander Ramos-Murguialday1, 2, Niels Birbaumer1, 3
4
1
5 Institute of Medical Psychology and Behavioral Neurobiology, University of Tubingen, Tubingen, Germany
2
6 TECNALIA, San Sebastian, Spain
3
7 Ospedale San Camillo, Istituto di Ricovero e Cura a Carattere Scientifico, Venezia Lido, Italy
8
9
10 Abstract— Non-invasive brain-computer–interfaces (BCI) coupled with prosthetic devices were recently introduced in the rehabilitation

11 of chronic stroke and other disorders of the motor system. These BCI systems and motor rehabilitation in general involve several motor

12 tasks for training. This study investigates the neurophysiological bases of an EEG-oscillations-driven BCI combined with a

13 neuroprosthetic device in order to define the specific oscillatory signature of the BCI-task. Controlling movements of a hand robotic

14 orthosis with motor imagery of the same movement generates sensorimotor rhythm oscillation changes and involves three elements of

15 tasks also used in stroke motor rehabilitation: passive and active movement and motor imagery and motor intention. We recorded EEG

16 while 9 healthy participants performed five different motor tasks consisting of closing and opening of the hand: 1) motor imagery

17 without any external feedback and without overt hand movement, 2) motor imagery which moves the orthosis proportional to the

18 produced brain oscillation change with online proprioceptive and visual feedback of the hand moving through a neuroprosthetic device

19 (BCI-condition), 3) passive and 4) active movement of the hand with feedback (seeing and feeling the hand moving) and 5) rest. During

20 the BCI-condition (2) participants received contingent on-line feedback of the decrease of power of the sensorimotor rhythm, which

21 induced orthosis movement and therefore proprioceptive and visual information from the moving hand. We analyzed brain activity

22 during the 5 conditions using time-frequency domain bootstrap based statistical comparisons and Morlet transforms. Activity during

23 rest was used as reference. Significant contralateral and ipsilateral event related desynchronization of sensorimotor rhythm was present

24 during all motor tasks, largest in contralateral-post-central, medio-central and ipsilateral-pre-central areas identifying the ipsilateral

25 pre-central cortex as an integral part of motor regulation. Changes in task specific frequency power when compared to rest were similar

26 between motor tasks and only significant differences in the time course and some narrow specific frequency bands were observed

27 between motor tasks. We identified EEG features representing active and passive proprioception (with and without muscle contraction)

28 and active intention and passive involvement (with and without voluntary effort) differentiating brain oscillations during motor tasks

29 that could substantially support the design of novel motor BCI-based rehabilitation therapies. The BCI task induced significantly

30 different brain activity compared to the other motor tasks indicating neural processes unique to the use of body actuators control in a

31 BCI-context.

32 Keywords-(Neuroprostheses; Motor action; EEG; Proprioceptive feedback)

33
34 I. INTRODUCTION

This work funded by the Deutsche Forschungsgemeinschaft (DFG) project KU 1453-1 and Koselleck to NB (BI 195/58-1), the Werner Reichardt Center of
Integrative Neuroscience (CIN) (2011-18 and 2008-6), the Baden Wuerttemberg Stiftung (ROB-1), the Bundes Ministerium für Bildung und Forschung (BMBF
Förderzeichen 01GQ0831), the European Union Information and Communication Technologies Framework Programme 7 (ICT-FP7) (HUMOUR 231724; WAY
288551) and the European Union COST action (TD1006).
Address for reprint requests and other correspondence: A. Ramos-Murguialday, Institute of Medical Psychology and Behavioral Neurobiology, University of
Tubingen, Silcher Str 5, 72076 Tubingen, Germany (ander.ramos@med.uni-tuebingen.de)
Copyright © 2015 by the American Physiological Society.
35 In an aging society motor rehabilitation is becoming increasingly important and one of the main focuses of the field is the use of

36 neuroimaging techniques to understand the effects of the different treatments. It has been demonstrated that changes in neural

37 networks correlate with improvements in motor function (Sehm et al 2010; Fritsch et al 2010; Koralek et al 2012) and that some

38 strategies exciting the motor neural networks in absence of movements through motor imagery determined motor improvements in

39 stroke patients (Sharma et al 2006). But a better understanding of the relevant neurophysiological processes involved during motor

40 rehabilitation is needed in order to improve treatment design, optimize efficacy and reduce intervention time. The use of BCI in

41 motor rehabilitation could help individuals without residual movements to control a rehabilitation robotic device and at the same

42 time serve as rehabilitation assessment tool (Ramos-Murguialday et al 2013). During BCI intervention one can track changes in

43 brain activity during treatment, the patients’ involvement during training (tracking volitional features in EEG or other neural

44 activities) and correlate these with functional and neuronal changes to predict or assess recovery. However, different motor

45 components produce very similar brain activity and as such the decoding of those different components is a complex problem.

46 Although, motor tasks and motor task components involved in motor rehabilitation (active and passive movement and motor

47 imagery) have been extensively studied these three motor tasks elements (active and passive movement and motor imagery) were

48 never studied in the same participants within the same experimental session minimizing the experimental variability.

49 Active (self-paced) limb movements have been mainly described to be accompanied by 3 different types of event-related

50 desynchronization and synchronization (ERD/ERS) patterns at the scalp EEG: (i) contralateral dominant sensorimotor rhythm

51 (SMR) in the alpha band and beta ERD prior to movement “pre-movement ERD” (Pfurtscheller & Neuper, 2006); (ii) bilateral

52 symmetrical alpha and beta ERD during execution of movement and (iii) contralateral dominant beta rebound (beta ERS) within the

53 first second after movement-offset (Pfurtscheller& Lopes da Silva, 1999). It has been shown that the topography of the alpha

54 changes is quite different for the low alpha band (8–10 Hz) and the high alpha band (10–12 Hz), suggesting that there may be two

55 different SMR-alpha rhythms contributing to the ERD during active movement (Pfurtscheller, 1989). The lower SMR-alpha ERD is

56 more prominent at parietal electrodes and the topography of the higher SMR-alpha ERD is more similar to that of the SMR-beta

57 ERD (range 17-30Hz), i.e. beginning in contralateral central regions becoming bilateral with the movement and extending to frontal

58 and parietal regions (Alegre et al 2004). Movement related slow cortical potentials during active movement have been very well

59 studied and divided into 2 main components: a) the slow cortical potentials (SCPs) occurring during intention or anticipation of an

60 upcoming movement which is also called the Bereitschaftspotential (BP) for self-paced movements (Barrett et al 1986) and b) the

61 motor potential (MP) occurring at the time of the execution (Deecke et al 1969).
62 Several studies demonstrated that after movement onset, synchronization (ERS) in the ipsilateral- as well as in surrounding, not

63 required, contralateral areas appears, whereas in the corresponding contralateral region the desynchronization continues until the

64 end of the motor sequence (ERD) (Alegre et al., 2004). This activation pattern has been associated with the concept of focal ERD

65 surrounded by ERS, whereby the less involved ipsilateral motor area neurons as well as the surrounding contralateral regions are

66 deactivated in order to increase the gain of the required active regions (Müller-Putz et al., 2007; Cassim et al., 2001; Hummel et al.,

67 2002; Pfurtscheller & Neuper, 2006), specific for the upper alpha frequency range. Furthermore, after movement offset a positive

68 slow cortical potential appears which may reflect the inhibition of the motor cortex after termination of the intended movement, as

69 well as afferent sensory projections to other motor-related areas acting as proprioceptive feedback (Birbaumer et al. 1990; Lee et

70 al., 1986; Müller et al., 2003; Müller-Putz et al., 2007).

71 On the other hand, during a passive movement voluntary planning and preparation are absent, yet the afferent somatosensory

72 components (skin mechanoreceptors, muscle spindles and joint receptors) produce SMR variations similar to those observed during

73 active movements (Alegre et al., 2002; Keinrath et al., 2006) with the absence of any preparatory ERD/ERS. In a PET study Mima

74 and his colleagues (1999) compared active and passive movements, during passive movements activity was observed in the

75 contralateral primary and secondary somatosensory cortex. While the classical motor network, including the primary motor cortex,

76 the supplementary motor area, the premotor cortex, the cerebellum and the basal ganglia were active during volitional movements.

77 They concluded that during passive movements there is a lack of sensorimotor attention (i.e. attention towards movements’ afferent

78 information), necessary to regulate movement control, which is present during active movements. Several groups observed similar

79 activation patterns, using EEG and ECoG data, between executed and passively performed movements (Lee et al., 1986; Lotze et

80 al. 2003, Müller-Putz et al., 2007, Cho et al., 2011). However, other studies have shown that primary motor cortex neurons receive

81 and process sensory afferent inputs from muscle spindles, without generating any active movements and thus are activated during

82 passive movement (Naito et al., 2002; Naito, 2004). However, most of these studies did not control for involuntary ipsilateral and

83 contraletreal upper-limb EMG activation, which could indicate unseen micromovements may have affected the observed EEG

84 activation (Ramos-Murguialday et al., 2010).

85 Motor imagery has been proposed as a tool for motor rehabilitation in combination with BCI (Birbaumer et al., 2008) and was

86 defined as rehearsal of movement without a joint overt action (Grezes & Decety, 2001). The cognitive process of motor imagery

87 does engage supraspinal structures, but usually lacks any spinal motor neuron activation (Stinear et al., 2006). However as of yet,

88 no description satisfactorily differentiates between the various forms of movement imagination, as the neural response clearly

89 differs depending on the perspective of the imagination (first “me” or third person “others”) and the strategy used (motor versus
 90 sensory for example). On the other hand, and most relevant for active brain control of rehabilitation robots, it has been proven that

91 kinaesthetic motor imagery, described as a “first person process” elicits very similar brain activity patterns to that obtained during

92 active movement execution (Neuper et al., 2005). Neurophysiologically the neural system used for the execution of a movement has

93 been presented as equivalent to the system activated during the kinaesthetic imagination of the same movements (Lotze et al.,

94 1999), although the activation is stronger during active movements (Jeannerod & Frak, 1999; Grezes & Decety, 2001; Pfurtscheller

95 & Neuper, 1997). Furthermore, the observation of movements was found to have an impact on the activity in sensorimotor areas,

96 reflecting processes of motor preparation and motor programming and activating the premotor cortex, the supplementary motor area

97 (SMA) and the pre-central cortex (Grezes et al., 1998; Halder et al 2012), indicating the importance of volitional initiation for a

98 visuo-motor task.

99 When addressing the relevant brain activation components in relation to active and passive movements, motor imagery follows

100 the same neurophysiological patterns as active movements (Gerardin et al., 2000; Ehrsson et al., 2002), except the overt execution

101 of the movement, no proprioception, no spinal activations etc. It is important to notice that imagery of unilateral right hand

102 movement can be accompanied by a contralateral beta ERD and an ipsilateral beta ERS (Pfurtscheller & Neuper, 1994, 1997)

103 similar to what has been observed for active movement. However, it is still unclear whether the primary motor cortex (M1) despite

104 being predominantly responsible for the motor command, is necessarily engaged during the imagination of a movement. Most

105 studies postulate the S1, the SMA, the cerebellum, the premotor- and the pre-central cortex to be active during motor imagery

106 (Sharma et al., 2006). While some studies suggest an apparent but weaker involvement of M1 (Grezes & Decety, 2001; Jackson et

107 al., 2003; Kasess et al., 2008; Lacourse et al., 2005), others dispute its role, arguing that motor imagery is primarily related to the

108 planning phase of a movement rather than the final execution (de Vries & Mulder, 2007; Dechent et al., 2004). Thus it might be that

109 the M1 is not necessary for an accurate movement representation in the brain, but is merely co-activated during imagery. However

110 it has to be stated, that most of these studies lacked appropriate controls for involuntary ipsilateral and contralateral upper-limb

111 EMG activation, which as previously mentioned could have affected the observed EEG activation (Ramos-Murguialday et al.

112 2010). This involuntary upper limb EMG activation may be caused by a plethora of reasons: a) due to undesired movement of the

113 hand during imagery of the same movement and cause EEG activity due to movement instead of only due to imagery; b) activity

114 during resting when they were asked to remain relaxed; c) bilateral muscle activity during active movement resulting in more

115 bilateral EEG activity when they were asked to move one hand and relax the other one.

