Neuropsychologia 49 (2011) 34843493

Contents lists available at SciVerse ScienceDirect

Neuropsychologia

journa l homepage: www.e lsev ier .com/ loca

On the zatinhibit nizHuntin

Christian rstea Institute for C sittssb Leibniz Reseac Department o any

a r t i c l e i n f o

Article history:Received 21 June 2011Received in reAccepted 24 AAvailable onlin

Keywords:Response inhiHuntingtons dCompatibilityNeural synchrEvent-related

a b s t r a c t

Fronto-striatal loops play an important role action selection processes, especially when discordantsensory and contextual information has to be integrated to allow adequate selection of actions.

1. Introdu

Cortico-related toWichmannloops for counder degethe basal gprocesses.tons diseascaused by

CorresponE-mail add

0028-3932/$ doi:10.1016/j.vised form 15 August 2011ugust 2011e 1 September 2011

bitioniseaseeffectsonizationpotentials (ERPs)

Neurodegeneration weakens neural inter-connectivity, which compromises the precision of neuralsynchronization processes. Yet, it is widely unknown how far changes in the precision of neural syn-chronization processes are induced by only slight dysfunctions of striatal neural inter-connectivity andin how far such slight changes may affect action selection processes.

We investigated these processes in a sample of 25 pre-HDs and case-matched controls in a modiedGo/Nogo task, while assessing neural synchronization processes bymeans of phase-locking factors (PLFs)as derived from event-related potentials (ERPs).

The results show that pre-HDs only encounter problems in response inhibition, when discordant con-textual information and sensory input have to be integrated. No decits were evident, when responseinhibition can be based onmore habitual stimulusresponsemappings, i.e., when contextual and sensoryinformation were congruent. While habitual action selection is unaffected by changes in striatal struc-tures inuencing reliability of neural synchronization processes, efcient controlled processes of actionseem to be closely dependent upon highly reliable neural synchronization processes. The neurophys-iological analysis suggests that especially pre-motor inhibition processes (Nogo-N2) are affected. Thiswas most strongly reected in a decline in the degree of phase-locking in the Nogo-N2 range. Decits inpre-HDs seem to emerge as a consequence of phase-locking-behavioural decoupling. Of clinical interest,declines in the precision of phase-locking depended on the amount of the individuals mutant huntingtinexposure and predicted the probability of disease manifestation in the next ve years. This suggeststhat phase-locking parameters may prove useful in future studies evaluating a possible function as abiomarker in Huntingtons disease.

2011 Elsevier Ltd. All rights reserved.

ction

striatal loops mediate several cognitive processesthe inhibition and selection of actions (DeLong &, 2007). To investigate the relevance of fronto-striatalgnitive processes the examination patients sufferingnerative basal ganglia disease provide insights intoanglia mechanisms that mediate specic cognitiveFronto-striatal loops are early affected in Hunting-e (HD), an autosomal, dominant neurological disorderan extension of the CAG-repeat length at the 4th

ding author. Tel.: +49 234 322 4323; fax: +49 234 321 4377.ress: christian.beste@rub.de (C. Beste).

chromosome (Huntingtons Disease Collaborative Research Group,1993). Neurodegeneration starts in the striatum and is evidentbefore motor symptoms of disease become manifest (i.e., pre-manifest HD; pre-HD) (e.g. Rosas et al., 2008; Tabrizi et al., 2009).In these early disease stages, neurodegeneration compromises theinter-connectivity of striatal medium-sized spiny neurons (MSNs)(e.g. Cepeda, Wu, Andre, Cummings, & Levine, 2007; Mitchell,Cooper, & Grifth, 1999; Tabrizi et al., 2009).

Even though it is well-known that degraded neural inter-connectivity compromises the efcacy and precision of neuralsynchronization (Kitano & Fukai, 2004; Lago-Fernandez, Corbacho,& Huerta, 2001), it is unknown how far changes in the precision ofneural synchronization processes are induced by only slight dys-functions of striatal neural inter-connectivity and in how far suchslight changes may affect action selection processes.

see front matter 2011 Elsevier Ltd. All rights reserved.neuropsychologia.2011.08.024role of fronto-striatal neural synchroniionEvidence from ERP phase-synchrogtons disease gene mutation carriers

Bestea,, Vanessa Nessa, Michael Falkensteinb, Caognitive Neuroscience, Department of Biopsychology, Ruhr-University Bochum, Univerrch Centre Dortmund, Aging and CNS Diseases, Germanyf Neurology, Huntington Centre NRW, St. Josef Hospital, Ruhr-University Bochum, Germte /neuropsychologia

ion processes for responseation analyses in pre-manifest

n Saft c

trasse 150, D-44780 Bochum, Germany

C. Beste et al. / Neuropsychologia 49 (2011) 34843493 3485

Action selection requires the integration of sensory inputs withcontextual information (e.g. Bar-Gad, Morris, & Bergman, 2003;Gurney, Prescott, Wickens, & Redgrave, 2004; Redgrave & Gurney,2006). This integration is especially importantwhenhabitual actioncontrol is nto allow adselection prthe early cHuntingtonable to inteselect the a

Event-recesses of reSaft, AndricKobayashi,tion or reviprocess (i.ebition (i.e.,Domschke,& Falkenste& RidderinIt has previare dysfun(Beste, Willual stimuluGo/Nogo tasite stimulucontextualpossibly haresponse inesize that frthepre-maninhibition acially respoplan/prograDziobek, Hishould be rshould be amutation ca

