R

Es

SAa

b

c

h

•••

a

ARRAA

KBEMF

1

t

Uf

gn

h0

Behavioural Brain Research 273 (2014) 106–115

Contents lists available at ScienceDirect

Behavioural Brain Research

jou rn al hom epage: www.elsev ier .com/ locate /bbr

esearch report

lectrophysiological correlates of mental navigation in blind andighted people

ilvia Erika Kobera,b,∗, Guilherme Wooda,b, Christiane Kampla, Christa Neupera,b,c,nja Ischebecka,b

Department of Psychology, University of Graz, AustriaBioTechMed-Graz, AustriaLaboratory of Brain-Computer Interfaces, Institute for Knowledge Discovery, Graz University of Technology, Austria

i g h l i g h t s

Oscillatory EEG signature of naturalistic navigation in blind and sighted participants differ.Functional reorganization leads to stronger activity in the visual cortex in blind than in sighted participants.Differences between blind and sighted participants were task-specific and cannot be explained by resting-state EEG or motor imagery performance.

r t i c l e i n f o

rticle history:eceived 28 May 2014eceived in revised form 11 July 2014ccepted 14 July 2014vailable online 21 July 2014

eywords:lindnesslectroencephalographyental navigation

unctional reorganization

a b s t r a c t

The aim of the present study was to investigate functional reorganization of the occipital cortex for amental navigation task in blind people. Eight completely blind adults and eight sighted matched con-trols performed a mental navigation task, in which they mentally imagined to walk along familiar routesof their hometown during a multi-channel EEG measurement. A motor imagery task was used as con-trol condition. Furthermore, electrophysiological activation patterns during a resting measurement withopen and closed eyes were compared between blind and sighted participants. During the resting mea-surement with open eyes, no differences in EEG power were observed between groups, whereas sightedparticipants showed higher alpha (8–12 Hz) activity at occipital sites compared to blind participantsduring an eyes-closed resting condition. During the mental navigation task, blind participants showed astronger event-related desynchronization in the alpha band over the visual cortex compared to sightedcontrols indicating a stronger activation in this brain region in the blind. Furthermore, groups showeddifferences in functional brain connectivity between fronto-central and parietal–occipital brain networks

during mental navigation indicating stronger visuo-spatial processing in sighted than in blind people dur-ing mental navigation. Differences in electrophysiological parameters between groups were specific formental navigation since no group differences were observed during motor imagery. These results indi-cate that in the absence of vision the visual cortex takes over other functions such as spatial navigation.

. Introduction

In sighted people, the visual cortex of the brain located inhe occipital lobe is primarily responsible for processing incoming

∗ Corresponding author. Department of Psychology Karl-Franzens University Grazniversitätsplatz 2/III A-8010 Graz Austria Tel.: +43 0 316/380 8497;

ax: +43 0 316/380 9808.E-mail addresses: silvia.kober@uni-graz.at (S.E. Kober),

uilherme.wood@uni-graz.at (G. Wood), christiane.kampl@gmx.at (C. Kampl),euper@tugraz.at (C. Neuper), anja.ischebeck@uni-graz.at (A. Ischebeck).

ttp://dx.doi.org/10.1016/j.bbr.2014.07.022166-4328/© 2014 Elsevier B.V. All rights reserved.

© 2014 Elsevier B.V. All rights reserved.

information coming from the visual path of the brain [1]. How-ever, there is evidence that in the absence of visual input thispart of the brain takes over other functions [2,3]. For instance, inblind people the primary visual cortex is activated by Braille read-ing and other tactile discrimination tasks [4–10], somatosensoryprocessing [11], verbal tasks [8,12], and auditory tasks [13,14]. Thisfunctional reorganization of the occipital cortex in blind people wasinterpreted as a form of cross-modal plasticity and may be involved

in functional compensation [4,9,15]. The role of the visual cortexin spatial navigation, where blind individuals have to compensatemissing visual information for successful navigation through space,is not fully clarified yet. In the present study, we investigated the

rain R

to

iaalnTebcttpisvvtt

ttcttsrapfcb

apebFiataoistgth[tfhpsPkaa

amdb

S.E. Kober et al. / Behavioural B

ask-dependent functional reorganization of the occipital cortexbserved in the blind during mental spatial navigation.

Generally, vision is highly relevant for successful navigation. Fornstance, salient visual cues or landmarks in the environment serves reference points to help recognizing paths and spatial locationsnd enable us to keep track of our movements [16,17]. Neverthe-ess, blind people are able to move independently in space and haveo major problems when localizing and reaching specific targets.here is behavioural evidence that the abilities to recognize trav-lled routes and to represent spatial information are comparable inlind and sighted individuals [18–24]. Apparently, blind people canompensate the missing visual information by using tactile, audi-ory, and olfactory cues, as well as motion-related cues arising fromhe vestibular and proprioceptive systems [21]. For instance, blindeople display higher auditory spatial abilities than sighted people

n an auditory spatial localization task, which indicates auditorypatial compensation in the blind [14,25–28]. Hence, the absence ofisual information is compensated by spatial information obtainedia other senses than vision, which might contribute positively tohe mental mapping of spaces and consequently, to successful spa-ial performance [29,30].

Based on the assumption that blind people rely more on idio-hetic cues and echolocation as well as on other senses than visiono move about independently in space [20], spatial navigation inompletely blind people is thought to involve different brain areashan those engaged in sighted people [21]. In sighted people, func-ional imaging studies determined key regions for navigation andpatial processing in the hippocampus, the medial and right infe-ior parietal cortex, parts of the basal ganglia, the parahippocampusnd the left prefrontal cortex [17,31–33]. Electroencephalogra-hy (EEG) studies showed an increased theta (4–8 Hz) power overronto-central brain regions during spatial processing [34–36]. Inontrast, the neuronal correlates of spatial navigation in completelylind people remain largely elusive.

A few prior neurophysiological studies investigated structuralnd functional correlates of spatial cognition in the brain of blindeople. These studies reported on contradicting results. Differ-nces in brain structure and function associated with spatial tasksetween blind and sighted people were found [5,18,24,37–39].or instance, differences in hippocampal volume [18], differencesn activation patterns in the hippocampus and parahippocampus,s well as the occipital cortex [5,24,39], and differences in func-ional brain connectivity [39] could be observed between blindnd sighted controls during diverse spatial tasks. In contrast, somether studies reported comparable brain activation patterns dur-ng spatial tasks in blind and sighted people [21,40]. These priortudies that investigated brain correlates of spatial cognition inhe blind used rather specific spatial tasks such as tactile navi-ation tasks using the tongue [21], tactile navigation tasks usinghe hand [24,38,39], auditory spatial localization tasks [14,27,28],aptic mental rotation [5], visuo-spatial imagery of grid patterns40], etc. Brain correlates of more ‘conventional’ spatial naviga-ion tasks, in which blind participants should for instance navigaterom a starting point to an endpoint in a specific environment,ave not been investigated so far. Therefore, the first aim of theresent study was to examine electrophysiological correlates ofpatial navigation on familiar routes in blind and sighted people.articipants were instructed to mentally navigate on different wellnown routes of their hometown. There is evidence that neuronalctivation patterns underlying navigation in real and mental spacere similar [40,41].