116 Regarding the timing of a motor task, one has to differentiate between the preparatory and the execution phases. During

117 unilateral hand movement, the preparatory phase is associated with a contralateral alpha and central beta event related
118 desynchronization (ERD) that appears 1–2 s prior to simple movement onset and becomes bilaterally symmetrical in the execution

119 phase (Stancak and Pfurtscheller, 1996a,b; Derambure et al., 1999). Conversely in the case of motor imagery of the same unilateral

120 hand movement, the contralateral ERD is present during the whole imagery process, in a manner similar to the preparation of a self-

121 paced hand movement. (Pfurtscheller & Neuper, 1997; Pfurtscheller et al., 2006c), These observations indicate that similar (but not

122 identical) neural structures in sensorimotor areas become activated in the course of preparation of a motor action and during the

123 mental simulation of the same action without its execution (Jeannerod & Decety, 1995). Generally mu and ⁄ or beta components

124 display a somatotopically organized activation (ERD) pattern, with both the pre-movement ERD and the imagination-related ERD

125 appearing to reflect similar types of brain states in premotor and motor areas responsible for movement preperation, regardless of

126 whether an action will be performed or imagined (Neuper & Pfurtscheller, 1999). However sensorimotor rhythms such as mu and

127 central beta can be modified by executed, passive, imagined or observed movement and other cognitive tasks (Jasper & Penfield,

128 1949; Salmelin & Hari, 1994; Leocani et al., 1997; Pfurtscheller & Neuper, 1997; Derambure et al., 1999; Neuper & Pfurtscheller,

129 1999; Alegre et al., 2002; Muthukumaraswamy & Johnson, 2004; Pineda, 2005, Weiller et al., 1996).

130 When addressing the relevant brain states during the use of robots for neurorehabilitation, passive motor trainings have not been

131 shown to result in gains in motor function because active participation and volition seems necessary to produce significant motor

132 function improvement (Hu et al 2009, Lotze et al 2003, Hogan et al 2006). The control of a robot or prosthetic device and the

133 feedback contingency, are of vital importance to enable neuro-motor-rehabilitation, as in an active motor rehabilitation treatment

134 the feedback plays an important role in coupling intention and action (Ramos-Murguialday et al 2013).

135 The critical elements of feedback have been distinguished according to three aspects; feedback modality (visual, auditory,

136 haptic, etc.), the timing of feedback (continuous or binary) and the transmission of the feedback (contingent or non-contingent). The

137 expression “continuous” refers to the whether feedback is presented throughout the entire trial; alternatively in a binary fashion the

138 feedback is presented once at the end of each trial. Since a therapist cannot accurately control all the aspects of feedback, robots

139 have been introduced as tool to improve repetition, precision and control during motor rehabilitation. Robots can be used as

140 tracking system (no friction and only tracks limb movements and forces), assisting (compensating gravity, guided paths etc),

141 resisting (increasing friction, perturbing trajectories etc). In order to control a robot during these three types of rehabilitation

142 scenarios a classifier (based on force, movement, EMG, EEG…) to detect action intention and to operate the robot joints is

143 necessary. When utilizing the commonly used linear classifier to control a robot, a threshold has to be defined based on force,

144 velocity, muscle or brain activity and feedback is provided each time the activation level crosses that activity threshold. This could

145 happen either at the end of the trial (binary) or during the ongoing trial (continuous). The distinction between contingent and non-
146 contingent defines the amount of association in time between the response (force, velocity, and muscle or brain activity) and the

147 feedback signal.

148 In this study we investigated the difference in brain activity during several commonly studied motor rehabilitation tasks: motor

149 imagery, passive and active movements. During these tasks we analysed EEG frequency changes over time. Furthermore, we added

150 an extra task linking brain oscillations (related to the imagination of hand opening and closing) with hand movements using a

151 robotic hand orthosis (haptic-BCI). The aim of our work was to analyze the cortical energy changes in the theta, alpha, beta and low

152 gamma bands during four motor tasks when compared to a resting condition. Three of the motor tasks were related to motor

153 rehabilitation (motor imagery, active and passive movement) and were compared with the BCI task, in which the three elements of

154 the tasks are mixed in order to link brain activity due to motor imagery of hand opening/closing and the exact same movements via

155 a BCI controlled robotic hand orthosis. Furthermore, we intended to study the relationship of these EEG oscillation changes with

156 the preparation, planning and execution of each of the motor tasks and hypothesized that: a) each motor task will present individual

157 EEG signatures (EEG power at locations L1, L2..Ln during time periods t1,t2…tn), i.e. characteristic EEG power at different locations

158 during different time periods when compared to the other motor taks (especially against rest that was used as reference); b) that

159 EEG activity during brain control of the orthotic device (BCI condition) would resemble neural activation generated during active

160 movement.

161 II. METHODS

162 A. Experimental Procedure

163 9 healthy volunteers (age range 26.6±4 years; all subjects were right-handed) were recruited for the experiments. Participants

164 were sitting in an upright position wearing a 128 channel EEG cap. The experimental protocol was approved by the ethics

165 committee of the University of Tubingen, Medical Faculty; with all participants providing written informed consent prior to

166 participating in this study. The predominant hand (right) of the participant was fixed to a hand orthosis (See. Fig.1.C) and

167 participants were presented randomly with 5 different auditory cues indicating the motor tasks to perform:

168 1. Motor imagery without direct control (MI) of the orthosis i.e. the participant imagined to move and imagined to feel the

169 hand moving without moving it and with no movement of the orthosis

170 2. Motor imagery with direct control (BCI task) of the orthosis i.e. the brain rolandic rhythm oscillations (SMR) during

171 imagining of hand opening and closing drove the movements of the orthosis attached to the hand
172 3. Passive movements of the orthosis i.e. the hand attached to the orthosis was opened and closed passively without any

173 concurrent mental task

174 4. Active movement i.e. the participant was required to open and close the hand attached to the orthosis, the orthosis following

175 the movements

176 5. Rest condition

177 The participants sensorimotor rhythm (SMR) oscillations were directly linked with the orthosis movements during the BCI task

178 only. Consequently during a BCI task trial, SMR decrease in power (desynchronization) was “rewarded” orthosis movement and

179 vice versus, an increase in SMR power (synchronization) was “penalized” with the restriction of orthosis movement.

180 The orthosis range of motion was adjusted for each participant before starting the experiment. Two seconds after hearing the

181 corresponding “task to perform” auditory cue, a “GO” cue was presented and the participant performed the appropriate motor task

182 for 5 seconds before movement was terminated by an auditory “end” cue at the end of the task period. All the spoken auditory cues

183 were previously recorded and normalized in pitch, volume and length to avoid cue differences. All participants had no prior BCI

184 experience and completed 5 experimental sessions; each session was undertaken on a different day and consisted of completing 10

185 runs of 25 trials each resulting in 50 trials per condition per session. An EEG-screening was performed the day before the first

186 training session and was used as a calibration session to identify the features, i.e. electrodes and frequency bins presenting higher r-

187 square values when comparing MI and rest, which were to be used by the BCI classifier. More information about the BCI

188 calibration and functioning can be found in Ramos-Murguialday et al. (2012).

189

190 B. Data Acquisition

191 EEG data were acquired using a BrainAmp 128-channel amplifier from Brainproducts GmbH, Munich Germany. An EasyCap

192 128-channel EEG cap (modified 10-20 system) from EASYCAP GmbH, Herrsching, Germany was used for EEG data acquisition;

193 referenced to the nasion, and grounded anteriorly to Fz. Only 61 EEG channels over the motor areas on both hemispheres were used

194 focusing on pre-motor, motor and post-central areas (See Fig.1.B). Additionally, we measured horizontal EOG on both eyes and

195 vertical EOG on the right eye for artefact correction and EMG on both upper- and fore-arms for the detection of unwanted muscle

196 contractions (e.g. muscle contraction during rest or MI). Data were sampled at 500 Hz and transmitted to a PC for both storage and

197 real-time signal processing, using the BCI2000 platform (www.bci2000.org). EMG data was acquired using 8 bipolar Ag/AgCl

198 electrodes from Myotronics-Noromed (Tukwila, WA, USA) and placed on antagonistic muscle pairs; one close to the external
199 epicondyle on the extensor digitorum (forearm extensor), one on the flexor carpi radialis (forearm flexor), one on the external head

200 of the biceps (upper arm flexor) and the last one placed on the external head of the triceps (upper arm extensor). The EEG and

201 EMG electrodes impedance was always kept under 5 and 20 kOhm respectively.

202 C. Signal Processing

203 We performed a time-frequency analysis using a 1.142 seconds sliding window with an overlap of 26 msec. EEG data were first

204 filtered using a band-pass filter (2 – 45 Hz) and segmented from 2 sec before to 5 sec after the “GO” cue, the baseline from -1.5 to

205 -1 seconds before the “GO” was removed from each segment to remove offsets and the event related spectrum perturbation was

206 then calculated using Morlet transforms (Daubechies, 1996). The Morlet transforms used 3 cycles at lowest frequencies and 23.04

207 at highest and a time window of 1114 msec and 30 msec overlap. A 200 msec time window from -1.5 to -1.3 sec before the go cue

208 was used as baseline for the event related spectra perturbation analysis (normalization). 85 linear-spaced frequencies from 3 Hz to

209 45 Hz and 200 time points were generated.

210 In order to test for significant differences in brain activation within-subjects and between-motor tasks, we used non-parametric

211 statistics (Bootstrapping) not assuming a particular distribution. We performed bootstrap analyses to compare brain activity (time-

212 frequency) during the motor tasks. Comparisons were performed always in a pairwaise manner, i.e. comparing 2 motor tasks and

213 adopting the comparison rest vs motor task (BCI, MI, passive and active movement) as reference for our analyses of each motor

214 task (See Table1).

215 When comparing “task a” to “task b”, we subtracted in a trial by trial basis “task a” from “task b” time-frequency spectra and then

216 we averaged these differences and created a time-frequency spectra representing the mean spectral difference between tasks.

217 (DIFF). For the spectral difference between tasks time-frequency plot (DIFF) generation, a baseline removal (power

218 normalization) was performed using a common power baseline calculated averaging both tasks baseline means. This common

219 baseline was then subtracted from each task time-frequency plot before computing the mean spectral difference time-frequency

220 plot. (See Fig.2.C).

221 Spectral differences between tasks not resulting as significant using the EEGLAB toolbox (http://sccn.ucsd.edu/eeglab/) were

222 masked and plotted green (See Fig.2). The null hypothesis distribution used to determine significance thresholds was estimated by

223 accumulating surrogate data from 200 bootstrap replications, shuffling the single trial spectral DIFF estimates using a two-tailed

224 bootstrap significance probability level (Burgess & Gruzelier, 1999; Efron & Tibshirani, 1994). The t 1 and 99 percentiles

225 obtained for every frequency and time bins (every “pixel” of the time-frequency plot) from the spectral difference between tasks

226 (DIFF) were calculated and averaged from all 200 bootstraps. This allowed us to determine at each time and frequency bin, the

227 power values determining significance (i.e. with a p-value=0.01). . No averaging of the percentiles’ values was performed in the
228 frequency dimension because at different frequencies we have different power values (higher at lower frequencies) and this would

229 produce false positives in low frequencies and false negatives in high frequencies. Averaging of the percentiles in time was

230 performed to see significant changes with respect to all changes in time during the entire 7 seconds of the task including

231 preparation, onset and task execution, to compute the bootstrap significance level (i,e. mask).

232 It is very important to note, that in this paper we considered significant differences in the time-frequency spectra between classes,

233 only clusters (group of neighboring pixels in time or frequency dimension) of more than 18 pixels presenting the same sign

234 significant difference in power with a minimum size in the X (0.5 Hz/pixel) and Y (33 msec/pixel) axes of 3 pixels. We consider

235 that the percentiles averaging over time and the cluster analysis compensate the lack of a multiple comparison correction to an

236 acceptable extent without losing information related to correlation and noise.

237 Our main goal of the paper is to describe spectral differences between motor tasks and describe features that might serve as

238 differentiating factors between conditions. If we use hypothesis testing and correct for multiple comparisons we would ignore and

239 lose class power (i.e. correlation and noise information are lost). Thus, some of the features alone would not result in significant

240 differences, but if used together, some features might become significant and help a classifier to differentiate between classes (i.e.

241 predictability based on recording). In order to demonstrate this effect, we used a nonparametric framework and determined the

242 Monte-Carlo estimates from the permutation distribution to compare spectra correcting for multiple comparisons. Permutation

243 statistics were computed for each sample in the time-frequency space as implemented in the Fieldtrip open source toolbox

244 (Oostenveld et al. 2011). We used 1000 random permutations on a dependent samples t-test and an adjusted alpha level of

245 p<0.025 per tail. Although these sample-wise statistics produce a massive number of comparisons (i.e. the product of the number

246 of channels (61) by times (200) by frequencies (85)), the cluster-based correction method was demonstrated to treat the

247 multiplicity problem effectively without losing sensitivity for spectral modulations (Maris and Oostenveld 2007) (See Fig9).

248

249 The EMG data were filtered using a high pass filter at 10 Hz, bipolarized and rectified. The waveform length of the signal (Ferry

250 et al 1996), which provided us information about amplitude and frequency of the EMG signal, was extracted for every motor task

251 and rest. After extracting the waveform length (WL) of the signal, the EMG data at rest were scanned for artefacts. Any trial with

252 EMG activity exceeding 3 standard deviations (SD) at any point in time, at any electrode, within the analyzed time window (1.5 to

253 4 sec with respect to the “GO” cue) was removed and the remaining “considered artefact-free” trials were used to calculate an

254 EMG activity threshold; this threshold was set at 3 SD from the mean of activity at rest. Any EMG activity exceeding this

255 threshold (from -2 to 5 sec with respect to the “GO” cue) for a continuous period of more than 25 msec was considered as muscle

256 contraction. Trials presenting muscle activity at any electrode during the resting task or on the resting hand during any motor task
257 and absence of activity during active opening and closing of the hand were excluded from the EEG analysis. A minimum number

258 of 15 trials per person was required to be included in the analysis, which resulted in 193, 182, 234,203 and 155 trials for MI, BCI,

259 passive and active movement and rest respectively.