On the bneural synctors (PLFs)Tallon-Baudtor (PLF) indis precise ocates that ttrials is leslocking ananeural syncgrate discoresponse indeterminedecits areeration inphase-lockiexposure (PTabrizi et atomatic staFalush, Pau

2. Materials a

2.1. Participan

A group ofdened by arecruited. Clas

expert raters assessments ofmotor signswhichwere not sufcient for the diagnosisof HD (Diagnostic Condence Level [DCL], item 17 of the UHDRS Motor Assessment)(Huntington StudyGroup, 1996). All HD participants underwent neurological inves-tigation and were scored according to the UHDRS items motor scale (MS), totalfunctional capacity (TFC) independence scale (IS) and the items verbal uency

bol dmmar-manO) witangbermulactinge. Theiers cAO])den s97; Taone timntrolage, s

udy. As werstudy, Germ.

k

used aonse ((i.e., Grticipaompatfor P

subjeain frog thisrespo

ssures alwen usrom reP stimwith tThe inpreschose

recor

signaordine locandwidorizondent cdata. Aitudeical alswaernies m=ion oft al. (1enta

analysnof thme inanticline.chievquencith th

ch (ti0ms bto stimeline wof trot adequate and controlled processes have to operateequate action selection (Redgrave et al., 2010). Actionocesses can be examined using Go/Nogo tasks. Due toortico-striatal dysfunctions in HD, even pre-manifests disease gene mutation carriers (pre-HDs) may notgrate sensory inputs with contextual information toppropriate actions.lated potentials (ERPs) reliably reect different subpro-sponse inhibition in basal ganglia diseases (e.g. Beste,h, Gold, & Falkenstein, 2008a; Bokura, Yamaguchi, &2005; Praamstra & Plat, 2001) that relate to the inhibi-sion of a motor plan/program before the actual motor., Nogo-N2), or the monitoring of the outcome of inhi-Nogo-P3) (see: Band & van Boxtel, 1999; Beste, Baune,Falkenstein, & Konrad, 2010a; Beste, Willemssen, Saft,in, 2010; Nieuwenhuis, Yeung, van den Wildenberg,khof, 2003; Roche, Garavan, Foxe, & OMara, 2005).ously been shown that response inhibition processesctions in manifest HD, but not in pre-manifest HDemssen, et al, 2010; Beste et al., 2008a), when habit-sresponse mappings underly action selection in ask. However, when one context requires the oppo-sresponse mapping than the previous context, theinformation interferes with previously strengthened,bitual stimulusresponse mappings and compromiseshibition processes. Under such conditions, we hypoth-onto-striatal circuits may become overstrained even inifest phaseofHD,which leadsdysfunctions in response

nd to a decline in the rate of correct rejections. As espe-nse inhibition processes and the revision of a motorm are more demanded in such conditions (see: Beste,elscher, Willemssen, & Falkenstein, 2009), these effectseected in the Nogo-N2 ERP component. The Nogo-N2ttenuated in pre-manifest Huntingtons disease generriers (pre-HDs).asis of these ERPs, we further examine properties ofhronization processes by means of phase-locking fac-(Beste, Kolev, et al., 2010; Roach & Mathalon, 2008;ry, Bertrand,& Fischer, 2001). Ahighphase-locking fac-icates that neural synchronization, or phase-synchronyr reliable in time and frequency, while a low PLF indi-iming and frequency of neural synchronization acrosss reliable (Roach & Mathalon, 2008). For the phase-lysis we hypothesize that timing and frequency ofhronization is less reliable, when pre-HDs have to inte-rdant contextual and sensory information to performhibition. These phase-locking decits are expected toresponse inhibition decits in this condition. As thesemost likely determined by severity of neurodegen-

pre-HDs, we expect that the individuals decline inng is related to the individuals mutant huntingtinenney, Vonsattel, MacDonald, Gusella, & Myers, 1997;l., 2009) and also on the probability to enter the symp-ge within a certain time period (Langbehn, Brinkman,lsen, & Hayden, 2004).

nd methods

ts

25 right-handed, pre-manifest HD gene mutation carriers (pre-HDs)positive gene test and absence of clinical motor symptoms weresication as absence of clinical motor symptoms was based on

test, symwere sueach preonset; Amodel (Lbehns foby subtronset aginto carronset [fEease buret al., 19(except

As cogroup inin the stprotocol

TheBochumHelsinky

2.2. Tas

Weand respversioneach pablock (c(Germanand theand refrexceedinafter thetime preblockwastrengthrefrain fthe STOto reactstimuli.trials arewas also

2.3. EEG

EEGtions accelectrodlter ba5k. Hindepenmentedan amplby technthe signaPerrin, Pof splinea precisPerrin e

Segmdomainsentatioin the tipeak quthis base

To atimefreepochs wysis epotill 60relatedand basnumber(p> .5).igit test, interference test, colour naming and word reading whichized as cognitive score (CS) (Huntington Study Group, 1996). Forifest participant the probability of estimated disease onset (Age ofhin 5 years was calculated according to Langbehn et al.s parametrichn et al., 2004). Also the expected AO was estimated using the Lang-

a. Years to disease onset (YTO) for the pre-HD subjectswere calculatedthe subjects age at the timeof investigation fromhis or her estimatedYTO for each subjectwereused todichotomizepre-manifestHDgrouplose to estimated age of onset [cEAO] and far from estimated age of(e.g. Tabrizi et al., 2009), by median-split. We also calculated the Dis-core (DBS= [CAG repeat35.5] age) for each subject (e.g. Penneybrizi et al., 2009). None of the mutation carriers was on medicationipramine 50mg and one uoxetine 20mg).