The second aim of the present study was to investigate EEG

ctivation patterns in sighted and blind participants during restingeasurements with open and closed eyes. General and unspecific

ifferences in brain structure [2,42,43] and activation patterns [44]etween blind and sighted individuals have often been reported in

esearch 273 (2014) 106–115 107

the literature. EEG studies provide strong evidence for differencein EEG frequencies between blind and sighted people during res-ting conditions [45,46] as well as during sleep [47]. Especially thealpha rhythm (8–12 Hz) is uncommon in the blind [45,46,48]. Oneexplanation for reduced or even missing alpha rhythm in the blindmay be that this rhythm is generally due to the spontaneous beat ofneurons in the occipital cortex usually dedicated to activities con-nected to visual feature processing. In sighted individuals, neuronsin the occipital cortex discharge spontaneously at a fixed rate whenthe visual area is unoccupied. This yields in increased alpha powerfor instance during an eyes-closed resting condition. In contrast,when vision is lost, the visual area does not remain unoccupiedand therefore has to become more accessible to incoming informa-tion from the rest of the brain [45,48]. Although there is evidenceof abnormalities in the alpha rhythm in the blind, it is assumedthat blind people generally show alpha oscillations in the EEG (see[45] for a review). Note that prior studies demonstrating differ-ences in the EEG activity during resting measurements betweenblind and sighted individuals were performed towards the middleof the 20th century [45,46,48], except for one recent study usingmagnetoencephalography (MEG) [49]. Recent EEG studies primar-ily focused on differences in EEG activation patterns between blindand sighted individuals during different tasks [5,39,50] indicatingreduced alpha power in the EEG of blind participants [50].

In the present study we investigated possible differences in EEGactivation patterns (absolute and relative power) between blindand sighted participants during resting measurements as well asduring mental navigation on familiar routes. We hypothesize thatblind and sighted participants show different EEG activation pat-terns during rest, particularly in the alpha rhythm [45,46,48,50].According to the literature, sighted participants should generallyshow more pronounced EEG alpha activity than the blind [50].Regarding the mental navigation task, we expect that in comparisonto sighted participants, blind participants should show an increasein activity in occipital brain areas during a spatial task. In line withthe evidence that blind people rely more on tactile and auditorycues during navigation [3,20,21] and also based on the findingsthat the visual cortex of blind people takes over tactile and audi-tory processing [3], we hypothesize that the visual cortex of blindpeople is also involved in spatial cognition. Furthermore, we inves-tigated possible differences in functional brain connectivity duringmental navigation between blind and sighted individuals. Althoughthere is some evidence for differences in connectivity measuresbetween blind and sighted people [39], this analysis is explorative.To investigate the specificity of brain activation patterns in blindand sighted people during mental navigation, we also performed amotor imagery task as a control condition. When increased activa-tion in the visual cortex is specific to the spatial task in the blind,no differences in brain activation patterns between groups shouldbe observed during the motor imagery task.

2. Material and methods

2.1. Participants

Eight blind adults (4 male, 4 female) recruited through localinstitutions for the blind took part in this study. The age of the blindparticipants ranged from 27 to 64 years (M = 45.75 yrs., SE = 5.34yrs.). All blind participants were completely blind. Detailed descrip-tion of the blind participants can be found in Table 1. In all cases,blindness was due to an inherited degenerative eye disease (Retini-

tis Pigmentosa) (N = 4), detached retina (N = 1), Nervus OpticusAtrophy (N = 2), or Glaucoma (N = 1). Preliminary analysis revealedno significant differences in EEG activation patterns between con-genitally blind participants and participants who became blind

108 S.E. Kober et al. / Behavioural Brain Research 273 (2014) 106–115

Table 1Characteristics of the blind participants.

Blind participant Age (years) Gender Handedness Mobility/navigational aids Aetiology of blindness Onset of blindness (in years)

1 62 F L White cane Nervus Opticus Atrophy Birth2 45 F R White cane, guide dog Retinitis Pigmentosa 203 46 M R White cane, guide dog Retinitis Pigmentosa 304 27 F R White cane Nervus Opticus Atrophy Birth5 33 M R White cane Glaucoma 13

cane

cane

cane,

ltTcaancow[

2

2

cptpf

wsqqsnttdfighNw

laermonra

2

iwtta

6 64 M R White7 29 M R White8 60 F R White

ater. All blind participants were mobile and navigated throughheir hometown (Graz, Austria) autonomously on a regular basis.he control group consisted of eight volunteers with normal ororrected to normal vision matched individually for gender, agend education (4 male, 4 female, age range 25–62 years, meange M = 45.75 yrs., SE = 5.00 yrs.). All participants had no history ofeurological illness and gave written informed consent. The ethicsommittee of the University of Graz, Austria approved all aspectsf the present study. The experiment was performed in accordanceith the ethical standards laid down in the Declaration of Helsinki

51].

.2. Experimental tasks

.2.1. Mental navigation taskFor the mental navigation task, 15 different routes of the inner

ity of Graz, Austria were used. The different routes were of com-arable complexity and length (between 190 and 250 m) withhe same amount of junctions and decision points and comprisedrominent landmarks of the city of Graz as start and end points (e.g.rom the main square to the Schlossberg/Castle Hill of Graz).

Before the experimental task, it was assured that the 15 routesere familiar to all participants. The different routes were pre-

ented to the participants beforehand. The participants had toualitatively rate their familiarity with the single routes (yes–nouestions, all participants said “yes”), imagine to walk from thetarting point to the end point and to report aloud on how toavigate from the start to the end point of each route. Hence,he participants were asked to recall the visual and sensorimo-or mental images of the walk along the route aloud. These verbalescriptions of the routes were recorded. Analysis of these audioles revealed no differences in verbal route description betweenroups (frequency analysis of used words). All participants wereighly familiar with each route and used these routes regularly.one of the participants experienced difficulties in reporting theay from the start to the end points.

During the mental navigation task, the names of two prominentandmarks of the city of Graz (the first one was the starting pointnd the last one was the end point of the route) were presented viaarphones to indicate the route segment that the participants wereequired to mentally simulate. The participants were instructed toentally navigate for a few seconds from the start to the end point

f the route. During the EEG recording, participants only mentallyavigated on the routes. No pronounced verbal description of theoutes was required during the EEG measurement to avoid musclend movement artefacts.

.2.2. Motor imagery taskDuring the motor imagery task, participants were instructed to

magine softly clenching a ball with either their right or left hand,

hile their arms were resting relaxed on an armrest. Therefore,

he verbal instructions “right hand” or “left hand” were presentedhrough earphones to indicate that they should imagine a right or

left hand movement respectively. Note that participants were

Retinitis Pigmentosa 39Retinitis Pigmentosa 10

guide dog Detached retina 18

asked to imagine their kinesthetic experience of movement ratherthan the visual experience while avoiding muscle tension (e.g., feelthe movements as they physically perform the task) [52]. Beforedata collection started, participants could practice the movementby physically clenching a ball with their left and right hand to facil-itate kinesthetic movement imagery. The motor imagery task wasperformed with no ball in the hand.

2.2.3. ProcedureBefore the EEG measurement, it was ascertained that the routes

presented were familiar to all participants. Furthermore, partici-pants could practice mental navigation on the different routes aswell as the motor task.

After the EEG setup, a 2-min eyes-open and 2-min eyes-closedresting EEG measurement followed, in which participants wereasked to relax and to avoid eye movements. Afterwards, the exper-imental tasks followed. The timing of a trial is presented in Fig. 1.Each trial started with the presentation of a beep tone followed by areference interval of 2 s, in which participants were asked to relax.This reference interval was used as baseline interval for later EEGanalysis. After the reference interval, either the names of two land-marks (mental navigation task) or the instruction “left hand”/“righthand” (motor imagery task) were presented via earphones. Thepresentation of these verbal instructions took approximately 2 sin each new trial. Then the participants were required to performthe mental task (either mental navigation or motor imagery) for5 s. A further beep tone indicated the end of the mental task andthe start of the variable pause interval (2–4 s). In sum, the 15 routeswere presented 6 times, motor imagery of the left hand was pre-sented 45 times and motor imagery of the right hand was alsopresented 45 times. The resulting 90 trials of mental navigation and90 trials of motor imagery were presented in a randomized order.Sighted people were instructed to look at a fixation cross duringthe whole experimental task presented in the centre of a screenin front of the participants to reduce eye movements. Blind peo-ple were instructed to try to reduce eye movements while keepingeyes open.