260 The Time-frequency changes analysis was divided in 3 time windows (See Fig.1.A): a) preparation period comprising the 1.5

261 seconds before the “GO” cue presentation starting 0.5 sec after the instruction period in which the participants were presented

262 with the task to be performed, b) time immediately after presentation of the “GO” cue comprising the 200 msec after the cue, c)

263 Time interval during the task 5 seconds the trial lasted and subdivided these time windows (a, b and c) in four frequency ranges:

264 a) theta (3-6 Hz), b) alpha (6-14 Hz), c) beta (15-30 Hz) and d) gamma (30-45 Hz).

265

266

267 III. RESULTS

268 Preparation period (Instructions period before the “GO” cue presentation):

269 During this time the participants hear an auditory cue instructing them which motor task will be performed and have 2 seconds to

270 prepare themselves before the imperative “GO” cue indicating the beginning of the motor task action.

271 Theta (3-6 Hz):

272 During the last 200 msec before the “GO” cue, a significant increase in power was observed in all electrodes and being larger at

273 contralateral pre-motor electrodes when comparing MI and rest (5-8Hz) (See Fig.3; electrode C1). Contrastingly when comparing

274 BCI with rest, a significant decrease of power (5-7Hz) occurred 1.5 to 1.2 sec prior to cue onset in the contralateral and pre-

275 central ipsilateral electrodes (See Fig.4; electrode C1). Furthermore when comparing preparation for active movement and rest,

276 significant theta (3-5Hz) increase in power was observed from 700 to 350 msec before the “GO” cue in all electrodes; however

277 this increase was relatively constant and stronger in parieto-central electrodes.

278 When comparing preparation for active movement with preparation for MI, BCI and passive movement we observed significant

279 higher power in theta frequencies during preparation for active movement in all electrodes (from 500 to 300 msec before the

280 “GO” cue when compared to MI and the last 300 msec when compared to BCI and passive movement preparation before the

281 “GO” cue) (See Fig.6). No significant difference was observed in the theta band before passive movement.

282

283 Alpha (6-14 Hz):

284 Significant increase in power was observed for the last 250 and 800 msec before the “GO” in all (mainly parietal) electrodes

285 preparing for MI and active movement tasks respectively (MI: 8-12Hz and Active: 6-10Hz). The significant increase in power in

286 the alpha band during the preparation for active movement started earlier and was much stronger than during the preparation for

287 MI, eliciting significant differences continuously until the “GO” cue presentation (See Fig.6). A 200 msec decrease in power was

288 observed in all electrodes from 1500 to 1300 msec before the “GO” cue when comparing preparing for passive movement and

289 preparing to rest (See Fig.5). This effect seems to be a prolongation of a baseline effect. Interestingly no significant difference in

290 power was observed when comparing the BCI task with the passive movement. Preparation for the BCI task evokes significant

291 power increase in the alpha rhythm frequency band when compared to preparation to rest in frontocentral contralateral electrodes

292 only (See Fig.4; electrode FC1), implying that expecting brain control of proprioceptive activation without muscle contraction (as

293 in the BCI task) does not elicit significant alpha power increase as during the preparation for an active movement or for a motor

294 imagery of that same movement in all electrodes.

295

296 Beta (15-30 Hz):

297 During the preparation for BCI and MI tasks when compared to preparation to rest, we observed from 500 to 200 msec before the

298 “GO” a significant increase in power in the 16-18Hz band in the precentral electrodes and in ipsilateral central electrodes

299 respectively (possibly related to attention (Klimesch et al 2012)). During preparation for BCI, passive and active movement, we

300 found significant power changes during the last 500 msec before the beginning of the task. Significant decreases in power during

301 prepariation for BCI and passive movement from 18 to 22Hz (BCI: all electrodes (mainly contralateral) and Passive: contralateral

302 electrodes) were observed. Significant increases in power during preparation for active movement in all electrodes being more

303 pronounced in post-central electrodes bilaterally (15-17Hz and 21-23Hz) were found.

304 Nevertheless, when comparing preparation for passive movements and BCI, two significant periods of significant lower power

305 preparing for BCI from -700 to -500 and the last 100 msec extending to 300 msec after the “GO” were found at 24-30 Hz and 18-

306 22Hz respectively. While preparation for MI just showed significant power increase right before the cue in low beta 16-18Hz,

307 preparation for active movement and BCI showed significant increase in power from 16-24Hz when compared to preparation for

308 passive movement. These results indicate 2 beta band components during task preparation:. a) the contralateral 18-23Hz activity

309 (decrease in power) might comprise preparation for passive proprioceptive (without muscle contraction) activation (present before

310 BCI and passive movement only); b) the ipsilateral pre-central 15-18Hz increase in power might correspond to attention to a task

311 or preparation to execute or imagine movement maintaining hand posture (present only before BCI, MI and active movements) in

312 line with Hatsopoulos & Donoghue, 2009 findings.

313

314 Gamma (30-45 Hz):

315 During preparation for all motor tasks, 30 to 45 Hz significant changes in power were found when compared to preparing to rest

316 in all electrodes.

317

318 Onset or first 200 msec after the “GO” cue presentation:

319 Theta & Alpha (3-14 Hz):

320 These two frequency bands presented similar activity and therefore the significant changes are reported in the same section. No

321 significant differences were found when comparing the rest task onset with the BCI and active movement tasks onsets. The MI

322 task showed a significant decrease in power (8-12Hz) in all electrodes (See Fig.3) at the onset. However, this decrease in power

323 was largest in the contralateral electrodes, predominantly in the post-central areas and in ipsilateral pre-central electrodes;

324 interestingly for the passive movement task, high alpha rhythm (12-14Hz) significant power decrease was found in the medial-

325 precentral electrodes (Cz,Fz,F1,F2,Fcz,Fc1,Fc2).

326

327 Beta (15-30 Hz):

328 All motor tasks but active movement showed significantly lower power in all electrodes in this frequency band when compared to

329 rest at onset. For the MI and BCI tasks onsets, the significant decrease in power was broad-banded (18-28Hz) predominantly over

330 contralateral regions and highest in post-central areas. Contrastingly during passive and active movement tasks onset, the

331 significant decrease in power was in a narrow band (24 to 27Hz) mainly over ipsilateral-post-central regions for active and across

332 all electrodes for passive movement. The largest differences were observed in contra-post-central electrodes (See Fig.5; electrode

333 CP3).

334 Interestingly when comparing BCI with passive and active movements tasks onsets, significantly larger contralateral power

335 decreases (18-22Hz) were seen during the BCI task onset. These decreases were only over the contralateral pre-central electrodes

336 when BCI was compared to passive onset.

337

338 Gamma (30-45 Hz):

339 Only a 100 msec period of significant power decrease (35-42 Hz) was observed in all electrodes during the first 200 msec after the

340 “GO” cue presentation during MI, BCI and passive movement onset being absent during active movement onset.

341
342 During the task (excluding the first 200 msec):

343 Theta (3-6 Hz):

344 During MI, BCI and passive movement tasks a significant power decrease was found in all electrodes from 750 to 900 msec after

345 the cue when compared to rest. However for the active movement task, a significant power increase was found during the same

346 interval (700 to 1000 msec) across all electrodes (See Fig.6). During the BCI and active movement tasks, segments of frequency

347 clusters presenting significant theta power increases (3-6Hz) were observed after the first 1800 msec until the end of the trial,

348 again across all electrodes. Furthermore in the active movement task, the power increase was larger than for the BCI tasks and

349 more pronounced in contralateral areas. But during the BCI task, the power increase was significantly larger compared to that of

350 the MI task. During passive movement 2 segments of frequency clusters presenting significant power decrease, between 1.2 to 1.5

351 and 3.6 to 3.8 sec, were found (3-7Hz) in all electrodes. These time points are exactly the times in which the orthosis changes

352 from opening to closing (i.e. orthosis stops for 150 msec to change from opening to closing changing velocity’s sign because

353 maximum open or close position was reached). In this frequency band (theta; 3-6 Hz) BCI presents similar brain activity to that

354 during active movement.

355

356 Alpha (6-14 Hz):

357 This is the frequency band in which significant decrease in power was more prominent for all the motor tasks when compared to

358 rest. During execution of all motor tasks alpha rhythm was the frequency range which presented more significant power changes

359 when compared to rest. The same spatial distribution was found in all tasks, with significant decreases in power observed over all

360 electrodes, but generally being more pronounced and continuous in ipsilateral electrodes for pre-central electrodes and

361 contralateral for post-central electrodes. Thus, identifying the ipsilateral pre-central cortex as an integral element of motor

362 regulation as expected. We will refer to this significant decrease in power spatial distribution as a “significance diagonal” from

363 contra-post-central to ipsi-pre-central electrodes in motor areas.

364 During MI and BCI tasks the significant difference in power occurred in the 7 to 10Hz frequency band (See Fig.3 and 4), during

365 the passive movement task this power difference occurred in the 6 to13 Hz frequency band (See Fig.5) and during the active

366 movement task it occurred in the 8 to 11 Hz frequency band (See Fig.6).

367 The MI task alone was the only motor task showing clearly significant power decreases from 200 msec before the onset until the

368 end of the trial (See Fig.3; electrode CP3), but only for the contralateral post-central electrodes. After the initial 300 msec,

369 significant decreases in power could be observed. The significant decrease in the ipsilateral electrodes displayed the earliest
370 response, particularly in parietal electrodes before appearing later in the medial and finally in the pre-central electrodes (posterior

371 to anterior).

372 During MI across the medio-pre-central electrodes there were two periods presenting significant decrease in power, one at the

373 beginning of the trial for 350 msec and the second starting 1.5 sec after the “GO” cue lasting until the end of the trial (See Fig.3;

374 electrode C2).

375 During the BCI task, the significant decrease in power starts early in contralateral post-central electrodes 300 msec after the “GO”

376 cue, then continues until 2100 msec after the beginning of the trial and after 350 msec of no significant change in power it

377 decreases significantly again until the end of the trial (See Fig.4). The significant decrease in power was observed for ipsilateral-

378 post-central electrodes throughout the entire trial.

379 During passive movement tasks we observed the beginning of significant spectral changes at around 200 msec in the contralateral

380 hemisphere when compared to rest, similar to what we observed when BCI was compared to rest (See Fig.5; electrode CP3). In

381 contrast, in the ipsilateral side the significant spectral changes start 150 msec earlier than during BCI task, at 200 msec in the post-

382 central areas and is delayed towards pre-central areas (See Fig.5; electrode CP4).

383 The significant decrease in power during passive movements was present until the end of the trial in medial-ipsilateral electrodes.

384 Active movement was the task with the longest latency to the first significant decrease in power of the alpha rhythm when

385 compared to rest, starting 800 msec after the “GO” until the end of the trial in post-central areas only. This late onset of significant

386 decrease in power could be explained due to a longer latency to the first movement onset for all users as demonstrated by the

387 EMG activity presented in Fig.3B. In the active movement case, the significant decrease in power was continuously present

388 towards the end of the trial only in the “significance diagonal” (contra-post-central ipsi-post-central and fronto-ipsilateral) and its

389 latency was the same in the post-central medio-lateral electrodes opposed to Takahashi et al (2011).

390 We found more (more time-frequency bins and/or larger difference in power) reduced power (7-10 Hz) during MI than during

391 BCI for the first 500 msec when compared to rest mainly in contralateral pre-central electrodes. However, we observed more

392 power decrease during BCI than during MI when compared to rest for the last 2.5 seconds of the trial in temporo-medial and

393 medio-post-central ipsilateral areas. Passive movement presented higher power than BCI during the first 2.5 sec (6-13Hz) in

394 ipsilateral central and medio central and pre-central.

395 During the last second of passive movement and MI task we observed smaller ERD (8-12Hz) in post-central areas (mainly

396 ipsilateral) when compared to the BCI task. On the other hand, during BCI we observed periods of lower power (8-10Hz) than

397 during active movement mainly in contralateral-post-central electrodes and in ipsilateral-post-central electrodes during the last 1.5

398 sec (See Fig. 3).

399 Summarizing, these data suggest that during BCI the significant power decrease in ipsilateral- medio- post-central electrodes at

400 frequencies between 8 and 10 Hz during the last second of the trial might be a unique effect of the haptic BCI task when

401 compared to the other motor tasks and rest.