s, a group of 25 right-handed, healthy subjectsmatched to the pre-HDex, educational status and socio-economic background was enrolledll participants gave written informed consent, before any of the studye commenced. The demographical information is given in Table 1.

was approved by the Ethics Committee of the Ruhr-Universityany. The study was conducted according to the Declaration of

modied Go/Nogo task, where the usual relation between stimulusi.e., Go respond; Nogo refrain from responding) and an invertedo refrain from responding; Nogo respond) was examined withinnt. The Go/Nogo task consisted of two different blocks. In the rstible) one out of two words was presented on a PC monitor: DRCKRESS; Go stimulus) and STOP (German for STOP; Nogo stimulus)cts were asked to respond within 500ms on the DRCK stimulusm responding on the STOP stimulus. In trials with reaction timesdeadline a feedback stimulus (1000Hz, 60dB SPL)was given 1200msnse. This warning stimulus had to be avoided by the subjects. Thiswas administered to strengthen response tendencies. The compatibleays presentedrst, to generate a strongpre-potent response set and toual stimulusresponse mappings. In the second block subjects had tosponding on the DRCK stimulus and were requested to respond onulus. In each block stimuli are displayed for 300ms. The subjects hadhe thumb to the Go stimuli and to refrain from responding to Nogoter-trial interval (ITI) was jittered between 1600 and 1800ms. 180ented in each block; 120 Go trials and 60 Nogo trials. This ITI lengthn to strengthen response tendencies.

ding and analysis

ls were recorded from 64 Ag-AgCl electrodes using standard posi-g to the extended 10/20 system (Pivik et al., 1993) against a referenceted on Cz. The sampling rate of all recordings was 1kHz, applying ath of 0.0580Hz to the EEG. Electrode impedances were kept belowtal and vertical eye-movements were corrected in the EEG usingomponent analysis (ICA) (infomax algorithm) applied to the unseg-rtifact rejection procedures were applied twice: automatically, with

threshold of 80V, and visually by rejecting all trials contaminatedrtifacts. Before quantifying ERPs the current source density (CSD) ofs calculated to achieve a reference-free evaluation (Nunez et al., 1997;r, Bertrand, & Echalier, 1989) using the following parameters: order4, and the maximum degree of the Legendre polynomials n=10, with2.727. The exact mathematical procedure is explained in detail in989).tion of the data was done twice: the rst segmentation (for timeis) was done segmenting epochs of 1200ms length (200ms till pre-eGoorNogostimulus at timepoint0). Baseline correctionwasappliedterval between 200ms and stimulus presentation. All subsequentation for standard ERP time-domain analysis was done relative to

e a reliable analysis of slow frequency components in subsequenty analyses the EEG was additionally segmented into 4096ms longe Go or Nogo stimulus presentation being in the center of the anal-

me point 0). The baseline was dened in the time window 800efore delivery of Go or Nogo stimulus, which is free of activityulus or response processing in the previous trial. These epochsere used for the time-frequency analysis (see details below). The

ials included in the analyses was similar for both segmentations

3486 C. Beste et al. / Neuropsychologia 49 (2011) 34843493

Table 1Demographical data of the pre-manifest HD gene mutation carriers group and the healthy control group. The mean and standard deviation (SD) are given.

Parameter pre-HD Control

N 25 25Age 39.12 (9.5)Sex 14 female/11 maleCAG NADisease burd NAYears to ons .75 NA5-Year prob 12.05 NAUHDRS moto NAUHDRS total NAUHDRS Instr NAUHDRS Cogn NA

On trials dnegative deeP3 was deneof the Go-N2the Nogo comDziobek, et al.electrode FCzthe usual P3b,

2.4. Timefreq

Timefreqmeans of a conMorlet waveleaccording to th

w(t, f ) = A exp

where t is timysis and TF-pland f is the wperformed in tvals. For differ2f , respectivef0 = 3Hz, 2t =each trial, thewas obtainedthe complex wfrequency of tthe lower alph2011).

Maximal Tthe time interto the responstime range ofAs the ITI is jittstimulus presewas log10-tra

In the cursis was restriccalculate the pchronization p2008; Tallon-B(Kolev & Yord1 indicating peclose to 0 reeto Beste, Kolev

2.5. Statistica

Kolmogorodistributed (aunivariate anafactors electroible) were usewas used as bwere adjustedadjusted usingcomplex intervariability therestricted to bdifferences be

ults

he following, the results of the behavioural and neurophys-al data analyses are presented. The results are presented ass: Trials presenting ,PRESS in the compatible condition and in the incompatible condition are referred to as ,Go-trials;resenting ,STOPP in the compatible condition and PRESS inompatible condition are referred to as Nogo-trials.

havioural data

the behavioural data, the absolute frequency of correctlyed and correctly inhibited responses for the pre-HD and con-oup is given in Fig. 1.