2.3. EEG recordings and analysis

The EEG was recorded by Ag/AgCl electrodes from 36 electrodepositions according to the extended 10–20 electrode placementsystem against a linked mastoid reference. The ground was placedat AFz (Fig. 2). For EEG recording, a BrainAmp Standard ampli-fier (Brain Products GmbH, Munich, Germany) was used. Verticaland horizontal EOG was recorded with three electrodes in total,two were placed on the outer canthi of the eyes and one wasplaced superior to the nasion. EEG and EOG signals were ampli-fied, digitized (1000 Hz) and pre-filtered with a 0.5 Hz high-pass, a100 Hz low-pass, and a 50 Hz notch filter. Electrode impedances

were kept below 5 k� for the EEG recording and below 10 k�for the EOG recording. Data pre-processing and analysis was per-formed with the Brain Vision Analyzer software (version 2.01, BrainProducts GmbH, Munich, Germany). Ocular artefacts (eye blinks,

S.E. Kober et al. / Behavioural Brain Research 273 (2014) 106–115 109

enta

edrwswatFCpP(

ibmctaicNibt

Fgr

Fig. 1. Timing of a trial of the experim

ye movements) were automatically corrected using the algorithmeveloped by Gratton et al. (1983) [53]. After ocular artefact cor-ection, automated rejection of other EEG artefacts (e.g. muscles)as performed (Criteria for rejection: >50.00 �V voltage step per

ampling point, absolute voltage value > ± 100.00 �V). All epochsith artefacts were excluded from the EEG analysis (11% of all tri-

ls). To aggregate the data, electrode sites were collapsed into 12opographical regions of interest (ROI) (Fig. 2): left frontal FL (AF3,3), middle frontal FM (Fz), right frontal FR (AF4, F4), left centralL (FC3, C3), middle central CM (Cz), right central CR (FC4, C4), leftarietal PL (CP3, P3), middle parietal PM (Pz), right parietal PR (CP4,4), left occipital OL (PO3, O1), middle occipital (Oz), right occipitalPO4, O2).

For statistical analysis of the EEG data the percentage changen alpha (8–12 Hz) and theta (4–8 Hz) band power between theaseline interval (1000–2000 ms of the reference interval) and theental task interval (0–1000 ms of the mental task interval) was

alculated, which has been termed event-related desynchroniza-ion/synchronization (ERD/ERS) [54,55]. To avoid that theta andlpha ERD/ERS is influenced by phase-locked EEG activities, whichnclude all types of event-related potentials (ERPs), the ERD/ERSalculation was based on the intertrial variance method [36,56].ote that a decrease (negative values/ERD) in the alpha band power

s associated with cortical activation, whereas a relative alphaand power increase (positive values/ERS) is associated with cor-ical deactivation [54,55,57]. Generally, relative changes in alpha

ig. 2. Layout of the locations of the 36 EEG scalp electrodes indicated in light grey,round electrode at AFz. The 12 ROIs used for statistical analysis are marked withectangles.

l tasks during the EEG measurement.

band power are valid indicators of cortical activation/deactivation[57–59]. In contrary, a synchronization in the theta band (positivevalues/ERS) is associated with cortical activation [60].

Furthermore, in order to investigate whether absolute alpha andtheta power was comparable between groups during the baselineinterval of the experimental task (1000–2000 ms of the referenceinterval), which was used for the ERD/ERS calculation, absolutetheta (4–8 Hz) and alpha (8–12 Hz) band power was extractedby means of complex demodulation [61] for this time window.Additionally, for the 2-min eyes-open and eyes-closed resting mea-surements before the experimental task absolute alpha and thetapower was extracted by means of complex demodulation andpower spectrum analyses were performed by using Fast FourierTransform (FFT, 1-s epochs, Hanning-window).

To investigate possible functional couplings between corticalareas, coherence analyses were applied for the mental task interval(0–1000 ms of the mental task interval) [39]. A functional rela-tionship between different brain areas is generally associated withsynchronous electrical activity in these regions. A quantitativemeasure for this synchrony is the EEG coherence between signalsrecorded from electrode pairs as a function of frequency [62]. Con-tinuous EEG during the mental task interval (0–1000 ms of themental task interval) was segmented into artefact-free epochs of1 s duration. For each segment, EEG power spectra were calculatedusing Fast Fourier Transformation (FFT). FFT was computed for eachdata segment with maximum resolution of ∼0.98 Hz. Furthermore,a 10% Hanning window was applied including a variance correc-tion to preserve overall power. Next, computation of coherence rbetween two channels c1 and c2 at a given frequency f was basedon FFT transformed complex data C according to Eq. (1)

r(c1, c2)(f ) =∣∣CS(c1, c2)(f )

∣∣2

∣∣CS(c1, c1)(f )∣∣ ∣∣CS(c2, c2)(f )

∣∣ (1)

with cross (respectively auto) spectra CS averaged over i datasegments given by Eq. (2)

CS(c1, c2)(f ) =∑

C1i(f )C2i(f )∗

(2)

[61]. Coherence between an electrode pair was defined as thecross spectral density function normalized by individual auto spec-tral density functions [63]. The coherence value r represents ageneralization of the Pearson product correlation coefficient tovariables expressed in the frequency domain [64]. Coherence is ameasure of the dependency of the data between two individualchannels over time. A resulting value of 0 indicates no correla-tion in frequency, whereas a resulting value around 1 indicatesan ideal constant correlation. The r values we present indicatethe coherence in the low beta band 13–18 Hz since prior studiesrevealed differences in EEG coherence between blind and sightedpeople especially in this frequency range [39]. Note that coher-

ence analyses in the theta (4–8 Hz) and alpha (8–12 Hz) bandrevealed no significant results in the present study. Coherence inthe low beta band is associated with information processing, visuo-motor activities, working memory, and visuo-spatial processing

110 S.E. Kober et al. / Behavioural Brain R

Fra

[(nlwt

2

tbmmfsRactibtvvtwpa

Fs

ig. 3. Grand average EEG power spectra at electrode position Pz for the 2-minesting measurements with open and closed eyes, presented separately for blindnd sighted participants.

39]. Coherence estimates were derived for four electrode pairsF3–C3, F4–C4, P3–O1, P4–O2) to assess if there is a functional con-ectivity in fronto-central and parietal-occipital networks in the

eft and right hemisphere. For statistical analyses, coherence valuesere Fisher’s z-transformed, and means were inverse transformed

o normalize the distribution of the correlation measures.

.4. Statistical analysis

To analyse whether there are differences in absolute alpha orheta power between blind and sighted participants during theaseline interval of the experimental task or during the resting EEGeasurements before the experimental task, univariate repeated-easures analyses of variance (ANOVA) with the between-subjects

actor GROUP (blind vs. sighted participants) and the within-ubjects factors FCPO (frontal vs. central vs. parietal vs. occipitalOI) and LMR (left vs. middle vs. right ROI) were calculated sep-rately for the alpha and theta power. The same ANOVAs werealculated for the dependent variable ERD/ERS in the alpha andheta band during the mental navigation task. For the motormagery task, a univariate repeated-measures ANOVA with theetween-subjects factor GROUP (blind vs. sighted participants) andhe within-subjects factors HAND (left vs. right hand), FCPO (frontals. central vs. parietal vs. occipital ROI) and LMR (left vs. middles. right ROI) were calculated for ERD/ERS in the alpha band. For

he coherence analysis, an univariate repeated-measures ANOVAith the between-subjects factor GROUP (blind vs. sighted partici-ants) and the within-subjects factors LR (left vs. right hemisphere)nd PAIR (F-C vs. P-O) was calculated for coherence in the beta

ig. 4. Topographic maps of (a) alpha band power (8–12 Hz) and (b) theta band power

eparately for blind and sighted participants.

esearch 273 (2014) 106–115

band during the mental navigation task and an univariate repeated-measures ANOVA with the between-subjects factor GROUP (blindvs. sighted participants) and the within-subjects factors HAND (leftvs. right hand), LR (left vs. right hemisphere) and PAIR (F-C vs. P-O) was calculated for coherence in the beta band during the motorimagery task.