402

403 Beta (15-30 Hz):

404 Contrary to most of the previous results commented in the literature (Pfurtscheller & Lopes da Silva, 1999, Pfurtscheller and

405 Neuper 1994,1997, Alegre et al. 2004, Stanack and Pfurtscheller 1996a,b, Derambere et al. 1999). Only limited and very short

406 periods of significant power changes were found during the active movement task in low beta (18-24Hz) when compared to rest

407 (See Fig.6; electrode C2). However, significant decreases in power during active movement was present mainly in medio-pre-

408 central and medio-post-central electrodes (SMA), which started before movement initiation and continued during movement,

409 mainly in the first 1.5 sec after muscle contraction started, in line with the previous work (Ohara et al. 2000, Pfurtscheller et al.

410 2003). A short 200 msec period of significant power increase was observed for all electrodes during active movement in the 14-16

411 Hz range, concurrently with the first significant increase in power in low-gamma high-beta frequency range.

412 During MI and BCI tasks a significant decrease in power when compared to rest was found for the first 450 msec (MI: 14-20 Hz

413 and BCI: 20-24 Hz) mainly across the significance diagonal, while during the passive movement task just the frequency range

414 from 20 to 24 Hz showed significant power decrease. Whilst during the MI and BCI tasks compared to rest segments of frequency

415 clusters presenting significant changes in power were found throughout the trial. Interestingly, we observed significant high beta

416 decrease in power (20-24Hz) only for the BCI task, in a manner similar to that observed during the passive movement task when

417 compared to rest.

418 During passive movements intermittent significant decreases in power were found in the 16 to 24 Hz frequency range becoming

419 more continuous from 1 to 3.5 sec after the “GO” (See Fig.5; electrode CP3) when compared to rest. The decrease in power in the

420 beta range during passive movements was larger than the decrease during the MI task in ipsilateral centro-post-central electrodes

421 during the entire trial and smaller in contralateral electrodes during the first 300 msec.

422 When comparing MI and BCI, we found significantly larger power decreases in the beta band (18-22Hz) for the BCI task. This

423 decrease occurred towards the end of the trial and was localised mainly to the ipsilateral-post-central electrodes. In contrast if BCI

424 and passive movement were compared, we again found significantly larger decreases in power (18-22Hz) during the BCI task,

425 however this was localised in contralateral medio-post-central electrodes and occurred during the first 500 msec of the trial. These

426 results reflect the influence of proprioception in the ipsilateral beta (20-24 Hz) decrease in power and of the motor volition in the

427 first milliseconds after the “GO” cue in the contralateral beta (20-24 Hz) decrease in power. Furthermore, these results indicate
428 either operant regulation or proprioceptive feedback, are both key factors in maintaining desynchronization in the 20 to 24 Hz beta

429 frequency range, as opposed to the main role of the MI in the significant beta desynchronization reported previously (Pfurtscheller

430 and Neuper 1997), which we found to happen in the 14 to 20 Hz. Furthermore significant beta band decreases in power were

431 larger and more continuous during MI than during active movement tasks.

432 When compared to rest, the passive movement produced more significant changes in beta frequency range than the active

433 movement, MI and BCI tasks. Significant changes in the 20 to 24 Hz frequency band were observed during BCI and passive

434 movement tasks only when compared to rest. All this suggests the implication of these rhythms (20-24Hz) in proprioception

435 without active muscle contraction; However, during the BCI task the significant power changes were less continuous.

436

437 Gamma (30-45 Hz):

438 During MI, BCI and Passive movement a 200 msec period of significant decrease in power (30-45Hz) was present in all

439 electrodes 800 msec after the “GO” cue when compared to rest. During active movement a significant increase in power (38-

440 45Hz) from 200 to 400 msec after the “GO” cue was observed. This significant power increase continued in contralateral post-

441 central electrodes, spreading to a broader frequency band (26-45Hz) starting earlier in contralateral-post-central electrodes and

442 sustained until the end of the trial.. The significant gamma increase in power in the 26 to 45Hz frequency range started

443 concurrently with significant decreases in power of the slow oscillations (4-10 Hz) in the contralateral hemisphere (See Fig.6), in

444 a manner consistent with previous research (Pfurtscheller et al. 1993, 2003b Pfurtscheller and Neuper 1994). Generally significant

445 increase in power (26-45Hz) started earlier in post-central contralateral electrodes compared to central and pre-central electrodes,

446 with the onset latency of the significant increase in power increasing together with the distance to post-central electrodes, thus the

447 more frontal the location of the electrode the larger the onset latency; However, in the ipsilateral hemisphere no delay between

448 post-central and pre-central areas was observed.

449

Motor Task Active Movement Passive Movement Imagined Movement BCI condition

Time Period Prior Onset During Prior Onset During Prior Onset During Prior Onset During

ERD (Power decrease)

3-6 Hz - - - - - X - X** X x - X

6-14Hz - - XXXX x X XXXX - XX XXXX - - XXXX

Contra. 15-30
- - X x XX XXX - XX XX XX XX XX
Hz
30-45
x - - X X x X X x X X x
Hz
3-6 Hz - - - - - X - - X x - X

6-14 Hz - - XXXX - X XXXX - X XXXX - X XXXX

Ipsi. 15-30
- - X - X XXX - X XX X X XX
Hz
30-45
x - - - X x X X x X X x
Hz

ERS (Power Increase)

3-6 Hz XX - X - - - X - - - - X

6-14 Hz XXXX - - - - - XX - - x - -
Contra. 15-30
XX - x - - - X - - - - -
Hz
30-45
X - XXXX X - - X - - x - -
Hz
3-6 Hz XX** - X - - - X - - - - X

6-14Hz XXXX - - - - - XX - - - - -
Ipsi. 15-30
X - x - - - X - - x - -
Hz
30-45
X - XXXX X - - X - - x - -
Hz
Table 1. Significant power change during motor tasks compared to rest. Significant decrease and increase in power before (preparation), at onset
(planning) and during (execution) the 4 different motor tasks (Active movement, Passive movement, motor imagery (MI) and motor imagery with
proprioceptive feedback (BCI)) when compared to the other conditions. X stands for significant power change intensity in power and time (i.e. difference in
number of pixels in the time-frequency plot showing significant power changes). The number and size of the X’s represent the difference in power in each
time period for each task (being x little and XXXX great significant power changes). Light shaded grey cells indicate significant changes during BCI
condition only and dark shaded cells indicate EEG activity changes related to the BCI condition and active movements only when compared to rest. Two
asterisks (**) represent unique task-specific significant changes in power, present during a specific time period (Prior, Onset, During) in that motor task
when compared to rest only (i.e, no significant power changes when comparing during that period of time any of the other three tasks to rest individually).
Prior, Onset and During stand for the time periods with respect to the “GO” cue from -1500 to 0 msec (preparation for a task), 0 to 200 msec (task initiation
in response to an imperative cue)and 200 to 4500 msec respectiveley (during task execution), as described in Fig1.

450 IV. DISCUSSION

451 We described changes in time and frequency recorded from EEG electrodes placed over the motor strip during 4 motor tasks and

452 rest and summarized our findings in 3 time periods (preparation, onset and during task) and some standard frequency bands (See

453 Table 1). X-values stand for significant power change, calculated as the number of pixels presenting significant power changes

454 when comparing that motor task to rest during that time period. The number and size of the X’s represent the difference in power

455 in each time period for each task, being x little and XXXX great significant power changes. We calculated the maximum and

456 minimum X-value for each time period over all the tasks and segmented it in 8 equally large Pow ranges (x, xx, xxx, xxxx, X,

457 XX, XXX, XXXX).

458 The presence/absence of X-values does not necessarily mean identical activity in tasks presenting/not-presenting X-values in the

459 same time period. (e.g. if X is present prior to BCI task and Passive Movement, it does not necessarily mean that prior to both

460 tasks result in identical power changes). The X values could result from significant activity changes during different time periods

461 or slightly different frequency bins when task-related power changes were compared to resting-related power changes. To

462 statistically compare between motor tasks power changes, additional analyses were carried out (e.g. See Fig.7).

463

464 The aim of our work was to analyse the cortical energy changes in the theta, alpha, beta and low gamma bands preparing for, at

465 onset and during the execution of four motor tasks when compared to a resting condition. In most of the previous literature

466 investigating spectral changes of motor tasks, time-frequency power changes have been analysed using a baseline to normalize

467 and to observe spectral changes with respect to that pre-defined baseline. Inter-trial intervals or instructions periods are commonly

468 used as baseline periods. In this work, although we did not correct for multiple comparisons, we used statistical analyses to

469 describe relevant spectral changes between motor tasks using a resting class as reference. Hypothesis testing correcting for

470 multiple comparisons ignores and lacks correlation and noise information, which indeed would mask features that combined may

471 help a classifier to differentiate between classes (i.e. predictability based on recording) (See Fig9). Our results could therefore be

472 taken as a study towards identifying the features that might help classify different motor tasks using EEG spectral changes.

473

474 The control for involuntary muscle contractions was performed because EMG activity has been observed in forearm extensor

475 muscles during contralateral forearm activity and during mental activity, without finger movements being required (Sogaard et al.,

476 2001). Similarly Whitham et al (2007) observed EMG activity in paralyzed subjects with a single limb excluded from paralysis

477 and as such it has been suggested that EMG activation during mental activity, may be related to evolutionarily important

478 mechanisms for preparedness to defend themselves or react, namely mental activity may constitutively activate protective

479 programs of readiness for action (Whitham et al 2007).

480

481 Preparation period (Instructions period before the “GO” cue presentation):

482 We observed that preparing for BCI does not resemble preparing for active movement or MI EEG activity. We found that

483 preparation for BCI (hand movements without muscle contractions), presented significantly lower preparatory theta and alpha

484 power compared to MI and active movement preparation (presenting significant increase in power) when compared to preparation

485 for rest. This might be explained by participants shifting their attention more towards the proprioception than towards the actual

486 MI during this preparatory phase. It has to be noticed that the comparisons here were made against preparing to rest, which indeed

487 presented power changes during preparation very similar to all the motor tasks (See Fig2.B).

488 In our data we demonstrated how preparing to move induced significant power increase in SMA and centro-parietal electrodes.

489 Numerous studies have implicated the SMA in the early representation of preparation for movement (Amador and Fried, 2004;

490 Brinkman, 1984; Erdler et al., 2000; Fried et al., 1991; Ikeda et al., 2002; Laplane et al., 1977; Lau et al., 2004a, 2004b; Lim et

491 al., 1994; Scepkowski and Cronin-Golomb, 2003; Shima and Tanji, 2000; Tanji, 1994; Thaler et al., 1995). In recordings with

492 intracranial EEG it has been observed that movement associated ERD is preceded by a weak ERS in a non-movement condition at

493 around 7-12 Hz, compared to a slightly stronger and focal early ERS in two electrode contacts from the left post-central Rolandic

494 region in a movement condition (Klopp et al. 2001). Klopp and colleagues suggested that the early ERS could reflect either

495 functional inhibition, in our case movement inhibition or movement preparation, whereas the late peri-Rolandic ERD represents a

496 state of transient functional activation. Klimesch et al. (2007) demonstrated that alpha ERS is elicited in situations where people

497 withhold or control the execution of a response and is obtained over sites that probably exert top-down control, for instance

498 corticospinal control of motorneurons. Thus, they proposed that alpha ERS reflects a topdown inhibitory control processes and

499 concluded that alpha ERS plays an active role for the inhibitory control and timing of cortical processing. Whereas ERD reflects

500 the gradual release of inhibition, associated with the emergence of complex spreading activation processes; therefore, a 7-12 Hz

501 ERS should be present during the preparation phase of all motor tasks. However when compared to preparation to rest, we found

502 mainly significant ERS during preparation for MI (centro-parietal electrodes), preparation for active movement (SMA and centro-

503 parietal electrodes) and preparation for BCI (minimally in contralateral fronto-central electrodes). These results follow Desmurget

504 et al (2009) observations that planning or intention of motor actions takes place in parietal cortex while actual movements are

505 produced by activation in the pre-motor cortex, although movement proprioception (feeling the movement) is induced through

506 activating parietal areas.

507 We did not observe the decrease in the mu and/or beta band power described in previous work (Stanack & Pfurtscheller 1996a,b;

508 Derambure 1999, Pfurtscheller 2006c; Pfurstcheller and Neuper 1997), probably due to our comparison to preparing for rest
509 instead of comparing it to resting. This can be seen in Fig.2.A, where a clear ERD before the “GO” cue can be observed. However

510 the goal of our study was to compare the different tasks that can occur during a rehabilitation training and therefore comparing it

511 to the task rest allowed us to subtract changes due to the protocol (mainly cue presentation and timing).

512 During preparation for all motor tasks, (MI, BCI, passive and active movement) we found significant changes in power in the beta

513 frequency band in contralateral post-central electrodes at 2 clearly differentiated narrow frequency bands. This suggests several

514 beta band components during task preparation: a) 18-23Hz power decrease in contralateral electrodes might comprise preparation

515 for proprioception of the movement without active muscle contraction (present before BCI and passive movement only) b) 18-

516 23Hz power increase might correspond to preparation for proprioception of the movement with active muscle contraction (present

517 before active movement only in all electrodes) c) 15-18Hz power increase in medio-post-central electrodes (present before MI,

518 BCI and active movement only) could correspond to preparation for a volitional motor action maintaining hand posture

519 (Hatsopoulos & Donoghue, 2009).