ANOVA revealed an interaction Go/Nogo condition(F(1,48) =5.63; p= .022, 2 = .11). Subsequent post hoc testsd that within the pre-HD and control group, accuracyttermpa) (rerfor-HDs), wnce39.36 (10.04)14 female/11 male42.08 (1.78)

en score (DBS) 253.5 (75.25)et (YTO) 15.66 (8.3) median=13ability 17.42 (17.51) median=r score (MS) 2.80 (2.78)functional capacity scale (TFC) 12.96 (0.2)umental Scale (IS) 99.4 (1.6)itive Score (CS) 329.40 (43.35)

enoting response inhibition the Nogo-N2 was dened as the mostction within the range of 150300ms after stimulus onset. The Nogo-d as the most positive deection from 320 till 500ms. Amplitudesand Go-P3 were measured at the corresponding time point, whereponent reached its maximum (Beste, Willemssen, et al., 2010; Beste,, 2009). The N2 was measured at electrode Fz, FCz and Cz. For the P3was used. However, to underline that the Nogo-P3 is different fromalso electrode Pz was included in analysis.

uency analysis and calculation of the phase-locking factor (PLF)

uency (TF) analysis of stimulus-related potentials was performed bytinuouswavelet transform (CWT) applyingMorletwavelets. Complexts w can be generated in the time domain for different frequencies, f,e equation:(

t222t

)exp(2ift),

e, A = (t

)1/2, t is the wavelet duration, and i =1. For anal-

ots, a ratio of f0/f =5.5 was used, where f0 is the central frequencyidth of the Gaussian shape in the frequency domain. The analysis washe frequency range 0.120Hzwith a central frequency at 0.5Hz inter-ent f0, time and frequency resolutions can be calculated as 2t andly.t andf are related by the equationt =1/(2f). For example, for425ms and2f =1.5Hz; for f0 = 5Hz, 2t =255ms and2f =2.5Hz. Fortime-varying power in a given frequency band was calculated, whichby squaring the absolute value of the convolution of the signal withavelet. Frequency-relevant TF powers were extracted in the central

he delta (f0 = 3Hz) and theta frequency band (f0 = 6Hz), as well as ina frequency band (f0 = 10Hz) (see: Ocklenburg, Gntrkn, & Beste,

F power and corresponding peak power latencies were measured invals used for ERP quantication. A time window of 600800ms priore was used to estimate background noise and wavelet power in theinterest was measured normalized to wavelet power at this baseline.ered between 1600 and 1800ms this baseline, relative to the begin ofntation (i.e., a new trial), is free of the before trial activity. TF powernsformed to normalize the distributions for statistical analyses.rent study, the total power of the ERP-signal was analyzed. Analy-ted to the total power of the signal, because this measure is used tohase-locking factor, giving a measure of the reliability of neural syn-rocesses in time and frequency across trials (PLF; Roach & Mathalon,audry et al., 2001). The PLF is independent of the signals amplitude

anova, 1997). The values of PLF vary between 0 and 1, with values ofrfect phase-locking or phase alignment across trials, whereas valuesct high phase variability. The whole analysis procedure is comparable, et al. (2010).

l analysis

3. Res

In tiologicfollow,STOPPtrials pthe inc

3.1. Be

Forexecuttrol gr

Thegrouprevealewas bethe cop< .001that pein prep< .001differevSmirnov tests revealed that all relevant variables were normallyll z .4; one-tailed). Data was analyzed using repeated andlyses of variance (ANOVAs). In the repeated measures ANOVAs, thede, trial type (Govs.Nogo) and context (compatible vs. incompat-d as within-subject factors. The factor group (pre-HD vs. controls)etween-subject factor. When appropriate, the degrees of freedomusing GreenhouseGeisser correction. Post hoc tests performed wereBonferroni correction, when necessary. In Section 3 always the mostaction between the different factors is described. As a measure ofstandard error of the mean (SEM) is given. These latter analyses wereehavioural and neurophysiological parameters that showed largesttween pre-HDs and controls in the initial ANOVAs.

Fig. 1. Mean aincompatiblebars representrials. Error baupon Go-trials than upon Nogo-trials presentation intible and the incompatible condition (all t(24) >10.1;fer Fig. 1). Between-condition comparisons revealed

mance on incompatible Nogo trials was dysfunctional(t(24) =28.45; p< .001) and controls (t(24) =10.88;

hen compared with the compatible condition. Thein Nogo trial performance between the compatiblebsolute frequency of correct responses in the compatible (top) and(bottom) block, separated for the control and pre-HD group. Whitet performance on Go-trials, black bars denote performance on Nogo-rs denote standard error of the mean (SEM).

C. Beste et al. / Neuropsychologia 49 (2011) 34843493 3487

and the incompatible condition was larger in the pre-HD group(t(24) =28.45; p< .001) than the control group (t(24) =10.88;p< .001). This is underlined, when using this difference score inan independent sample t-test showing that performance declinewas strongeOn Go trialtion were e(t(24) =8.43formance band controlmance decl

Within tsubjects cloindicated bbetween sucondition athe dichtomand lprob)(RTs) did nothere wereF .4

To analychanges inblock was bmance betwor an intera

3.2. Neurop

Stimulusible and incand control

N2 effecaction elep< .005; 2

Go/Nogowas only etrodes (Fs 11.7;

In the iated N2 oncontrols: group diffe(t(48) =1.3Nogo-N2 wpatible con(calculatedible) was la(t(48) =5.11the N2 datasubjects clothe dichotop> .3).

P3-effecttrodeGo/2 = .14). EGo/Nogotrode FCz (FPz (F(1,48)was analyzboth groupspatible (t(4

p> .5) (refer Fig. 2). Upon Nogo-trials, the pre-HD group revealedlower amplitudes than controls in both conditions (compatible:t(48) =4.23; p< .001: incompatible: t(48) =4.11; p< .001) (referFig. 2A). There were generally no latency effects in the P3 data (all