The probability of a Type I error was maintained at p = 0.05.For the ANOVA’s, Mauchly’s tests of sphericity were carried out onthe repeated-measures variables, and where violated, Greenhouse-Geisser correction was applied. For post-hoc analyses, Bonferronicorrections for multiple comparisons were applied.

3. Results

3.1. EEG power during resting measurements

In Fig. 3, EEG power spectra for the 2-min resting measurementswith open and closed eyes are depicted for blind and sighted par-ticipants. As one can see in the power spectra, blind and sightedparticipants showed a peak in the alpha range (8–12 Hz). Hence,sighted as well as blind individuals exhibited alpha oscillationsin their EEG. Generally, the alpha rhythm is predominant in theEEG when the individual is relaxed and has the eyes closed. Alphadisappears or is “blocked” when the eyes are opened [45,65]. Thisphenomenon of “alpha blocking” could be seen in sighted partici-pants. Alpha power was higher during the eyes-closed than duringthe eyes-open condition in sighted people (Fig. 3). In contrast, alphablocking was not evident in the blind. Alpha power did not changebetween the eyes-open and eyes-closed condition in blind partici-pants (Fig. 3).

To test statistically whether blind and sighted people werecomparable in their EEG power during the resting measurements,we compared the alpha and theta power of both groups dur-ing the 2-min eyes-open and 2-min eyes-closed resting conditionbefore the experimental task. For the eyes-open resting con-dition, the 2 × 4 × 3 ANOVA (GROUP × FCPO × LMR) revealed nosignificant group differences neither for alpha (F(1,14) = 0.003, ns.,�2 = 0.00) nor for theta (F(1,14) = 1.47, ns., �2 = 0.09) power. Forthe eyes-closed resting condition, the 2 × 4 × 3 ANOVA (GROUPx FCPO x LMR) revealed no group differences in theta power(F(1,14) = 0.77, ns., �2 = 0.05). However, alpha power during the

eyes-closed resting condition was higher in sighted than inblind participants as indicated by a significant main effectof GROUP (F(1,14) = 11.25, p < 0.01, �2 = 0.45). The interactioneffect GROUP × FCPO × LMR was significant as well (F(6,84) = 4.27,

(4–8 Hz) during the resting measurements with open and closed eyes, presented

S.E. Kober et al. / Behavioural Brain Research 273 (2014) 106–115 111

Table 2Means and standard errors of absolute alpha and theta power during the baseline interval before the experimental tasks presented separately for blind and sighted participantsin each of the 12 ROIs.

Baseline interval (1000–2000 ms of the reference interval)

Blind participants Sighted participants

Theta powermean (SE) Alpha powermean (SE) Theta powermean (SE) Alpha powermean (SE)

Regions of interest ROIs FL 4.42 (1.03) 3.97 (1.16) 3.34 (0.47) 3.42 (0.98)FM 5.88 (1.21) 4.78 (1.34) 4.12 (0.63) 4.19 (1.09)FR 4.27 (0.83) 3.90 (1.08) 3.34 (0.50) 3.53 (0.94)CL 4.60 (1.06) 5.93 (2.20) 3.37 (0.46) 4.87 (1.69)CM 6.01 (1.18) 6.31 (2.03) 4.43 (0.65) 5.00 (1.48)CR 4.48 (0.87) 6.19 (2.37) 3.39 (0.55) 5.29 (1.73)PL 4.49 (1.14) 9.56 (4.01) 3.07 (0.45) 6.00 (1.81)PM 5.06 (1.22) 9.83 (4.30) 3.42 (0.53) 6.01 (1.77)PR 4.11 (0.92) 7.47 (2.99) 3.02 (0.49) 6.71 (2.10)

42 (4.26 (1.67 (2.

ppwbspaca

3

itomL(cglfobawfprgt

3

tmatsapiadt

was significant (F(6,84) = 4.02, p < 0.01, �2 = 0.22). Post-testsrevealed that alpha ERD was stronger over left centro-parietalregions (around the motor cortex) during a right hand motor

OL 3.67 (1.05) 9.OM 2.63 (0.74) 5.OR 3.11 (0.75) 6.

< 0.01, �2 = 0.23) indicating higher alpha power values at occipital-arietal sites compared to fronto-central sites in sighted peoplehereas no significant topographical difference was observed in

lind people (see Fig. 4a). Nevertheless, blind participants alsohowed slightly increased alpha power over parietal-occipital com-ared to frontal brain regions during the eyes-closed condition,s illustrated in Fig. 4a. Theta power had its maximum at fronto-entral sites around Fz and Cz in both groups during the eyes-opens well as the eyes-closed condition (Fig. 4b).

.2. EEG power during the baseline interval

To test whether blind and sighted participants were comparablen their EEG power during the baseline interval (1000–2000 ms ofhe reference interval), we compared the alpha and theta powerf both groups during this reference interval before the experi-ental tasks by calculating a 2 × 4 × 3 ANOVA (GROUP x FCPO x

MR). Neither for alpha (F(1,14) = 0.34, ns., �2 = 0.02) nor for thetaF(1,14) = 1.27, ns., �2 = 0.08) power significant group differencesould be found. Hence, differences in ERD/ERS values betweenroups or conditions cannot be attributed to differences in abso-ute EEG power values during the baseline interval, which was usedor ERD/ERS calculations. In Table 2 means and standard errorsf absolute alpha and theta power during the baseline intervalefore the experimental task are presented separately for blindnd sighted participants. Since variability among blind participantsas rather high compared to sighted controls, we additionally per-

ormed permutation tests (Mann-Whitney U test) comparing theower values between groups. These non-parametric tests alsoevealed no significant differences in alpha or theta power betweenroups. Generally, the most important advantage of permutationests is that the results are also reliable for small samples.

.3. ERD/ERS during the mental navigation task

For ERD/ERS in the alpha band during mental navigation,he 2 × 4 × 3 ANOVA (GROUP × FCPO × LMR) revealed a significant

ain effect of GROUP (F(1,14) = 12.73, p < 0.01, �2 = 0.48) indicating stronger ERD in the alpha band in blind than in sighted par-icipants. Furthermore, the interaction effect GROUP × FCPO wasignificant (F(3,42) = 4.89, p < 0.01, �2 = 0.26). Post-tests revealed

stronger cortical activation (higher alpha ERD values) in blindarticipants compared to sighted participants especially at occip-

tal sites. The topographical plots of alpha ERD/ERS for blindnd sighted participants during the mental navigation task areepicted in Fig. 5a. The topographical distribution of ERD/ERS inhe alpha band differed across groups. Blind participants showed

21) 2.37 (0.25) 5.17 (1.49)93) 2.01 (0.20) 4.15 (1.16)76) 2.43 (0.30) 5.33 (1.64)

a desynchronization (ERD) in the alpha band over the whole cor-tex, with a maximum activation over occipital brain areas, whereassighted participants showed the strongest activation (alpha ERD)at more right fronto-central sites. At occipital areas, sighted par-ticipants even showed a synchronization (ERS) in the alpha band,which is generally associated with cortical deactivation.