520

521

522 Onset or first 200 msec after the “GO” cue presentation:

523 The contralateral decrease in power in the alpha frequency range at MI onset continued until the end of the task, while in all the

524 other motor tasks there was no significant alpha power change at onset. This might be a result of neural processes involving motor

525 afferents. On the other hand, in the beta frequency range all motor tasks but active movement presented significant power

526 decrease at task onset. All these significant power decreases in beta oscillations (22-26 Hz) were observed during passive

527 movement, MI and BCI suggesting a simultaneous involvement of passive proprioception related neural networks during

528 kinaesthetic MI onset.

529

530 During the task (excluding the first 200 msec):

531 During the execution of all the tasks we observed a significant alpha power decrease spatial distribution when compared to rest.

532 “significance diagonal”. Contra-post-central, medio-central and ipsi-pre-central electrodes in motor areas comprise this spatial

533 distribution of significant alpha power changes being in line with previous work (Pfurtscheller & Berghold, 1989; Toro et al.,

534 1994; Pfurtscheller et al., 2000, Babiloni et al. 1999; Crone et al. 1998). However, the ipsilateral activation has not been

535 investigated further.

536 During MI, BCI and passive movement tasks when compared to rest ERD starts to be significant in parietal areas earlier during

537 the execution of the task in comparison to central areas and in central areas in comparison with frontocentral areas. Significant
538 decrease in power appearance occurred at the same time in contralateral electrodes. This might be interpreted as a propagation

539 effect, referring only to the onset of significant alpha ERD when MI, BCI and passive movement tasks are compared to the rest

540 task.

541 In Fig 8 we can see that the alpha ERD is continuous during the BCI task after the onset. However, these alpha power changes are

542 not marked as significant continuously. Nevertheless, this effect needs further investigation and analysis to prove its significant

543 appearance.

544 Topographical analysis of the phase of ongoing oscillations between recording sites has led to the detection of ‘traveling’ waves,

545 best documented for alpha frequencies moving in a task dependent manner e.g. from anterior to posterior sites (Sauseng et al.,

546 2005b; for a review cf. Nunez et al., 2001; Silberstein, 1995, Takahashi et al., 2011). The fact that ‘travel’ speed is in the range of

547 neural transmission (Schack et al., 2003) indicates that spreading direction of alpha reflects communication between brain areas.

548 This suggests that the observed significant ipsilateral alpha wave decrease in power starting in posterior sites earlier and in

549 anterior sites later during passive movements, MI and BCI task, might indicate sensory information propagating from parietal to

550 motor and pre-motor areas and not motor intention or planning, because the delayed towards frontal electrodes latency of

551 significant alpha power changes is not present during active movement; a conclusion again supporting previous findings

552 (Schwartz 1994, 2004). During a BCI task we could expect a similar effect to the one observed during active movement because

553 the movements of the hand with the orthosis, where controlled by motor related oscillations on the contralateral hemisphere as in

554 the active movement task, however this was not the case probably due to delay between efferent (brain command) and afferent

555 (BCI guided orthosis movement) activity. The fact that during motor imagery (MI and BCI) and passive movement the brain does

556 not use precise somatosensory feedback, could delay the appearance of significant alpha power decrease in ipsilateral brain areas

557 via uncrossed afferent fibers arriving directly to pre-central central areas and post-central areas. During active movements the

558 ipsilateral cortex is activated at the same time with movement related sensory feedback.

559 The significant decrease of ipsilateral hemispheric alpha rhythm (6 – 10 Hz) activation during MI might be due to different

560 events: active inhibition of movements, reception of an efference copy from the contralateral hemisphere or inhibitory inputs.

561 Regarding the sustained ipsilateral significant decrease in alpha (8 – 10 Hz) power during active movements, this might be due to

562 a combination of interhemispheric flow (efference copy), active suppression of mirror movements, and an afferent somatosensory

563 input projecting also to the ipsilateral hemisphere via uncrossed fiber tracts (Korvenoja et al., 1995).

564

565 It has been demonstrated that alpha ERD during passive movements disappeared if local nerve block was administered and

566 proprioception of the upper limb was blocked (Chou-Ching K et al., 2003), indicating that this frequency is directly related to
567 movement proprioceptive input. Therefore we should not see significant alpha decrease in power during MI, however we do

568 indeed observe a significant decrease in power in the alpha band during MI, suggesting that either alpha comprises 2 components

569 and one is related to motor volition or there is an afferent recall during kinaesthetic MI.

570 It has been proposed that ERS should be produced for planning of action and ERD during sensory processing (Klimesch et al

571 2012). The mu rhythm is expected to desynchronize over the respective areas of the homunculus, reflecting the downstream

572 modulation of motor neurons (Pineda 2005). But areas (such as the SMA) that are involved in planning of movements should

573 exhibit an increase in power. Indeed with people observing another person doing a complex motor task, Muthukumaraswamy and

574 Johnson (2004) and Muthukumaraswamy et al. (2004) found ERD over sensory-motor areas, but ERS over the SMA. However,

575 we observed alpha band ERS during preparation for active movement, BCI and MI in SMA but not during the execution of these

576 motor tasks. This effect could be due to the need of planning before the task starts, while once the task has started no more

577 planning is needed (too simple task) and only motor correction responding to sensory activity takes place.

578 Neuper et al (2008) demonstrated that abstract feedback (e.g. a thermometer on a screen) from brain oscillations during a BCI task

579 produces similar mu ERD/ERS compared to watching a video of a hand moving. Since our participants watched the hand being

580 moved by the orthosis during the active and passive movements and BCI conditions, the alpha ERD during the task we observed

581 can be a combination of “mirror movement” and proprioceptive produced activity.

582

583 It has been suggested that different dynamics in the temporal pattern of oscillatory change in alpha and beta bands indicate that

584 parallel processes may be occurring in the neuronal network configurations involved in generating these rhythms during

585 movement, with those generating beta rhythms having more of a motor component, while those generating alpha rhythms having

586 more of a somatosensory component (Jurkiewicz et al. 2006). In line with these results, Patino et al. (2006) showed that with

587 chronic deafferentation, there is a larger ERD during active movement performance in the alpha rhythm only and speculated to be

588 an effort-related effect since it should be more difficult to perform a movement without sensory feedback, whereas the beta

589 dynamics remained unaffected after chronic deafferentation and in transient ischemia (Schnitzler et al., 1997). Although in these

590 studies it was not possible to distinguish whether the stronger alpha power decrease was due to increased effort or due to the

591 removal of the expected sensory input i.e. larger prediction error requiring increased neural processing, it was shown that the

592 removal of afferent feedback does not significantly influence the strength of beta dynamics and at least does not weaken the alpha

593 rhythm dynamics during active movements. However in Schnitzler’s et al. work a complete alpha range was used in the analysis,

594 ignoring the different sub-components within alpha frequency range, therefore it could have affected the motor component of the

595 alpha rhythm only. Our data shows larger significant beta decrease in power during passive movement, MI and BCI than during
596 active movement, suggesting a more inhibitory activity modulated by sensory input in the beta frequency range i.e. the participant

597 feels or imagines feeling the hand moving and inhibits muscle contraction.

598 However, results from human electroencephalographic studies suggest that an increase (and not a decrease) in power in the beta

599 range is associated with inhibition of the excitatory state of the motor cortex (Gilbertson et al 2005). There is also clinical

600 evidence regarding the origin of this inhibition in patients with frontal lobe damage that exhibit 'unwilled' automatic movements

601 (Archibald et al. 2001). These clinical studies suggest that the prefrontal, anterior cingulate, and supplementary motor cortices

602 may contribute the necessary inhibition to prevent triggering of movement commands generated in activated motor and premotor

603 cortical areas. Following these studies we should have observed beta ERD in these areas during active movement. However,

604 during active movement only sparse significant decrease in power were observed in the low beta frequency range mainly at

605 medio-central and ipsilateral electrodes as opposed to the majority of the literature results (Pfurtscheller& Lopes da Silva, 1999,

606 Pfurtscheller and Neuper 1994,1997, Alegre et al. 2004, Stanack and Pfurtscheller 1996a,b, Derambere et al. 1999). In our study

607 we used an extra resting task with the same protocol (warning cue with instructions “Rest”, imperative cue “GO” and “end cue”)

608 and compared it time point by time point and frequency band by frequency band as opposed to the baseline used in those studies

609 (normally an average from each frequency band during some seconds before the task) and this could have influenced the

610 statistical comparison.

611

612 We observed several spectral components only related to: a) volition (i.e. all volitional tasks - BCI, MI, Active movement -

613 showing the same activation; b) passive proprioception only (i.e. without muscle contraction - BCI and passive movement -

614 showing the same activation; c) proprioception (i.e. all proprioceptive tasks -BCI, passive movement and active movement-

615 showing the same activation (See Table 2).

616 The frequency range from 20 to 24 Hz appears to be related to proprioception without muscle contraction since only passive

617 movement and BCI tasks showed significant activity in this band at the time orthosis movement started, therefore, we decided to

618 subdivide the frequency ranges used in Table 1 to analyse these effects more in detail. Furthermore, the BCI condition presented

619 significantly higher activity in this band when compared to passive movement implying that movement imagery linked to passive

620 movements increased the activity in this frequency band in line with previous BCI control results (Ramos-Murguialday et al.,

621 2012). This could be explained following the results observed by Kayser et al. (2009), in which they found feedback interactions

622 during multisensory processing are associated with beta-band activity. However, our results regarding the 20 to 24 Hz band,

623 indicate this somatosensory processing is related to proprioception without muscle contraction, because it was absent during

624 active movement.

625 Beta band activity has been also identified as related to ‘keeping online’ the mapping between somatosensory inputs and motor

626 actions required by the context. Indeed, beta oscillations have been related to top-down attention and sensorimotor decision-

627 making (Siegel et al., 2012), Thus since the active movement task was a simple opening and closing of the hand, this could have

628 been the reason for the absence of more beta activity deysnchronization during active movement and therefore we suggest that

629 beta ERD is related to proprioceptive information processing and/or the inhibition of muscle contractions.

630 Summarizing, we observed alpha and beta significant power differences before, at onset and during the motor tasks that related to

631 passive proprioception or volition only (See Table 2): before the task we observed volition only related ERS (6-10 Hz in

632 contralateral frontal electrodes and passive proprioceptive related ERD (19-24Hz in contralateral electrodes). At task onset we did

633 not observe any specific volition only or proprioception only related ERD or ERS. During the task we observed passive

634 proprioception only related ERD 19-24 Hz in all electrodes.

635
 Time Prior Onset During

Freq. band Elec. Contra. Med. Ipsi. Contra. Med. Ipsi. Contra. Med. Ipsi.

Front. V(ERS) - - - - - - -
6-10
Hz Centr. - - - - - - - - -

Alpha. Post-Centr. - - - - - - - - -
α
Front. - - - - - - - - -
10-14
Hz Centr. - - - - - - - - -

Post-Centr. - - - - - - - - -

Front. - V(ERS) V(ERS) - - - -

15-18
Hz Centr. - - - - - - - - -

Beta Post-Centr. - - - - - - - -
β Front. PP(ERD) - - - - - PP(ERD) PP(ERD) PP(ERD)
19-24
Hz Centr. PP(ERD) - - - - - PP(ERD) PP(ERD) PP(ERD)

Post-Centr. PP(ERD) - - - - - PP(ERD) PP(ERD) PP(ERD)

Table 2. Significant power change related to propriocpetion or volition only. Significant decrease (ERD) and increase (ERS) in power before
(preparation), at onset (planning) and during (execution) for the 4 different motor tasks (Active movement, Passive movement, motor imagery (MI) and
motor imagery with proprioceptive feedback (BCI)) when compared to rest allowed us to identify significant changes in frequency at different electrodes
related to: a) Volition (V), i.e. all volitional tasks (BCI, MI, Active movement) showing the same activation; b) passive proprioception only (PP) i.e. without
muscle contraction (BCI and passive movement); c) proprioception (P), i.e. all proprioceptive tasks (BCI, passive movement and active movement).

636 In addition to the main investigated motor related frequency bands using EEG (alpha and beta), in subdural recordings,

637 contralateral dominant gamma clusters (gamma ERS) have been observed during the execution phase. Our strict filtering of trials

638 containing EMG activity during tasks might have made the detection of gamma activity more reliable. It is important to notice

639 that induced gamma and beta oscillations embedded in desynchronized alpha band activity have been already observed previously

640 (Pfurtscheller et al. 1993, Pfurtscheller and Neuper 1994). We found EEG gamma significant power changes (24-45 Hz) during

641 the task to be concurrent with muscle contraction and co-occurring with slow oscillations (theta and alpha) significant changes in

642 power (See Fig.6). Whilst we have not statistically proven the correlation between theta and gamma activity, recent results using

643 intracranial EEG reported delta and gamma band to contribute significantly more strongly than other frequency bands when

644 decoding muscle activity (Shin et al. 2012) and end point trajectories (Nakanishi et al 2013). Furthermore, it has been shown that

645 theta oscillations modulated power at gamma frequency such that the gamma oscillations co-occur with the incoming sensory

646 information, i.e. during active sensing (Siegel et al., 2012). Therefore, we expected to observe co-occurring gamma and theta

647 oscillations during active movement and BCI task, however this was not the case during BCI.
648 We concluded that we did not observe slow oscillations co-occurring with gamma oscillations during the BCI task because a) The

649 BCI system delay linking afferent and efferent activity was not optimal, as naturally happening during active movement and the

650 slow oscillations could not modulate power at gamma frequency and b) gamma significant changes in power are related to muscle

651 contraction only.