; p>ummredn ta2 innditeffectd ER3) ases ithe

ionstatiosingsesed in8, w

lp tod th(1,48rouponaldo nouraompwas a

urop

efrponsce t

s desof th

todGo/N. Sub

baidenc teswas: 5.30.3)(t(2aveltrolstibleatedof an co0.3)

nce0.3)0.3;3.77were .1).he pre-HD group, there were no differences betweense and far to EAO on Go and Nogo trial performance, asy a repeated measures ANOVA using cEAO and fEAO asbject factor and Go and Nogo-trials in the incompatibles within subject factor (all F .2). Similarly, alsoization in high and low 5-year onset probability (hprob

did not reveal effects (all F .3). Reaction timest differ between the conditions (all F .3) and

also no other signicant main or interaction effects (all).ze behavioural parameters with respect to possibleperformance throughout the incompatible block, thisisected into two halves. Comparing behavioural perfor-een the halves did not reveal any main effect halves,ction with the factor group (all F .6).

hysiological data: time domain analysis

-lockedpotentials forGoandNogo trials in the compat-ompatible conditions are shown in Fig. 2A for pre-HDss separately.ts: The repeated measures ANOVA revealed an inter-ctrodeGo/Nogo conditiongroup (F(1,48) =9.23;= .19). Subsequent ANOVAs revealed that an interactionconditiongroup (F(1,48) =14.73; p< .001; 2 = .24)vident for electrode FCz, but not for the other elec-1.1; p> .3). Therefore only electrode FCz was analyzed

compatible Go and Nogo-trials, pre-HDs and controlsfer in their N2 amplitudes (Go: t(48) =0.52; p= .5;=0.5; p= .6) (refer Fig. 2A). Within each group the N2als was always larger in the compatible condition (allp< .001) (refer Fig. 2A).

ncompatible condition, pre-HDs revealed an attenu-Nogo-trials, compared to controls (pre-HDs: 8.80.7;14.50.6) (t(48) =6.77; p< .001) (refer Fig. 2A). Norences were evident for Go-trials in this condition; p= .16). Compared to the compatible condition, theas attenuated in pre-HDs and controls in the incom-dition (t(24) >4; p< .001). The degree of attenuationas the difference between compatible minus incompat-rger in pre-HDs (9.30.6), than in controls (4.50.4); p< .001). There were generally no latency effects in(all F .3). There were no differences betweense and far from EAO (all F .3), and when usingmization of the 5-year onset probabilities (all F .7; 2 = .002). Therefore, only electrode FCzed further. Independent sample t-tests revealed thatdid not differ in amplitudes upon Go-trials in the com-

8) = .18; p> .8), and incompatible condition (t(48) = .55;

Fs < 0.7In s

compabetweeNogo-Ntask co

N1data anNogo-Pprocestion ofconditpresenprocesrespondepictand POby scarevealePO8 (Fwith gattentihencebehaviwhencThere

3.3. Ne

Timing resFCz, sinysis, aresults

N2revealeband2 = .1)quencywas evpost hopowerpre-HD2.77groupstotal wto concompaattenudegreebetwee(2.49differe(2.85(t(24) =Nogo (There(all Fshigh otherefoanalys.4).ary, the Nogo-P3 was attenuated in the pre-HD group,

to the control group. This attenuation did not differsk conditions. Opposed to that, an attenuation of thethe pre-HD groupwas only evident in the incompatibleion and parallels the behavioural results pattern.s: To ensure that the effects observed in the behaviouralPs reecting response inhibition processes (Nogo-N2,

re not biased with respect to different visual attentionaln pre-HDs and controls, we examined the modula-N1 between both groups in the different experimental(compatible vs. incompatible) on Go and Nogo-trialn. The N1 is a reliable marker of visual attentionalreecting modulations of the magnitude of neural

to incoming stimuli (e.g. Mangun, 1995). The N1 isFig. 2B. The N1 was measured at electrodes PO7

hich revealed the maximum potentials as suggestedpography maps. The repeated measures ANOVA onlyat the N1 was larger at electrode PO7, compared to) =9.1; p< .001). All other main and interaction effects were not signicant (all F .5), suggesting thatprocesses are not different between the groups and

ot modulate response inhibition processes. As with thel data, neither of the above ERP-parameters differed,aring therst and secondhalf of the incompatible block.lso no interaction with group (all F .6).

hysiological data: timefrequency analysis

equencydecompositionwasperformed for ERPs reect-e inhibition processes and were restricted to electrodehis electrode reected effects in the time-domain anal-cribed above. The effects in this analysis parallel thee time-domain analysis.

tal power: The repeated measures ANOVAa signicant interaction frequency

ogo compatibilitygroup (F(2,96) =4.18; p< .05;sequent ANOVAs revealed that only in the delta fre-nd an interaction Go/Nogo compatibilitygroupt (F(1,48) =4.94; p= .01; 2 = .1). Bonferroni-correctedts revealed that in the compatible block total delta bandgenerally higher in Nogo trials (controls: 5.340.2;90.3) than in Go trials (controls: 2.880.3; pre-HD:(all t(24) >4.9; p< .001) with no difference between4) =0.3; p> .3). In the incompatible condition, Nogo-N2et power was weaker in pre-HDs (2.90.2) compared(3.770.1) (t(48) =2.99; p< .01). Compared to thecondition, the Nogo-N2 total wavelet power wasin pre-HDs and controls (t(24) >4; p< .001). Yet, thettenuation in Nogo trials (calculated as the differencempatible minus incompatible) was larger in pre-HDs, than in controls (1.570.2) (t(48) =3.01; p< .01). Noin total delta band power was evident between Goand Nogo-trials (2.90.2) with the pre-HD groupp> .2), while in controls the delta power was higher on0.1), than Go-trials (2.830.3) (t(24) =6.3; p< .001).no differences between subjects close and far to EAO