For ERD/ERS in the theta band during mental navigation,the 2 × 4 × 3 ANOVA (GROUP × FCPO × LMR) revealed a significantmain effect of GROUP (F(1,14) = 5.03, p < 0.05, �2 = 0.26) indicating astronger ERS in the theta band in sighted than in blind participants.The topographical distribution of theta ERD/ERS in both groups isdepicted in Fig. 5b. Sighted participants showed the strongest thetaERS in fronto-central sites during mental navigation, which is asso-ciated with cortical activation. In contrast, blind participants evenshowed a desynchronization (ERD) in the theta band over the wholecortex.

3.4. ERD/ERS during the motor imagery task

For ERD/ERS in the alpha band during the motor imagery task,the 2 × 2 × 4 × 3 ANOVA (GROUP × HAND × FCPO × LMR) revealedno group differences. The interaction effect HAND × FCPO × LMR

Fig. 5. Topographic maps of (a) ERD/ERS in the alpha band (8–12 Hz) and (b)ERD/ERS in the theta band (4–8 Hz) during the mental navigation task, presentedseparately for blind and sighted participants.

112 S.E. Kober et al. / Behavioural Brain R

Fis

iatmtbaisocw

3

vr�(CttbtPpt

t

Faas

ig. 6. Topographic maps of ERD/ERS in the alpha band (8–12 Hz) during motormagery of a left (upper panel) and right (lower panel) hand movement, presentedeparately for blind and sighted participants.

magery task compared to the left hand motor imagery task. Leftnd right hand motor imagery tasks led to the strongest activa-ion (alpha ERD) over bilateral centro-parietal regions around the

otor cortex (C3, CP3, C4, CP4) in both groups (see Fig. 6). Hence,he activation patterns during motor imagery were comparableetween blind and sighted participants. The topographical plots oflpha ERD/ERS for blind and sighted participants during the motormagery task are depicted in Fig. 6. Although sighted participantshowed more specific and focused activation patterns (alpha ERD)nly over motor areas compared to blind participants, no signifi-ant group differences in the activation level during motor imageryere observed.

.5. Coherence analysis

The 2 × 2 × 2 ANOVA (GROUP × LR × PAIR) for the dependentariable coherence in the beta band during mental navigationevealed a significant main effect PAIR (F(1,14) = 23.46, p < 0.01,2 = 0.63). Coherence between parietal-occipital (P-O) brain areasr = 0.56; SE = 0.04) was higher than between fronto-central (F-) areas (r = 0.33; SE = 0.03). The main effect GROUP was byrend significant (F(1,14) = 4.11, p = 0.06, �2 = 0.23). Sighted par-icipants (r = 0.51; SE = 0.04) had higher coherence values thanlind participants (r = 0.38; SE = 0.04). Furthermore, the interac-ion GROUP × LR was significant (F(1,14) = 5.01, p < 0.05, �2 = 0.26).ost-tests revealed that coherence between fronto-central and

arietal–occipital electrode pairs was significant higher in sightedhan in blind participants only in the left hemisphere (see Fig. 7).

The 2 × 2 × 2 × 2 ANOVA (GROUP x HAND x LR x PAIR) forhe dependent variable coherence in the beta band during motor

ig. 7. Coherence values (r) in the beta band (13–18 Hz) during mental navigationveraged over fronto-central and parietal-occipital electrode pairs, presented sep-rately for electrode pairs in the left and in the right hemisphere and for blind andighted people. Significant differences are marked with asterisks (*p < 0.05).

esearch 273 (2014) 106–115

imagery revealed a significant main effect PAIR (F(1,14) = 19.08,p < 0.01, �2 = 0.58). Coherence between parietal-occipital (P-O)brain areas (r = 0.57; SE = 0.05) was higher than between fronto-central (F-C) areas (r = 0.38; SE = 0.04). Furthermore, the main effectHAND was significant (F(1,14) = 4.87, p < 0.05, �2 = 0.26). Coher-ence was higher during imagery of a left hand movement (r = 0.48;SE = 0.04) than during imagery of a right hand movement (r = 0.46;SE = 0.04). No significant differences in coherence values could befound between groups in this control task.

4. Discussion

In the present study we investigated brain activation patternsduring mental navigation in blind compared to sighted people.Hence, we were interested in the functional reorganization ofthe occipital cortex in the blind associated with a mental naviga-tion task. Accordingly, sighted and blind participants performeda mental navigation task and a motor imagery task, which wasused as a control task, during a multi-channel EEG measure-ment. Furthermore, we compared EEG patterns during restingmeasurements between blind and sighted participants to exam-ine task-independent alterations in electrophysiological activationpatterns in the blind. During the resting measurement, blind andsighted participants did not differ in their EEG activation patternsexcept in alpha power during an eyes-closed resting period. How-ever, fundamental differences in cortical activation patterns andbrain connectivity between blind and sighted participants werefound during mental navigation but not during the motor imagerycontrol task. This indicates a functional reorganization of occipi-tal brain areas in the blind, which is specifically related to mentalnavigation. In the following, we will discuss these results in moredetail.

In a first step, we analysed the EEG of blind and sighted par-ticipants during 2-min eyes-open and 2-min eyes-closed restingmeasurements to investigate possible general and unspecific alter-ations in the EEG in blind compared to sighted individuals. Duringthe eyes-open resting measurement, no differences in the EEGactivity could be observed between blind and sighted participants.Both groups showed alpha oscillations in the EEG when their eyeswere open as indicated by a predominant peak in the alpha rangein the power spectra of blind and sighted participants (Fig. 3). Wefound no deviations or abnormalities in alpha power activity inblind compared to sighted people during an eyes-open condition.Prior studies reported that up to 60% of blind participants show noalpha activity in the EEG at all [45,48]. However, there is also evi-dence that alpha activity in blind participants is related to differentfactors such as cause of blindness or time of onset of blindness [45].For instance, the alpha rhythm in the EEG during resting measure-ments was more common in children who had not been blind frombirth [45]. In the present study, only two blind participants werecongenitally blind. Although these two participants also showedEEG alpha activity, the variations in the onset and cause of blindnessin the present sample might be a reason why no differences in EEGalpha activity between blind and sighted participants during theeyes-open condition could be found. The results of the eyes-closedcondition were in line with prior findings [45,46,48,50]. Duringthis condition, sighted participants showed a significant increasedalpha power over parietal–occipital brain areas, which could not beseen in blind participants, although the blind also showed slightlyincreased alpha power over parietal-occipital compared to frontalbrain regions. Hence, we observed the well-known alpha blocking

phenomenon in sighted but not in blind individuals [45,65]. Jeav-ons et al. (1964) also reported that the most common alpha rhythmpattern in the blind was one similar to that seen in sighted con-trols with open eyes. In line with that, we found that alpha activity

rain R

pdstosR

pstWhpsasamspttigitnsptfwr

iirde(svafTifvs(utnpts

pnit(pSd

S.E. Kober et al. / Behavioural B

atterns were comparable between blind and sighted participantsuring an eyes-open condition, whereas alpha power was higher inighted than in blind people during an eyes-closed condition. In theheta power we found no group differences, neither for the eyes-pen nor for the eyes-closed condition. Both groups showed thetrongest theta power over fronto-central brain areas [34,36,60,66].eports on theta rhythm in blind people are rare [45].