652

653 Another limitation of the study is to what extent affects the feedback in the BCI task the Time-frequency modulations. We cannot

654 discern which of the differences in time-frequency activity between MI and BCI were due to proprioception or to the feedback

655 (visual and proprioceptive) of the BCI, which would strengthen the ERD control and therefore SMR activity as demonstrated

656 previously (Buccino et al 2001, Ramos-Murguialday et al 2012, Lapenta et al 2013).

657 During passive movement and active movement the proprioception is continuous, during BCI (reflected in our results in the theta

658 band during the task), during BCI the control of the orthosis will vary depending on the SMR modulation. Therefore, we cannot

659 conclude that if there is not a modulation at the same time during BCI and passive or active movement there is not a common

660 signature between BCI and these 2 classes (passive and active movement)..

661

662 Relevant for SMR based BCI control of orthoses is the fact that theta (3 - 6 Hz) and alpha (8 - 10 Hz) frequency band power

663 changes during MI, BCI and active movement do not show significant differences. This means that classification of EEG changes

664 due to active movement or MI could be differentiated from passive movement generated desynchronization when using a

665 combination of theta and alpha frequency band features from medio-post-central electrodes on the contralateral hemisphere.

666 Furthermore, our findings showed that active movement could be easily decoded using low gamma band activity after adequate

667 limb and head EMG artefact filtering. We found theta power increase during the trial to be the feature differentiating BCI and

668 active movement from MI and passive movement (significantly higher power during BCI and MI compared to passive movement

669 and MI) and could be related to voluntary hand movement generation.

670

671 V. CONCLUSION

672 We analyzed the cortical energy changes in the alpha, beta and low gamma bands before, at onset and during four motor tasks,

673 from which three of the 4 motor tasks were related to motor rehabilitation (MI, active and passive movement) and were compared

674 with the other motor task (BCI), in which the three of them are mixed. During the BCI task we linked brain activity due to MI of

675 hand opening/closing and the exact same movement via a BCI controlled robotic hand orthosis. We demonstrated that spectral
676 power changes in alpha and beta SMR frequency ranges before and during each of the motor tasks (motor imagery, BCI and

677 active and passive movements) presented low and high components for each band, separating the involvement of these low and

678 high components of alpha and beta band in proprioception and volitional motor commands.

679 Furthermore we demonstrated that each motor task presented EEG signatures, which were significantly different from the other

680 motor tasks, except for passive movement. The BCI task incorporated all EEG signatures from passive movement, while

681 presenting some EEG signatures shared only with active movement and others related only to the BCI task. On the other hand,

682 and somehow opposite to our goal when closing the loop (BCI), we observed that EEG activity during brain control of the orthotic

683 device in the BCI task, did not resemble neural activation generated before (ERS in theta and alpha bands) and during (ERS in

684 beta and low gamma bands) active movement. This was probably due to the absence of muscle contraction during BCI (low

685 gamma ERS), the presence of proprioceptive feedback false positives/negatives, the delay caused by MI classification not being

686 100% correct and because signal processing presents a larger delay compared to natural cortico-muscular activations. However as

687 mentioned above, we found specific significant power changes at frequencies and time windows present during BCI task and

688 active movement only. We observed EEG signatures present during the BCI task only, indicating that the BCI task involves

689 particular neural oscillations not present during MI, passive or active movement tasks. These findings raise the question of the

690 existence of a different brain state when brain-controlling (BCI) afferent pathways (haptic feedback) bypassing subcortical, spinal

691 and peripheral effectors.

692 In this work we pin-pointed EEG signatures related to motor tasks and motor task components involved in motor rehabilitation,

693 which could be used towards the design and development of an EEG based user involvement assessment tool for motor

694 rehabilitation interventions, that could provide the therapist with information regarding the involvement of the patient during the

695 rehabilitation motor training by being able to classify if the patient is not trying to move and only passive components are present

696 or if any volition component is present and the patient is trying to move, but movement is not possible. These features

697 differentiating brain oscillations during motor tasks involved in motor rehabilitation, could contribute substantially in the

698 development of more advanced motor rehabilitation therapies like brain controlled and shared controlled (EEG-EMG) orthotics. It

699 is important to notice that significant power changes were found before the beginning of the trial in all frequency ranges,

700 suggesting the prediction of motor intention is very plausible and therefore so is the development of faster EEG based BCI

701 control.

702 The proprioceptive feedback broadened the significant changes in frequency power mainly in alpha and beta bands and therefore

703 increased BCI performance, biasing it as the power at these frequency bands was used as inputs for the linear classifier. We know

704 from previous work (Ramos-Murguialday et al., 2012, 2013) that healthy participants and patients can learn to control the
705 proprioceptive BCI, but an adequate selection of features could optimize learning by avoiding afferent bias and for this purpose

706 we suggest, following the work of Siegel et al., (2012) in using spectral fingerprints to distinguish between similar EEG activity in

707 different behavioural levels.

708

709 ACKNOWLEDGMENT

710 The authors thank Sebastian Halder, Juergen Mellinger and Juergen Dax for their hard work and dedication towards development

711 of the BCI-Orthosis platform. The authors would also like to thank Eva Maria Hammer and Silvia Kofler for their assistance with

712 the EEG experiments.