; p> .4), and also no differences between subjects with5-year onset probability. The N2 total power data,

ompletely parallels the N2-data in the time-domain

3488 C. Beste et al. / Neuropsychologia 49 (2011) 34843493

Fig. 2. (A) StimHDs. Time poiThe scalp topodifferent colou

P3 totalinteraction(F(2,96) =5.Go/Nogodelta frequepatible con(controls: 95.440.4;showing lo(t=4.1; df =higher onthan on Go-(t(24) =2.66revealed lowp< .05). TheEAO (all F .5) andwith high or low5-year onset probability

(all F .5). In summary, the results of P3 total powerwaveletrallel the results of the time-domain analysis. Similar-domain analysis, no main or interaction effects weremparing the two halves (all F .3).

hysiological data: phase-locking analysis

se-locking factors (PLFs) for each condition (Go vs. Nogotible vs. incompatible) are depicted in Fig. 3 for the pre-trol group for electrode FCz. The PLF was quantiedrespective peak in the time-domain analysis (ERPs).The repeated measures ANOVAs across the phase-tors (PLFs) in the N2 range revealed a signicantfrequency bandGo/Nogo compatibilitygroup

14; p< .001; 2 = .17). Only for the delta frequencyinteraction Go/Nogo compatibilitygroup was

C. Beste et al. / Neuropsychologia 49 (2011) 34843493 3489

Fig. 3. Plots oThe degree of pboundaries rel

evident (F(frequency b

Bonferrohigher for Npre-HDsandid not diff(t .effect in PLFGo trials (t(on compatiin pre-HDsp< .001).

Similarthere weref the phase-locking factor (PLF) at electrode FCz, separated for the control and pre-HD grhase-locking is colour-coded. The ordinate denote the frequency (in Hz), the abscissa denated to the delta (1.53.5Hz), theta (48Hz) and alpha frequency band (812Hz). The tim

1,48) =7.94; p= .007; 2 = .14). Hence, only the deltaand was analyzed further.ni-corrected post hoc tests revealed that the PLF wasogo than for Go-trials in the compatible condition in

dcontrols (all t(24) >7.24;p< .001) andalso thegroupser from each other in the PLF on Go and Nogo trials9). For the incompatible condition, the between groupwas larger for Nogo (t(48) =11.63; p< .001), than for

48) =5,83; p< .001). Furthermore, the reduction of PLFble compared to incompatible Nogo trials was stronger(0.240.02), than in controls (0.050.01) (t(48) =8.01;

to the N2 amplitude and total wavelet power data,no differences between subjects close and far to EAO

(all F .2). Contrary to pre-HDs, PLFs were higher upon, compared to Go-trials in the control group in thele condition (t(24) =6.75; 24; p< .001) (refer Fig. 3).a frequency band PLFs were generally lower in pre-ared to controls (main effect group: F(1,48) =52.37;= .52). Also, PLFswere generally lower for the incompat-red to the compatible condition (main effect condition:6.68; p< .001; 2 = .69).Analyzing phase-locking in the P3 range (i.e., from 320using the above ANOVAs did not reveal any signicant .3).

the total power analyses, no main or interaction effectsed,whencomparing the twohalvesof the incompatible

3490 C. Beste et al. / Neuropsychologia 49 (2011) 34843493

Fig. 4. Correla o (leftcondition (bot

block (all F .4). In summary, thephase-lockingdata parallelobserved in the time domain and total wavelet power

ion analyses

on analyses were performed to examine the interrela-hase-locking processes and behavioural performance.linear regression models were calculated using phase-the delta frequency band and group as regressorsbehavioural performance. As can be seen in Fig. 4,ees of phase-locking were related to better behaviourale (more correct rejections) on Nogo trials in thecondition (PLF= .674; t=6.17; p< .001), not differingoups (group =.001; t=0.01; p> .9) (F(2,47) =19.55;

r, group differences were evident when applying thel to predict behavioural performance in incompati-ials (group = .344; t=2.28; p< .001) (F(2,47) =157.02;g. 4 shows that higher degrees in phase-locking werebetter behavioural performance in controls (r= .581;.007), but no correlation was evident in pre-HDs inatible condition (r= .06; R2 = .1; p> .3). The lack oftween the degree of phase-locking and behaviourale seems therefore to be specic for the incompati-n. Using total wavelet power as regressor instead of

did not reveal any signicant regression model (all3). Also, when correlating the individual decline inng (i.e., PLFNogo-compatible PLFNogo incompatible) with theeclines in behavioural performance across conditions

(i.e., Nogocthe healthyphase-lockiformance bgroup no cofor a phase

Additionine howPLFNogo-comuals diseasprobabilitymodel by Laysis are giv

Fig. 5Abetween thto enter theexponentiaexplained vlinear regreWhen usingdictor the dto the comexponentiaear regressexplained aonly 2% betstantial corthe degree dgesting thastronger de) and Nogo-trials (right) for the compatible (top) and incompatibleompatible Nogoincompatible) there was a correlation incontrol group suggesting that a larger difference in

ngwas related to a larger difference in behavioural per-etween groups (r= .44; R2 = .17; p= .012). In the pre-HDrrelation was evident (r= .18; R2 = .2; p> .2), suggesting-locking-behavioural decoupling in pre-HDs.al regression analyses were performed to exam-far the observed declines in phase-locking (i.e.,patible PLFNogo incompatible) are related to the individ-e status. To this end, we calculated the 5-year onsetfor each pre-HD subject according to the predictionngbehn et al. (2004). The results of the regression anal-

en in Fig. 5.suggests an exponential decline of phase-lockinge experimental conditions with increasing probabilitysymptomatic disease stage in the next ve years. An

l regression function revealed a 13% higher amount ofariance (F(1,23) =34.56; p< .001; r= .79; R2 = .62) than assion model (F(1,23) =22.59; p< .001; r= .66; R2 = .43).years-to-onset (YTO) (Langbehn et al., 2004) as pre-ecline in phase-locking in the incompatible comparedpatible condition was higher with fewer YTO. Both, anl (F(1,23) =37.72; p< .001; r= .80; R2 = .64) and a lin-ion function (F(1,23) =41.43; p< .001; r=0.79; R2 = .62)comparable amount of variance with a difference of

ween the models (refer Fig. 5B). There was also a sub-relation between the disease burden score (DBS) andecline in phase-locking (r= .764;R2 = .57; p< .001), sug-

t genemutation carrierswith a higher DBS also revealedcline in phase-locking (refer Fig. 5C).