In a second step, we investigated differences in brain activationatterns between the blind and sighted controls during a mentalpatial navigation task, since there is evidence that the visual cor-ex generally takes over other functions in the absence of vision.

hen mentally navigating on a familiar route within their ownometown, blind participants showed in comparison to sightedarticipants a stronger cortical activation especially at occipitalites, as indicated by higher alpha ERD values. Hence, the blindctivated the visual cortex more during mental navigation thanighted people. There is evidence that blind people rely more onuditory and tactile stimuli during navigation to compensate for theissing visual inputs [3,20,21]. Furthermore, prior neuroimaging

tudies found an increased activation in the visual cortex of blindeople when performing tactile or auditory tasks indicating thathis brain area takes over the processing of other sensory stimuli inhe absence of vision [3,14,24,27,28,38,39,67]. Hence, when blindndividuals primarily use auditory and tactile information as navi-ational aids, it is assumed that the visual cortex should be involvedn spatial navigation processes in the blind. In line with this assump-ion, our results indicate that the visual cortex is involved in spatialavigation in blind participants but not in sighted controls. Priortudies also found an involvement of the occipital cortex of blindeople in diverse spatial tasks [5,14,21,24–28,37–40]. However, tohe best of our knowledge this is the first study providing evidenceor the functional reorganization of the visual cortex in blind peoplehen using a more conventional mental navigation task on familiar

outes.Although brain correlates of active navigation have often been

nvestigated in sighted people [17,31–36], reports on modulationsn EEG power during mental navigation in sighted individuals areare. Friedrich et al. (2012) analysed changes in EEG power duringifferent mental tasks such as mental spatial navigation in familiarnvironments. They found a desynchronization (ERD) in the lower13–20 Hz) and upper (20–30 Hz) beta band over fronto-centralites in sighted individuals during mental navigation, but no acti-ation in the occipital cortex [68]. In a further study by the sameuthors, no ERD in the alpha band over occipital areas could beound in sighted individuals performing mental navigation [69].hese results are in line with the present findings in sighted partic-pants, who also showed stronger activation (alpha ERD) over moreronto-central areas especially in the right hemisphere and no acti-ation in the occipital cortex during mental navigation. Generally,patial navigation tasks are considered a right hemispheric taske.g. [17,70,71]). There is evidence that neuronal activation patternsnderlying navigation in real and mental space are fundamentallyhe same [40,41]. Hence, the visual cortex in the occipital lobe doesot seem to play an important role in spatial navigation in sightedeople. Therefore, the increased activity (alpha ERD) over occipi-al regions in the blind indicates that the visual cortex takes overpatial navigation functions in the absence of vision.

In the theta band ERS, we found a stronger increase in thetaower (ERS) in sighted than in blind participants during mentalavigation over fronto-central sites, which is associated with an

ncreased cortical activity in sighted than in blind people. In con-rast, blind participants even showed a relative power decrease

ERD) in the theta band over the whole cortex. Generally, thetaower is found to increase during spatial tasks [34–36,66,72].ince there were no differences in theta power between groupsuring the resting measurement or during the baseline interval

esearch 273 (2014) 106–115 113

before the mental navigation task, these differences in relative thetapower during mental navigation are specifically related to navi-gation. Prior studies in sighted individuals linked increased thetapower observed during spatial navigation tasks to sensorimotorintegration [36,66,72]. When a visual cue or landmark, which isused for navigation, is visible, theta oscillations normally increase.It is assumed that the incoming sensory landmark informationupdates the motor plan and leads to a decision on the direction[36]. Hence, an increase in theta power observed during spatialnavigation might be related to incoming visual cues. In the presentstudy, sighted participants may have used such visual landmarksto mentally navigate on the familiar routes, which might have ledto increased theta ERS. In contrast, blind individuals might haveused other sensory cues than visual ones for mentally retrievingthe familiar routes, probably leading to theta ERD. Note that differ-ences in ERD/ERS values between the groups cannot be attributedto differences in general EEG power values since the groups did notdiffer in their absolute alpha or theta power during the baselineinterval before the experimental tasks. Hence, group differences inrelative event-related power increases (ERS) or decreases (ERD) arenot influenced by differences in baseline power levels.

In order to investigate the specificity of brain activation patternsin blind and sighted people during mental navigation, we also per-formed a motor imagery task as control condition. In contrast tothe results of the mental navigation task, no group differences incortical activation patterns could be observed in the control task.During motor imagery, both groups showed increased activationover motor areas. These results are in line with prior motor imagerystudies showing increased activation during motor imagery of handor foot movements over contralateral motor areas in the brain. Gen-erally, motor imagery activates comparable brain areas than motorexecution [52,73,74]. Hence, the functional reorganization of theoccipital cortex in blind participants seems to be specific for mentalnavigation.

In a final step, we analysed functional brain connectivity inblind and sighted participants during mental navigation. Sightedparticipants showed compared to blind participants stronger func-tional connectivity (coherence in the beta band 13–18 Hz) betweenfronto-central and parietal–occipital electrode pairs – particularlyin the left hemisphere. Prior studies reported heterogeneous resultsconcerning the coherence in the EEG of blind and sighted peo-ple. For instance, Leclerc et al. (2005) found increased connectivitybetween central and posterior cortical regions in blind comparedto sighted individuals during a sound localization task indicatingthat the occipital cortex, which is normally reserved for visualprocessing in sighted individuals, is more integrated into the audi-tory attention and processing system of blind individuals [27].Campus et al. (2012) found increased coherence in the beta band(14–18 Hz) in both blind and sighted individuals, which emergedamong fronto-central, centro-parietal, and occipito-temporal elec-trode sites associated with visuo-spatial processing. In that study,participants either passively or actively explored the contour ofobjects by touch. During active exploration, blind participantsshowed lower brain connectivity compared to sighted controls.However, group differences in coherence values were comparablebetween electrode pairs in the left and right hemisphere [39]. Thereis evidence that different functional networks within the brainplay an important role in visuo-spatial processing. Particularly,parietal–occipital circuits seem to integrate information equallyfrom central and peripheral visual fields. It is assumed that thiscircuit transforms visual representations into additional referenceframes relative to parts of the body as well as the eye [75]. Hence,

increased connectivity in sighted people compared to blind peoplemay reflect such visuo-spatial processes during mental navigation.Since direct visual representations of the environment are absentin the blind, it seems obvious that these networks are less active

1 Brain R

rsfc

apbcaSciiwiavolpoomaivm

5

rEnbdtoiTti

A

iu(cTmc

R

[

[

[

[

[

[

[[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

14 S.E. Kober et al. / Behavioural

esulting in a reduced coherence in the beta band in blind than inighted participants. Results of the coherence analysis were specificor mental navigation, since no group differences in beta coherenceould be observed during the motor imagery control task.

Preliminary analysis revealed no significant differences in EEGctivation patterns between congenitally blind participants andarticipants who became blind over the lifespan, so-called latelind participants. In the present study, only two participants wereongenitally blind. Prior studies reported on differences in brainctivation patterns between early and late blind people [10,45,76].adato et al. (2002) found that the first 16 years of life represent aritical period for a functional reorganization of the visual cortexn blind individuals. The primary visual cortex (V1) was activatedn blind people who lost their sight before the age of 16 years,

hereas activation in V1 was suppressed during a tactile discrim-nation task in blind people who lost their sight after 16 years ofge. Furthermore, the cause of blindness also influences brain acti-ation patterns [45]. The small and relatively heterogeneous groupf blind participants concerning onset and cause of blindness is aimitation of the present study. Furthermore, we did not assess lighterception in the blind participants, which might have an influencen EEG activity in the visual cortex as well. However, the resultsf the present study did not indicate differences in ERD/ERS duringental navigation between the two congenitally blind participants

nd the six late blind participants. Future studies are needed tonvestigate the effects of onset and cause of blindness on EEG acti-ation patterns during mental navigation as well as during restingeasurements in more detail.