713 REFERENCES
714
715 Alegre M, Labarga A, Gurtubay IG, Iriarte J, Malanda A, Artieda J. Beta electroencephalograph changes during passive movements: Sensory afferences contribute
716 to beta event-related desynchronization in humans. Neuroscience Letters 331: 29-32, 2002.
717 Alegre M, Gurtubay IG, Labarga A, Iriarte J, Malanda A, Artieda J. Alpha and beta oscillatory activity during a sequence of two movements. Clinical
718 Neurophysiology 115: 124-130, 2004.
719 Archibald SJ, Mateer CA, Kerns KA: Utilization behavior: clinical manifestations and neurological mechanisms. Neuropsychol Rev 2001, 11:117-130.
720 Babiloni C, Carducci F, Cincotti F, Rossini PM, Neuper C, Pfurtscheller G, Babiloni F. Human movement-related potentials vs desynchronization of EEG alpha
721 rhythm: a high-resolution EEG study. Neuroimage 10(6): 658-65, 1999.
722 Barrett G, Shibasaki H, Neshige R. Cortical potentials preceding voluntary movement: evidence for three periods of preparation in man. Electroencephalogr Clin
723 Neurophysiol. 63: 327-339, 1986.
724 Birbaumer N, Elbert T, Canavan AGM, Rockstroh B. Slow Potentials of the Cerebral Cortex and Behavior 70: 1-40, 1990.
725 Birbaumer, N., Ghanayim, N., Hinterberger, T., Iversen, I., Kotchoubey, B., Kübler, A., Perelmouter, J., Taub, E. & Flor, H. A spelling device for the paralysed.
726 Nature 398: 297-298, 1999.
727 Birbaumer N, Ramos Murguialday A and Cohen L. Brain-Computer-Interface (BCI) in paralysis. In: Current opininon in Neurology 21(6): 634-638, 2008.
728 Burgess A, Gruzelier J. Methodological advances in the analysis of event-related desynchronization data: reliability and robust analysis. In: Handbook of
729 electroencephalography and clinical neurophysiology, revised series, Vol 6 (Pfurtscheller G, Lopes da Silva FH, eds), pp 139–158. Amsterdam Elsevier
730 Science, 2008.
731 Cassim F, Monaca C, Szurhaj W, Bourriez JL, Defebvre L, Derambure P, Guieu JD. Does post-movement beta synchronisation reflect an idling motor cortex?
732 Neuroreport 12(17): 3859-3863, 2001.
733 Cho W, Vidaurre C, Hoffmann U, Birbaumer N, Ramos-Murguialday A; Afferent and Efferent Activity Control in the design of Brain Computer Interfaces for
734 Motor Rehabilitation. Engineering in Medicine and Biology Society (EMBC) Proceedings, 2011 Annual International Conference of the IEEE:7310-5,
735 2011.
736 Chou-Ching K, Ming-Shaung L, Chih-Hsu Huang. Influence of Passive Thumb movements on mu Wave. J. of Medical and Biol. Eng. 23(1): 37-44, 2003.
737 Crone NE, Miglioretti DL, Gordon B, Sieracki JM, Wilson WT, Uematsu S and Lesser RP. Functional mapping of human sensorimotor cortex with
738 electrocorticographic spectral analysis: I Alpha and beta event-related desynchronization. Brain 121:. 2271-2299, 1998.
739 Daubechies I. Where do wavelets come from?---A personal point of view. Proceedings of the IEEE, Special Issue on Wavelets 84 (4): 510—513, 1996.
740 Dechent P, Merboldt KD, Frahm J. Is the primary motor cortex involved in motor imagery? Cognitive Brain Research 19: 138-144, 2004.
741 Deecke L, Scheid P, Kornhuber HH. Distribution of readiness potential, pre-motion positivity, and motor potential of the human cerebral cortex preceding
742 voluntary finger movements. Exp Brain Res. 7: 158-68, 1969.
743 Derambure, P., Defebvre, L., Bourriez, J.L., Cassim, R. & Guieu, J.D. Event-related desynchronization and synchronization. Reactivity of electrocortical rhythms
744 in relation to the planning and execution of voluntary movement. Neurophysiol. Clin. 29: 53–70, 1999.
745 Desmurget M, Reilly KT, Richard N, Szathmari A, Mottolese C, Sirigu A. Movement Intention After Parietal Cortex Stimulation in Humans. Science.324: 811,
746 2009.
747 d Vries S, Mulder T. Motor imagery and stroke rehabilitation: a critical discussion, J Rehabil Med. 39(1): 5-13, 2007.
748 Efron B, and Tibshirani RJ. An Introduction to the Bootstrap. Boca Raton, FL, CRC Press, 1994.
749 Ehrsson, H.H., Kuhtz-Buschbeck, J.P. & Forssberg, H. Brain regions controlling nonsynergistic versus synergistic movement of the digits: a functional magnetic
750 resonance imaging study. J. Neurosci. 22: 5074–5080, 2002.
751 Farry KA,Walker ID,Baraniuk RG. Myoelectric teleoperation of a complex robotic hand. IEEE trans. On Robotics and Automation 12(5): 775-88, 1996.
752 Fritsch B, Reis J, Martinowich K, Schambra HM, Ji Y, Cohen LG, Lu B. Direct current stimulation promotes BDNF-dependent synaptic plasticity: potential
753 implications for motor learning. Neuron 66(2): 198-204, 2010.
754 Gerardin, E., Sirigu, A., Lehe´ricy, S., Poline, J.B., Gaymard, B., Marsault, C., Agid, Y. & Bihan, D.L. Partially overlapping neural networks for real and
755 imagined hand movements. Cereb. Cortex 10: 1093–1104, 2000.
756 Gilbertson T, Lalo E, Doyle L, Di LV, Cioni B, Brown P: Existing motor state is favored at the expense of new movement during 13–35 Hz oscillatory synchrony
757 in the human corticospinal system. J Neurosci 25:7771-7779, 2005.
758 Grezes J, Costes N, Decety J. Top down effect of the strategy to imitate on the brain areas engaged in perception of biological motion: a PET investigation.
759 Cognitive Neuropsychology 15: 553–582, 1998.
760 Grezes J, and Decety J. Functional Anatomy of Execution, Mental Simulation, Observation, and Verb Generation of Actions: A Meta-Analysis. Human Brain
761 Mapping 12: 1-19, 2001.
762 Halder S, Agorastos D, Veit R, Hammer EM, Lee S, Varkuti B, Bogdan M, Rosenstiel W, Birbaumer N, Kubler A. Neural mechanisms of brain-computer
763 interface control. Neuroimage 55(4): 1779-90, 2011.
764 Hatsopoulos NG, Donoghue JP. The science of neural interface systems. Annu Rev Neurosci. 32: 249-66, 2009.
765 Hogan N, Krebs HI, Rohrer B, Palazzolo JJ, Dipietro L, et al. Motions or muscles? Some behavioral factors underlying robotic assistance of motor recovery. J
766 Rehabil Res Dev 43: 605-618, 2006.
767 Hummel F, Andres F, Altenmüller E, Dichgans J, Gerloff C. Inhibitory control of acquired motor programmes in the human brain. Brain:125: 404-420, 2002.
768 Hu XL, Tong KY, Song R, Zheng XJ, Leung WW. A comparison between electromyography-driven robot and passive motion device on wrist rehabilitation for
769 chronic stroke. Neurorehabil Neural Repair 23: 837-846, 2009.
770 Jasper, H.H. & Penfield, W. Electrocorticograms in man: effect of the voluntary movement upon the electrical activity of the precentral gyrus. Arch. Psychiatr. Z
771 Neurol. 183: 163–174, 1949.
772 Jeannerod, M. & Decety, J. (1995) Mental motor imagery: a window into the representational stages of action. Curr. Opin. Neurobiol. 5: 727–732, 1995.
773 Jeannerod M, and Frak V. Mental Imaging of Motor Activity in Humans. Current Opinion in Neurobiology 9: 735-739, 1999.
774 Jackson PL, Lafleur MF, Malouin F, Richards CL, Doyon J. Functional cerebral reorganization following motor sequence learning through mental practice with
775 motor imagery. Neuroimage 20: 1171-1180, 2003.
776 Jurkiewicz MT, Gaetz WC, Bostan AC, Cheyne D. Post-movement beta rebound is generated in motor cortex: Evidence from neuromagnetic recordings.
777 NeuroImage 32(3): 1281-1289, 2006.
778 Kasess CH, Windischberger C, Cunnington R, Lanzenberger R, Pezawas L, Moser E. The suppressive influence of SMA on M1 in Motor imagery revealed by
779 fMRI and dynamic causal modeling. Neuroimage 40: 828-837, 2008.
780 Kayser, C. & Logothetis, N. K. Directed interactions between auditory and superior temporal cortices and their role in sensory integration. Front. Integr Neurosci.
781 4: 3-7, 2009.
782 Keinrath C, Wriessnegger S, Mueller-Putz GR, Pfurtscheller G. Post-movement beta synchronization after kinaesthetic illusion, active and passive movements.
783 Int. Journal of Psychophysiology 62: 321-327, 2006.
784 Klimesch W. Alpha-band oscillations, attention, and controlled access to stored information. Trends in Cogn. Scien 16(12): 606-617, 2012.
785 Klimesch W, Sauseng P and Hanslmayr S. EEG alpha oscillations: The inhibition-timing hypothesis.Brain Research Reviews 53: 63-88, 2007.
786 Klopp J, Marinkovic K, Clarke J, Chauvel P, Nenov V, Halgren E. Timing and Localization of Movement-Related Spectral Changes in the Human Peri-Rolandic
787 Cortex: Intracranial Recordings. Neuroimage14(2): 391-405, 2001.
788 Koralek AC, Jin X, Long JD, Costa RM, Carmena JM. Corticostriatal plasticity is necessary for learning intentional neuroprosthetic skills. Nature 483(7389): 331-
789 5, 2012.
790 Korvenoja A, Wikstrom H, Huttunen J, Virtanan J, Laine P, Aronen HJ, Seppalainen AM, Ilmoniemi RJ. Activation of ipsilateral primary sensorimotor cortex by
791 median nerve stimulation. Neuroreport 6(18): 2589-93, 1995.
792 Lacourse MG, Orr ELR, Cramer SC, Cohen MJ. Brain Activation during Execution and motor imagery of novel and skilled sequential hand movements.
793 Neuroimage 27: 505-519, 2005.
794 Lee BI, Lüders H, Lesser RP, Dinner DS, Morris HH. Cortical potentials related to voluntary and passive finger movements recorded from subdural electrodes in
795 humans. Annual Neurology 20: 32-37, 1986.
796 Leocani, L., Toro, C., Manganotti, P., Zhuang, P. & Hallett, M. Eventrelated coherence and event-related desynchronization ⁄ synchronization in the 10 Hz and 20
797 Hz EEG during self-paced movements. Electroencephalogr. Clin. Neurophysiol. 104: 199–206, 1997.
798 Lotze M, Montoya P, Erb M, Hülsmann E, Flor H, Klose U, Birbaumer N, Grodd W. Activation of Cortical and Cerebellar Motor Areas during Executed and
799 Imagined Hand Movements: An fMRI Study. Journal of Cognitive Neuroscience 11(5): 491-501, 1999.
800 Lotze M, Braun C, Birbaumer N, Anders S, and Cohen LG. Motor learning elicited by voluntary drive. Brain 126: 866–872, 2003.
801 Maris E, Oostenveld R. Nonparametric statistical testing of EEG- and MEG-data. J Neurosci Methods 164:177-190, 2007.
802 Mima T, Sadato N, Yazawa S, Hanakawa T, Fukuyama H, Yonekura Y, Shibasaki H. Brain structures related to active and passive finger movements in man.
803 Brain 122: 1989-1997, 1999.
804 Müller GR, Neuper C, Keinrath C, Gerner HJ, Pfurtscheller G. Event-related beta EEG changes during wirst movements induced by functional electrical
805 stimulation of forearm muscles in man. Neurscience Letters 340: 143-147, 2003.
806 Müller-Putz GR, Zimmermann D, Graimann B, Nestinger K, Korisek G, Pfurtscheller G. Event-related beta EEG-changes during passive and attempted foot
807 movements in paraplegic patients. Brain Research 1137: 84-91, 2007.
808 Muthukumaraswamy, S.D., Johnson, B.W. Changes in rolandic mu rhythm during observation of a precision grip. Psychophysiology 41: 152–156, 2004.
809 Muthukumaraswamy, S.D., Johnson, B.W., McNair, N.A. Mu rhythm modulation during observation of an object-directed grasp. Cogn. Brain Res. 19: 195–201,
810 2004.
811 Naito, E. Sensing limb movements in the motor cortex: how humans sense limb movement. Neuroscientist 10(1): 73–82, 2004.
812 Naito, E., Ehrsson, H.H. Kinesthetic illusion of wrist movement activates motor-related areas. Neuroreport 12: 3805–3809, 2001.
813 Nakanishi Y, Yanagisawa T, Shin D, Fukuma R, Chen C, Kambara H, Yoshimura N, Hirata M, Yoshimine T, and Koike Y. Prediction of Three-Dimensional Arm
814 Trajectories Based on ECoG Signals Recorded from Human Sensorimotor Cortex. PLoS One 8: e72085, 2013.
815 Neuper, C. & Pfurtscheller, G. Motor imagery and ERD. In Pfurtscheller, G. & Lopes da Silva, F.H. (Eds), Event-Related Desynchronization and Related
816 Oscillatory Phenomena of the Brain. Handbook of Electroencephalography and Clinical Neurophysiology, revised Edn., Vol. 6. Elsevier, Amsterdam, pp.
817 303–325, 1999.
818 Neuper C, Scherer R, Reiner M, Pfurtscheller G. Imagery of motor actions: Differential effects of kinaesthetic and visual-motor mode of imagery in single-trial
819 EEG. Cognitive Brain Research 25: 668-677, 2005.
820 Neuper C, Scherer R, Wriessnegger S, Pfurtscheller G. Moor imagery and action observation: odulation of sensorimotor brain rhythms during mental control of a
821 brain computer interface. Clin Neurophysiol. 120(2): 239-47, 2008.
822 Ohara S, Ikeda A, Kunieda T, Yazawa S, Baba K, Nagamine T, Taki W, Hashimoto N, Mihara T, and Shibasaki H. Movement-related change of
823 electrocorticographic activity in human supplementary motor area proper. Brain 123:1203-1215, 2000.
824 Oostenveld R, Fries P, Maris E, Schoffelen JM. FieldTrip: Open source software for advanced analysis of MEG, EEG, and invasive electrophysiological data.
825 Comput Intell Neurosci 2011, Article ID 156869, 9 pages, 2011. doi:10.1155/2011/156869, 2011.
826 Patino L, Chakarov V, Schulte-Mönting J, Hepp-Reymond MC, Kristeva R. Oscillatory cortical activity during a motor task in a deafferented patient. Neurosci
827 Lett. 401(3): 214-8, 2006.
828 Pfurtscheller G, and Berghold A. Patterns of cortical activation during planning of voluntary movement. Electroencheph clin Neurophysiol 72: 250-258, 1989.
829 Pfurtscheller G, Neuper C and Kalcher J. 40-Hz oscillation during motor behavior in man. Neurosci Lett 164: 179-182, 1993.
830 Pfurtscheller G, and Neuper C. Event-related synchronization of mu rhythm in the EEG over the cortical hand area in man. Neurosci Lett 174: 93-96, 1994.
831 Pfurtscheller G, and Neuper C. Motor Imagery activates Primary Sensorimotor area in Humans. Neuroscience Letters 239: 65-68, 1997.
832 Pfurtscheller G, and Lopes da Silva FH. Event-related desynchronization, Handbook of electroencephalography and clinical neurophysiology. Vol. 6, Elsevier,
833 Amsterdam, 1999.
834 Pfurtscheller G, Neuper C, and Krausz G. Functional dissociation of lower and upper frequency mu rhythms in relation to voluntary limb movement. Clin
835 Neurophysiol 111: 1873-1879, 2000.
836 Pfurtscheller G, Woertz M, Supp G and Lopes da Silva FH. Early onset of post-movement beta synchronization in the supplementary motor area (SMA) during
837 self-paced finger movement in man, Neurosci Lett. , 2003.
838 Pfurtscheller G, Neuper C. Future prospects of ERD/ERS in the context of brain-computer interface (BCI) developments. Prog Brain Res. 159: 433-437, 2006.
839 Pfurtscheller, G., Brunner, C., Schlo¨ gl, A. & Lopes da Silva, F.H. Mu rhythm (de)synchronization and EEG single-trial classification of different motor imagery
840 tasks. NeuroImage 31: 153–159, 2006c.
841 Pineda, J.A. The functional significance of mu rhythms: translating “seeing” and “hearing” into “doing”. Brain Res. Brain Res. Rev. 50: 57–68, 2005.
842 Ramos-Murguialday A; Soares E; Birbaumer N; Upper Limb EMG Artefact Rejection in Motor Sensitive BCIs; Engineering in Medicine and Biology Society
843 (EMBC) Proceedings, 2010 Annual International Conference of the IEEE 2010:1 – 6, 2010
844 Ramos-Murguialday A; Schürholz M, Caggiano V, Wilgruber M, Caria A, Hammer EM, Halder S, Birbaumer N. Propriocpetive feedback and brain computer
845 interface (BCI) based neuroprostheses. PLoS ONE 7(10): e47048, 2012.
846 Ramos-Murguialday A, Broetz D, Rea M, Laeer L, Yilmaz O, Brasil FL, Liberati G, Curado MR, Garcia-Cossio E, Vyziotis A, Cho W, Agostini M, Soares E,
847 Soekadar S, Caria A, Cohen LG, Bibaumer N. Brain-Machine-Interface in Chronic Stroke Rehabilitation: A Controlled Study. Annals of Neurology 74:
848 100-108, 2013
849 Salmelin, R. & Hari, R. Spatiotemporal characteristics of sensorimotor neuromagnetic rhythms related to thumb movement. Neuroscience 60: 537– 550, 1994.
850 Schnitzler A, Salenius S, Salmelin R, Jousmäki V, Hari R. Involvement of primary motor cortex in motor imagery: a neuromagnetic study. Neuroimage 6(3): 201-
851 8, 1997.
852 Schwartz AB. Distributed motor processing in cerebral cortex. Cur Opinion in Neurobiology 4(6): 840-846, 1994.
853 Scwartz AB. Cortical Neural Prosthetics. Annu. Rev. Neurosci. 27: 487-507, 2004.
854 Sehm B, Perez MA, Xu B, Hidler J, Cohen LG. Functional neuroanatomy of mirroring during a unimanual force generation task. Cereb Cortex 20(1): 34-45, 2010.
855 Sharma N, Pomeroy VM, Baron JC. Motor Imagery: A backdoor to the motor system after stroke? Stroke 37: 1941-1952, 2006.
856 Shin D, Watanabe H, Kambara H, Nambu A, Isa T, Nishimura Y, and Koike Y. Prediction of muscle activities from electrocorticograms in primary motor cortex
857 of primates. PLoS One 7: e47992, 2012.
858 Siegel M, Donner TH, Engel AK. Spectral fingerprints of large-scale neuronal intercations. Nature Reviews 13:121:131, 2012.
859 Stancak, A. Jr & Pfurtscheller, G. The effects of handedness and type of movement on the contralateral preponderance of mu-rhythm desynchronisation. Clin.
860 Neurophysiol. 99: 174–182, 1996a.
861 Stancak, A. & Pfurtscheller, G. Event-related desynchronisation of central beta-rhythms during brisk and slow self-paced finger movements of dominant and
862 nondominant hand. Brain Res. Cogn. Brain Res. 4: 171–183, 1996b.
863 Takahashi K, Saleh M, Penn RD, Hatsopoulos NG. Propagating waves in human motor cortex. Front. Hum. Neurosci. 5(40): 1-8, 2011.
864 Toro C, Deuschl G, Thatcher R, Sato S, Kufta C and Hallett M. Event-related desynchronization and movement-related cortical potentials on the ECoG and EEG.
865 Electroenceph. clin. Neurophysiol. 93: 380-389, 1994.
866 Weiller, C., Juptner, M., Fellows, S., Rijntjes, M., Leonhardt, G., Kiebel, S., Muller, S., Diener, H.C., Thilmann, A.F., Brain representation of active and passive
867 movements. NeuroImage 4: 105–110, 1996.
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883 Figure1. Experimental Protocol. A) Channel CP3 time-frequency event related spectral perturbation (ERSP) during the BCI task. We divided our analysis in
884 three time periods: preparation (-1500 to 0 msec); onset (0 to 200 msec; shaded in light grey); during task (200 to 4500 msec); B) EEG 128 channel cap used for
885 the recordings. Shaded in green the 61 electrodes used during the experiments over the motor areas. C) Close look at the Orthosis with the fingers attached.
886 Bowden cables pull and push participants fingers (open and close the hand) depending on the task to execute.