C. Beste et al. / Neuropsychologia 49 (2011) 34843493 3491

Fig. 5. Correpatible andresponses inPLFNogo-compatiblocking declinincompatible tlocking with tPart (B) denotPart (C) denotein phase-locki

4. Discussi

We examto integrateresponse infar changes

are induced by only slight dysfunctions of striatal neural inter-connectivity and in how far such slight changes may affect actionselection processes.

-manifest HD gene mutation carriers (pre-HDs) revealedntiall infpropuslyes, deion,Presubstatextuathe apprevioferencconditlation of the difference in phase-locking between the com-incompatible conditions in trials requiring the inhibition ofpre-HD subjects. The difference in phase-locking is calculated asle PLFNogo incompatible and reects the degree as to which phase-es by increasing difculty of stimulusresponse mapping in theask. Part (A) denotes a exponential relationof this difference inphase-he probability to enter manifest disease stage in the next ve years.es a linear relation with the years until the estimated age of onset.s the correlation of the disease burden score (DBS) with the decline

ng in pre-manifest gene mutation carriers.

on

ined pre-manifest HD gene mutation carriers abilitysensory inputs with contextual information to drivehibition processes. This was done to examine, howin the precision of neural synchronization processes

Therefore,tion of respexperimentsis revealedat the beharesponse inwavelet pofactor (PLFprocesses inlocking waamount ofburden scothe probabtime periodnot be dueet al., 2003seen througtask. Also ket al., 2011observed.

In the coative) on NBoth grouprevealed athat was clrange. Thesprocessing,Falkenstein2003) are spatible condof the Nogosuggests anto respondthat a highequency (i.ebetter behato the pre-reects the(Beste, Dzio1999). The pthe pattern2010). Alsoet al., 2010)controls inwas attenudifferent beformance ahave to bedue to the lbe ruled ou

In the inIn pre-HDsthe compaincompatibcompatibleboth groupresponse inhibition decits, when changes in con-ormation alter mappings of a visual stimulus toriate response and interferes with an existing andstrengthened stimulusresponse mapping. Group dif-pending on compatible vs. incompatible Go/Nogo taskwere evident on Nogo trials but not on Go trials.especially response inhibition opposed to the execu-onses was differentially affected by variations in theal conditions. An in-depth neurophysiological analy-that response inhibition decits in pre-HDs observedvioural level were mediated by specic alteration inhibition subprocesses: the N2 ERP amplitude and totalwer was attenuated and also the related phase-locking), reecting the reliability of neural synchronizationtime and frequency was weaker. The degree in phase-

s related to task performance and depended on thethe individuals mutant huntingtin exposure (diseasere; Penney et al., 1997; Tabrizi et al., 2009) as well asility to enter the symptomatic stage within a certain(Langbehn et al., 2004). The changes observed can-

to known decits in task switching in pre-HD (Aron), since response inhibition decits in pre-HDs werehout the incompatible blockwith no effects of time-onnown dysfunctions in attentional processes (e.g. Stout

) may be less relevant, since no effects in the N1 were

mpatible condition, the N2 was larger (i.e., more neg-ogo, compared to Go-trials in controls and pre-HDs.s did not differ in their Go and Nogo-N2 amplitudes andsimilar high level of response inhibition performanceosely related to the degree of phase-locking in the N2-e results suggest that pre-motor inhibition or conictas reectedby theNogo-N2 (Beste,Dziobek, et al., 2009;, Hoormann, & Hohnsbein, 1999; Nieuwenhuis et al.,imilarly efcient in pre-HDs and controls in the com-ition. According to thepre-motor inhibitionhypothesis-N2 (Falkenstein et al., 1999) a more negative Nogo-N2increase in pre-motor inhibition reducing the tendencyon Nogo-trials (Falkenstein et al., 1999). The ndingr reliability of neural synchronization in time and fre-., high PLF) in the Nogo-N2 time range is related tovioural performance (i.e., fewer false alarms) ts wellmotor inhibition hypothesis, stating that the Nogo-N2

inhibition of a mistakenly selected motor programbek, et al., 2009; Beste et al., 2008a; Falkenstein et al.,attern of results in the compatible condition replicatesfound in a previous study (Beste, Willemssen, et al.,in line with this previous study (Beste, Willemssen,the Nogo-P3 was attenuated in pre-HDs, compared to

the compatible condition and also total wavelet powerated in pre-HDs. The degree of phase-locking was nottween the groups and did not affect behavioural per-s indicated by regression analyses. The Nogo-P3 effectstreated cautiously, as oddball-effects that may emergeower frequency of Nogo, compared to Go trials cannott to modulate the P3 effects.compatible condition, the pattern of results changed:, the Nogo-N2 amplitude was attenuated relative totible condition and also compared to controls in thele condition. The Nogo-N2 attenuation, relative to thecondition, was stronger in pre-HDs than in controls. Ins, this was paralleled by decreases in the rate of correct

3492 C. Beste et al. / Neuropsychologia 49 (2011) 34843493

rejections,whichwasalsohigher inpre-HDs.As theNogo-N2atten-uationand theattenuationofNogo-N2-PLFwas stronger inpre-HDsthan in controls, pre-motor inhibition processes (Falkenstein et al.,1999) appear to bemore compromised in pre-HDs than in controls,whendiscograted. Thethat performis in contraperformancThis decoupin pre-HDssory and cpre-HDs.