. Conclusion

In the present study we focused on electrophysiological cor-elates of mental navigation in the blind. We found differences inEG activation patterns and connectivity measures during a mentalavigation task between blind and sighted individuals. Especially,lind participants showed increased activation in the visual cortexuring mental navigation, which could not be seen in sighted con-rols. Hence, our study provides further evidence for the plasticityf the visual cortex. When the visual cortex receives no visual input,t takes over other functions such as spatial navigation processes.he present study demonstrated the functional reorganization ofhe occipital cortex for a more conventional mental navigation taskn familiar environments in blind people.

cknowledgement

This work was partially supported by the Games and Learn-ng Alliance (GaLA)–Network of Excellence for Serious Gamesnder the European Community Seventh Framework ProgrammeFP7/2007 2013), Grant Agreement no. 258169. Possible inaccura-ies of information are under the responsibility of the project team.he text reflects solely the views of its authors. The European Com-ission is not liable for any use that may be made of the information

ontained therein.The authors declare that they have no competing interests.

eferences

[1] Birbaumer N, Schmidt R. Biologische Psychologie. 6th ed. Heidelberg: Springer;2006.

[2] Noppeney U. The effects of visual deprivation on functional and structuralorganization of the human brain. Neurosci Biobehav Rev 2007;31:1169–80.

[3] Cattaneo Z, Vecchi T, Cornoldi C, Mammarella I, Bonino D, Ricciardi E, Pietrini

P. Imagery and spatial processes in blindness and visual impairment. NeurosciBiobehav Rev 2008;32:1346–60.

[4] Cohen L, Celnik P, Pascual-Leone A, Corwell B, Falz L, Dambrosia J, Honda M,et al. Functional relevance of cross-modalplasticity in blind humans. Nature1997;389:180–3.

[

[

esearch 273 (2014) 106–115

[5] Röder B, Rösler F, Hennighausen E. Different cortical activation patterns in blindand sighted humans during encoding and transformation of haptic images.Psychophysiology 1997;34:292–307.

[6] Sadato N, Pascual-Leone A, Grafman J, Ibanez V, Deiber M, Dold G, Hallett M.Activation of the primary visual cortex by Braille reading in blind subjects.Nature 1996;380:526–8.

[7] Sadato N, Pascual-Leone A, Grafman J, Deiber M, Iban V, Hallett M. Neuralnetworks for Braille reading by the blind. Brain 1998;121:1213–29.

[8] Burton H, Snyder A, Conturo T, Akbudak E, Ollinger J, Raichle M. Adaptivechanges in early and late blind: a fMRI study of Braille reading. J Neurophysiol2002;87:589–607.

[9] Ptito M, Moesgaard S, Gjedde A, Kupers R. Cross-modal plasticity revealedby electrotactile stimulation of the tongue in the congenitally blind. Brain2005;128:606–14.

10] Büchel C, Price C, Frackowiak R, Friston K. Different activation patterns in thevisual cortex of late and congenitally blind subjects. Brain 1998;121:409–19.

11] Röder B, Rösler F, Hennighausen E, Näcker F. Event-related potentials dur-ing auditory and somatosensory discrimination in sighted and blind humansubjects. Cogn Brain Res 1996;4:77–93.

12] Armedi A, Raz N, Pianka P, Malach R, Zohary E. Early ‘visual’ cortex activationcorrelates with superior verbal memory performance in the blind. Nat Neurosci2003;6:758–66.

13] Arno P, Volder Ade, Vanlierde A, Wanet-Defalque M, Streel E, Robert A,Sanabria-Bohórquez S, et al. Occipital activation by pattern recognition inthe early blind using auditory substitution for vision. NeuroImage 2001;13:632–45.

14] Weeks R, Horwitz B, Aziz-Sultan A, Tian B, Wessinger C, Cohen L. A positronemission tomographic study of auditory localization in the congenitally blind.J Neurosci 2000;20:2664–72.

15] Bavelier D, Neville H. Cross-modal plasticity: where and how? Nat Neurosci2002;3:443–52.

16] Tolman EC. Cognitive maps in rats and men. Psychol Rev 1948;55:189–208.17] Dudchenko PA. Why People Get Lost: The Psychology and Neuroscience of

Spatial Cognition. New York: Oxford Univ. Press; 2010.18] Fortin M, Voss P, Lord C, Lassonde M, Pruessner J, Saint-Amour D, Rainville C,

et al. Wayfinding in the blind: larger hippocampal volume and supranormalspatial navigation. Brain 2008;131:2995–3005.

19] Loomis JM, Klatzky RL, Golledge RG, Cicinelli JG, Pellegrino JW, Fry PA. Nonvisualnavigation by blind and sighted: assessment of path integration ability. J ExpPsychol Gen 1993;122:73–91.

20] Thinus-Blanc C, Gaunet F. Representation of space in blind persons: vision as aspatial sense? Psychol Bull 1997;121:20–42.

21] Kupers R, Chebat DR, Madsen KH, Paulson OB, Ptito M. Neural correlatesof virtual route recognition in congenital blindness. Proc Natil Acad Sci2010;107:12716–21.

22] Cornoldi C, Cortesi A, Preti D. Individual differences in the capacity limitations ofvisuospatial short-term memory: research on sighted and totally congenitallyblind people. Mem Cognition 1991;19:459–68.

23] Haber R, Haber L, Levin C, Hollyfield R. Properties of spatial representations:data from sighted and blind subjects. Percept Psychophys 1993;54:1–13.

24] Gagnon L, Schneider FC, Siebner HR, Paulson OB, Kupers R, Ptito M. Activationof the hippocampal complex during tactile maze solving in congenitally blindsubjects. Neuropsychologia 2012;50:1663–71.

25] Després O, Candas V, Dufour A. Spatial auditory compensation in early-blindhumans: Involvement of eye movements and/or attention orienting? Neu-ropsychologia 2005;43:1955–62.

26] Després O, Candas V, Dufour A. The extent of visual deficit and auditory spatialcompensation: evidence from self-positioning from auditory cues. Cogn BrainRes 2005;23:444–7.

27] Leclerc C, Segalowitz SJ, Desjardins J, Lassonde M, Lepore F. EEG coherence inearly-blind humans during sound localization. Neurosci Lett 2005;376:154–9.

28] Voss P, Gougoux F, Lassonde M, Zatorre R, Lepore F. A positron emissiontomography study during auditory localization by late-onset blind individuals.NeuroReport 2006;17:383–8.

29] Lahav O, Mioduser D. Blind persons’ acquisition of spatial cognitive mappingand orientation skills supported by virtual environment. Int J Disabil Hum Dev2005;4:231–7.

30] Passini R, Proulx G. Wyafinding without vision – an experiment with congeni-tally totally blind people. Environ Behav 1988;20:227–52.

31] Maguire EA, Burgess N, Donnett JG, Frackowiak RSJ, Frith CD, O’Keefe J.Knowing where and getting there a human navigation network. Science1998;280:921–4.

32] Maguire EA, Burgess N, O’Keefe J. Human spatial navigation: cognitive maps,sexual dimorphism, and neural substrates. Curr Opin Neurobiol 1999;9:171–7.

33] Janzen G, van Turennout M. Selective neural representation of objects relevantfor navigation. Nat Neurosci 2004;7:673–7.

34] Kahana MJ, Sekuler R, Caplan JB, Kirschen M, Madsen JR. Human thetaoscillations exhibit task dependence during virtual maze navigation. Nature1999;399:781–4.

35] Kahana MJ, Seelig D, Madsen JR. Theta returns. Curr Opin Neurobiol2001;11:739–44.

36] Kober SE, Neuper C. Sex differences in human EEG theta oscillations duringspatial navigation in virtual reality. Int J Psychophysiol 2011;79:347–55.