887 Figure 2. Significance Mask. A) Time-Frequency plots for the active movement task (baseline removed using Bootstrap analysis). B) Time-Frequency plots for
888 the active movement task (baseline removed using Bootstrap analysis). C) Difference between active movement and rest (Rest-Active power values) tasks using a
889 common baseline. D) Significance mask obtained during the bootstrap analysis comparing active movement and rest tasks. Everything not in green is statistically
890 significant (p<0.01). The vertical dashed line indicates the time in which the “GO” cue was presented. Red color indicates desynchronization (DS) (power
891 decrease in decibels (dB)) and blue synchronization (S) (power increase in dB) during BCI compared to EEG activity during the rest condition as expressed in the
892 color bar.

893 Figure 3. Motor Imagery Vs Rest spectral power. In the upper panel of the figure, we can see the bootstrap analysis (grand average) comparing motor imagery
894 (MI) task and rest of 52 electrodes from the 61 used (for clarity 9 electrodes were excluded from the figure). Everything not in green is statistically significant
895 (p<0.01). The vertical dashed line indicates the time in which the “Go” cue was presented. Red color indicates desynchronization (DS) (power decrease in decibels
896 (dB)) and blue synchronization (S) (power increase in dB) during MI compared to EEG activity during the rest condition as expressed in the color bar. In the lower
897 panel of the figure we can see enlarged views of different channels time-frequency power difference between these 2 classes.

898 Figure 4. Bootstrap analysis (grand average) comparing motor imagery with feedback (BCI task) and rest. In the upper panel of the figure, we can see the
899 bootstrap analysis (grand average) comparing the BCI task and rest of 52 electrodes from the 61 used (for clarity 9 electrodes were excluded from the figure).
900 Everything not in green is statistically significant (p<0.01). The vertical dashed line indicates the time in which the “Go” cue was presented. Red color indicates
901 desynchronization (power decrease in decibels (dB)) and blue synchronization (power increase in dB) during BCI compared to EEG activity during the rest
902 condition as expressed in the color bar. In the lower panel of the figure we can see enlarged views of different channels time-frequency power difference between
903 these 2 classes.

904 Figure 5. Bootstrap analysis (grand average) comparing passive movement and rest. In the upper panel of the figure, we can see the bootstrap analysis (grand
905 average) comparing the passive movement task and rest of 52 electrodes from the 61 used (for clarity 9 electrodes were excluded from the figure). Everything not
906 in green is statistically significant (p<0.01). The vertical dashed line indicates the time in which the “Go” cue was presented. Red color indicates
907 desynchronization (power decrease in decibels (dB)) and blue synchronization (power increase in dB) during BCI compared to EEG activity during the rest
908 condition as expressed in the color bar. In the lower panel of the figure we can see enlarged views of different channels time-frequency power difference between
909 these 2 classes.

910 Figure 6. Bootstrap analysis (grand average) comparing active movement and rest. In the upper panel of the figure, we can see the bootstrap analysis (grand
911 average) comparing the active movement task and rest of 52 electrodes from the 61 used (for clarity 9 electrodes were excluded from the figure). Everything not
912 in green is statistically significant (p<0.01). The vertical dashed line indicates the time in which the “Go” cue was presented. Red color indicates
913 desynchronization (power decrease in decibels (dB)) and blue synchronization (power increase in dB) during BCI compared to EEG activity during the rest
914 condition as expressed in the color bar. In the lower panel of the figure we can see enlarged views of different channels time-frequency power difference between
915 these 2 classes. Below EEG electrodes in the topographical view raw EMG activity during active movement task. We can observe the significant activity in
916 gamma and mu rhythm to correspond to the presence of muscle recruitment.

917 Figure 7. Active Movement Vs BCI spectral power comparison. Bootstrap analysis (grand average) comparing motor imagery with feedback (BCI) task and
918 active movement. Everything not in green in statistically significant (p<0.01). The vertical dashed line indicates the time in which the “Go” cue was presented to
919 the participants. Red color indicates desynchronization (DS) (power decrease in decibels (dB)) and blue synchronization (S) (power increase in dB) during BCI
920 compared to EEG activity during the active moevement condition as expressed in the color bar. Below the last channel line (parietal electrodes) we plotted raw
921 EMG activity acquired on top of the extensor digitormum muscle during the active movement condition. The EMG amplitude increase corresponds with gamma
922 significnat higher power in the low gamma region during active movement compared to the BCI condition.

923 Figure 8. BCI Vs Rest spectral power comparison without significance mask. Time-frequency topographical plot of the difference between active movement
924 and rest (Rest-Active power values) tasks using a common baseline. The common baseline (-1.5 to -1.3 sec) has been subtracted from each task time-frequency
925 before subtraction between tasks was done. Everything not in green in statistically significant (p<0.01) compared to the common baseline. The vertical dashed line
926 indicates the time in which the “Go” cue was presented to the participants. Red color indicates desynchronization (DS) (power decrease in decibels (dB)) and blue
927 synchronization (S) (power increase in dB) during BCI compared to EEG activity during rest as expressed in the color bar.

928 Figure 9. Spectral time-frequency measures statistical comparison. In the upper panel we can see a topographical distribution of time-frequency plots for the
929 difference in power between the BCI and the resting class. A) shows the time-frequency spectral perturbations data; B) time-frequency spectral perturbations
930 shown in A but using a significance mask clacultaed using bootstraping techniques and corrected for multiple comparisons. Everything not in green is statistically
931 significant.; C) and D) enlarged channel CP3 time-frequency spectra without and with significance mas respectively. The vertical dashed line indicates the time in
932 which the “GO” cue was presented. ERSP stands for event related spectral power perturbation measured in decibels (dB)difference during BCI task compared to
933 EEG activity during the rest condition. The areas contoured by a black line represent significant activity spectral changes before multiple comparison correction.

934
935
936
937
938
939
940
941

Figure1. Experimental Protocol. A) Channel CP3 time-frequency event related spectral perturbation (ERSP) during the BCI task. We divided our analysis in
three time periods: preparation (-1500 to 0 msec); onset (0 to 200 msec; shaded in light grey); during task (200 to 4500 msec); B) EEG 128 channel cap used
for the recordings. Shaded in green the 61 electrodes used during the experiments over the motor areas. C) Close look at the Orthosis with the fingers attached.
Bowden cables pull and push participants fingers (open and close the hand) depending on the task to execute.

942
943
944
945
946
947

Figure 2. Significance Mask. A) Time-Frequency plots for the active movement task (significant “p≤0.01”power chnages compared to baseline are masked in
green after bootstrapping). B) Time-Frequency plots for the resting task (significant “p≤0.01”power chnages compared to baseline are masked in green after
bootstrapping). C) Difference between active movement and rest (Rest-Active power values) tasks using a common baseline. D) Significance mask obtained
during the bootstrap analysis comparing active movement and rest tasks. Everything not in green is considered to be statistically significant (p<0.01) without
correcting for multiple comparisons. The vertical dashed line indicates the time in which the “GO” cue was presented. Red color indicates desynchronization
(DS) (power decrease in decibels (dB)) and blue synchronization (S) (power increase in dB) during BCI compared to EEG activity during the rest condition as
expressed in the color bar (same for all subplots).

948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975

Figure 3. Motor Imagery Vs Rest spectral power. In the upper panel of the figure, we can see the bootstrap analysis (grand average) comparing motor
imagery (MI) task and rest of 52 electrodes from the 61 used (for clarity 9 electrodes were excluded from the figure). Everything not in green is statistically
significant (p<0.01). The vertical dashed line indicates the time in which the “Go” cue was presented. Red color indicates desynchronization (DS) (power
decrease in decibels (dB)) and blue synchronization (S) (power increase in dB) during MI compared to EEG activity during the rest condition as expressed in
the color bar. In the lower panel of the figure we can see enlarged views of different channels time-frequency power difference between these 2 classes.
976
977
978
979
980
981
982
983
984
985

Figure 4. Bootstrap analysis (grand average) comparing motor imagery with feedback (BCI task) and rest. In the upper panel of the figure, we can see
the bootstrap analysis (grand average) comparing the BCI task and rest of 52 electrodes from the 61 used (for clarity 9 electrodes were excluded from the
figure). Everything not in green is statistically significant (p<0.01). The vertical dashed line indicates the time in which the “Go” cue was presented. Red
color indicates desynchronization (power decrease in decibels (dB)) and blue synchronization (power increase in dB) during BCI compared to EEG activity
during the rest condition as expressed in the color bar. In the lower panel of the figure we can see enlarged views of different channels time-frequency power
difference between these 2 classes.

986
987
988
989
 990
991
992
993
994
995
996
997
998
999
1000
1001
1002

Figure 5. Bootstrap analysis (grand average) comparing passive movement and rest. In the upper panel of the figure, we can see the bootstrap analysis
(grand average) comparing the passive movement task and rest of 52 electrodes from the 61 used (for clarity 9 electrodes were excluded from the figure).
Everything not in green is statistically significant (p<0.01). The vertical dashed line indicates the time in which the “Go” cue was presented. Red color indicates
desynchronization (power decrease in decibels (dB)) and blue synchronization (power increase in dB) during BCI compared to EEG activity during the rest
condition as expressed in the color bar. In the lower panel of the figure we can see enlarged views of different channels time-frequency power difference between
these 2 classes.

1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017

Figure 6. Bootstrap analysis (grand average) comparing active movement and rest. In the upper panel of the figure, we can see the bootstrap analysis (grand
average) comparing the active movement task and rest of 52 electrodes from the 61 used (for clarity 9 electrodes were excluded from the figure). Everything not
in green is statistically significant (p<0.01). The vertical dashed line indicates the time in which the “Go” cue was presented. Red color indicates
desynchronization (power decrease in decibels (dB)) and blue synchronization (power increase in dB) during BCI compared to EEG activity during the rest
condition as expressed in the color bar. In the lower panel of the figure we can see enlarged views of different channels time-frequency power difference between
these 2 classes. Below EEG electrodes in the topographical view raw EMG activity during active movement task. We can observe the significant activity in
gamma and mu rhythm to correspond to the presence of muscle recruitment.

1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028

Figure 7. Active Movement Vs BCI spectral power comparison. Bootstrap analysis (grand average) comparing motor imagery with feedback (BCI) task
and active movement. Everything not in green in statistically significant (p<0.01). The vertical dashed line indicates the time in which the “Go” cue was
presented to the participants. Red color indicates desynchronization (DS) (power decrease in decibels (dB)) and blue synchronization (S) (power increase in
dB) during BCI compared to EEG activity during the active moevement condition as expressed in the color bar. Below the last channel line (parietal
electrodes) we plotted raw EMG activity acquired on top of the extensor digitormum muscle during the active movement condition. The EMG amplitude
increase corresponds with gamma significnat higher power in the low gamma region during active movement compared to the BCI condition.

1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041

Figure 8. BCI Vs Rest spectral power comparison without significance mask. Time-frequency topographical plot of the difference between active
movement and rest (Rest-Active power values) tasks using a common baseline. The common baseline (-1.5 to -1.3 sec) has been subtracted from each task
time-frequency before subtraction between tasks was done. Everything not in green in statistically significant (p<0.01) compared to the common baseline. The
vertical dashed line indicates the time in which the “Go” cue was presented to the participants. Red color indicates desynchronization (DS) (power decrease in
decibels (dB)) and blue synchronization (S) (power increase in dB) during BCI compared to EEG activity during rest as expressed in the color bar.

1042
1043
1044
1045
 Figure 9. Spectral time-frequency measures statistical comparison. In the upper panel we can see a topographical distribution of time-frequency plots for
the difference in power between the BCI and the resting class. A) shows the time-frequency spectral perturbations data; B) time-frequency spectral
perturbations shown in A but using a significance mask clacultaed using bootstraping techniques and corrected for multiple comparisons. Everything not in
green is statistically significant.; C) and D) enlarged channel CP3 time-frequency spectra without and with significance mas respectively. The vertical dashed
line indicates the time in which the “GO” cue was presented. ERSP stands for event related spectral power perturbation measured in decibels (dB)difference
during BCI task compared to EEG activity during the rest condition. The areas contoured by a black line represent significant activity spectral changes before
multiple comparison correction.

1046
1047
1048
1049
1050