37] Deutschländer A, Stephan T, Hüfner K, Wagner J, Wiesmann M, Strupp M,Brandt T, et al. Imagined locomotion in the blind: an fMRI study. NeuroImage2009;45:122–8.

rain R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

S.E. Kober et al. / Behavioural B

38] Fiehler K, Burke M, Bien S, Roder B, Rosler F. The human dorsal action controlsystem develops in the absence of vision. Cerebral Cortex 2009;19:1–12.

39] Campus C, Brayda L, Carli Fde, Chellali R, Fama F, Bruzzo C, Lucagrossi L, et al.Tactile exploration of virtual objects for blind and sighted people: the role ofbeta 1 EEG band in sensory substitution and supramodal mental mapping. JNeurophysiol 2012;107:2713–29.

40] Vanlierde A, Volder AGde, Wanet-Defalque M, Veraart C. Occipito-parietal cor-tex activation during visuo-spatial imagery in early blind humans. NeuroImage2003;19:698–709.

41] Buzsáki G, Moser EI. Memory, navigation and theta rhythm in the hippocampal-entorhinal system. Nat Neurosci 2013;16:130–8.

42] Ptito M, Schneider FCG, Paulson OB, Kupers R. Alterations of the visual pathwaysin congenital blindness. Exp Brain Res 2008;187:41–9.

43] Chebat D, Chen J, Schneider F, Ptito A, Kupers R, Ptito M. Alterations in right pos-terior hippocampus in early blind individuals. Neuroreport 2007;18:329–33.

44] Volder AGde, Bol A, Blin J, Robert A, Arno P, Grandin C, Michel C, et al. Brainenergy metabolism in early blind subjects: neural activity in the visual cortex.Brain Res 1997;750:235–44.

45] Jeavons P. The electro-encephalogram in blind children. Brit J Ophthal1964;48:83–101.

46] Jeavons P, Harding G, Ferries G, Thompson C. Alpha rhythm in totally blindchildren. Brit J Ophthal 1970;54:786–93.

47] Bértolo H, Paiva T, Pessoa L, Mestre T, Marques R, Santos R. Visual dreamcontent, graphical representation and EEG alpha activity in congenitally blindsubjects. Cogn Brain Res 2003;15:277–84.

48] Adrian E, Matthews B. The Berger rhythm: potential changes from the occipitallobes in man. Brain 1934;57:355–85.

49] Hawellek DJ, Schepers IM, Roeder B, Engel AK, Siegel M, Hipp JF. Alteredintrinsic neuronal interactions in the visual cortex of the blind. J Neurosci2013;33:17072–80.

50] Kriegseis A, Hennighausen E, Rösler F, Röder B. Reduced EEG alpha activity overparieto-occipital brain areas in congenitally blind adults. Clin Neurophysiol2006;117:1560–73.

51] WMA. (World Medical Association). Declaration of Helsinki. Ethical prin-ciples for medical research involving human subjects. J Indian Med Assoc2009;107:403–405.

52] Neuper C, Scherer R, Reiner M, Pfurtscheller G. Imagery of motor actions: differ-ential effects of kinesthetic and visual–motor mode of imagery in single-trialEEG. Cogn Brain Res 2005;25:668–77.

53] Gratton G, Coles MG, Donchin E. A new method for off-line removal of ocularartifact. Electroencephalogr Clin Neurophysiol 1983;55:468–84.

54] Pfurtscheller G. Spatiotemporal analysis of alpha frequency components withthe ERD technique. Brain Topogr 1989;2:3–8.

55] Pfurtscheller G, Lopes da Silva FH. Event-related EEG/MEG synchronization anddesynchronization: basic principles. Clin Neurophysiol 1999;110:1842–57.

56] Kalcher J, Pfurtscheller G. Discrimination between phase-locked and non-phase-locked event-related EEG activity. Electroencephalogr Clin Neurophysiol1995;94:381–4.

57] Klimesch W, Sauseng P, Hanslmayr S. EEG alpha oscillations: the inhibition-timing hypothesis. Brain Res Rev 2007;53:63–88.

[

[

esearch 273 (2014) 106–115 115

58] Laufs H, Kleinschmidt A, Beyerle A, Eger E, Salek-Haddadi A, PreibischC, Krakow K. EEG-correlated fMRI of human alpha activity. NeuroImage2003;19:1463–76.

59] Laufs H, Krakow K, Sterzer P, Eger E, Beyerle A, Salek-Haddadi A, KleinschmidtA. Electroencephalographic signatures of attentional and cognitive defaultmodes in spontaneous brain activity fluctuations at rest. Proc Natl Acad Sci2003;100:11053–8.

60] Klimesch W. EEG alpha and theta oscillations reflect cognitive and memoryperformance: a review and analysis. Brain Res Brain Res Rev 1999;29:169–95.

61] Brain Products GmbH. BrainVision Analyzer 2.0.1 User Manual, (3rd ed.).Munich, Germany; 2009.

62] Varela F, Lachaux J, Rodriguez E, Martinerie J. The brainweb: phase synchro-nization and large-scale integration. Nat Rev Neurosci 2001;2:229–39.

63] Nunez PL, Srinivasan R, Westdorp AF, Wijesinghe RS, Tucker DM, SilbersteinRB, Cadusch PJ. EEG coherency. I: statistics, reference electrode, volume con-duction, Laplacians, cortical imaging, and interpretation at multiple scales.Electroencephalogr Clin Neurophysiol 1997;103:499–515.

64] Thornton KE, Carmody DP. Traumatic brain injury rehabilitation: QEEG biofeed-back treatment protocols. Appl Psychophysiol Biofeedback 2009;34:59–68.

65] Kononen M, Partanen JV. Blocking of EEG alpha activity during visual per-formance in healthy adults. A quantitative study. Electroencephalogr ClinNeurophysiol 1993;87:164–6.

66] Caplan JB, Madsen JR, Schulze-Bonhage A, Aschenbrenner-Scheibe R, NewmanEL, Kahana MJ. Human theta oscillations related to sensorimotor integrationand spatial learning. J Neurosci 2003;23:4726–36.

67] Gougoux F, Zatorre RJ, Lassonde M, Voss P, Lepore F. A functional neuroimag-ing study of sound localization: visual cortex activity predicts performance inearly-blind individuals. Plos Biol 2005;3:e27.

68] Friedrich EV, Scherer R, Neuper C. The effect of distinct mental strategies onclassification performance for brain–computer interfaces. Int J Psychophysiol2012;84:86–94.

69] Friedrich EV, Scherer R, Neuper C. Long-term evaluation of a 4-class imagery-based brain–computer interface. Clin Neurophysiol 2013;124:916–27.

70] Kolb B, Wishaw I. Fundamentals of Human Neuropsychology. 4th ed. New Yorkand Oxford: Freeman and Company; 1996.

71] Cutmore TRH, Hine TJ, Maberly KJ, Langford NM, Hawgood G. Cognitive and gen-der factors influencing navigation in a virtual environment. Int J Hum-ComputSt 2000;53:223–49.

72] Bland BH, Oddie SD. Theta band oscillation and synchrony in the hippocam-pal formation and associated structures: the case for its role in sensorimotorintegration. Behav Brain Res 2001;127:119–36.

73] Pfurtscheller G, Neuper C. Motor imagery activates primary sensorimotor areain humans. Neurosci Lett 1997;239:65–8.

74] Jeannerod M, Decety J. Mental motor imagery: a window into the representa-tional stages of action. Curr Opin Neurobiol 1995;5:727–32.

75] Kravitz DJ, Saleem KS, Baker CI, Mishkin M. A new neural framework for visuo-spatial processing. Nat Rev Neurosci 2011;12:217–30.

76] Sadato N, Okada T, Honda M, Yonekura Y. Critical period for cross-modalplasticity in blind humans: a functional MRI study. NeuroImage 2002;16:389–400.