NeuralconnectivitLago-Fernaability of nsubsequentlikely emerin pre-manCepeda et arecent studthe N2 in retem modulare importation (e.g. BGurney, 20affected inpossible thneuron deglack of grouto attentionemerge at avisual inputstages. In Hhuntingtin,(e.g. Thomajects with aburden scostronger deburden. SinMSN functithat the meto MSN dyspatible condonly becomdemandingible conditinetworks.

Becauserelated disedisease maof disease monset (YTOdeclines inlower probathe slope ofhigher onsedeclines infrequency mjects with ain the nextHD subjectwell as higsubgroupsuous CAG-a

possibly sensitive pre-manifest disease progression markers (seealso: Langbehn et al., 2004).

The current study adds to growing evidence that electrophys-iological techniques may serve as biomarkers in HD (for review:

n, Brpectect

Saft,Falkountn etulatctionotheses tinof fe dosteiningt forion aa difhke,allymle difanifepre-pro

ummter py inpcitsabitnsorys frple

fecteal syn seonizhat ed. Thse-lorge af cli

ded ore a

nextroveomar

wled

resem (Aound

nces

, Beniccategoy. Dem. R., W(2003)licatio642.rdant contextual andsensory informationhas tobe inte-phase-locking factor (PLF) analyses further revealedance in pre-HDs was also not related to the PLF, which

st to the compatible condition. In controls, behaviourale was still related to the degree of phase-locking.ling of behavioural performance from phase-lockinglikely reects the dysfunctional integration of sen-ontextual inputs in the incompatible condition in

synchronization processes depend on the inter-y between neural assemblies (Kitano & Fukai, 2004;ndez et al., 2001). The observed decrease in the reli-eural synchronization in time and frequency withdysfunctions in response inhibition processes most

ge due to already evident neurodegenerative changesifest HD compromising neural inter-connectivity (e.g.l., 2007; Rosas et al., 2008; Tabrizi et al., 2009). In ay by our group we argued that processes reected bysponse inhibitionmay depend on the nigro-striatal sys-ating MSNs (Beste, Willemssen, et al., 2010). As MSNsnt for the integration of various streams of informa-ar-Gad et al., 2003; Gurney et al., 2004; Redgrave &06; Redgrave, Prescott, & Gurney, 1999) and are earlyHD (Cepeda et al., 2007; Mitchell et al., 1999) it is

at the cognitive decit observed may be due to MSNeneration andweakening ofMSN interconnectivity. Thep differences in neurophysiological processes relatedal selection (i.e., N1) also suggest that decits observedpoint in processing where contextual information andare integrated and not already at attentional selectionD, MSNs reveal an increased vulnerability to mutantwhich is higher than vulnerability of cortical neuronss et al., 2011). Our correlational results show that sub-higher lifetime exposure to mutant huntingtin (diseasere) (Penney et al., 1997; Tabrizi et al., 2009) revealedclines in phase-locking than pre-HD with lower diseasece mutant huntingtin especially compromises striataloning (e.g. Thomas et al., 2011), this result underlineschanisms leading to the observed effects may be duefunction. The fact that response inhibition in the com-ition was not affected suggests that MSN dysfunctions

e critical,whenaction selectionprocesses becomemore. More habitual stimulusresponse mappings (compat-on) seem to be resolved even in compromised MSN

of a strong interrelation of the CAG-repeat lengthase burden score with measures giving an estimate ofnifestation in the next years (i.e., a higher probabilityanifestation in the next ve years and fewer years to

) (Langbehn et al., 2004)), these were also related tophase-locking. Interestingly, in pre-HD subjects withbility to enter the manifest stage in the next ve years,the regression curve was atter than in subjects witht probability in the next ve years. This suggests thatthe reliability of neural synchronization in time anday be particularly sensitive to changes in pre-HD sub-higher probability to enter the manifest stage withinve years. The fact that no differences between pre-

s dichotomized in close to EAO and far to EAO ash (hprob) and low 5-year onset probability (lprob)were evident, suggests a higher sensitivity of contin-ge measures over dichotomized measures to identify

Nguyethis resERPs reBeste,Saft, &ies accNguyeof moddysfunwhileprocesor again somFalkenexaminsuggesinhibitated viDomscferentipossibpre-mring indisease

In sencounsensorNo demore hand sepre-HDthe comis unafof neurof actiosynchrgests taffecteof phato emepling. Odepenexposuin themay pas a bi

Ackno

TheBochuCHDI f

Refere

Antal, A.tualstud

Aron, AW.Imp629adshaw, Stout, Croft, & Georgiou-Karistianis, 2010). Inseveral studies accounted for robust declines in variousing different cognitive processes (e.g. Antal et al., 2003;Andrich, Gold, & Falkenstein, 2008b; Beste, Willemssen,enstein, 2009; Mnte et al., 1997). Yet, only a few stud-ed for alterations in the pre-manifest stage (for review:al., 2010). In this stage, the neurophysiological patternion is less clear: some processes seem to be alreadyal (e.g. Beste et al., 2008b; van der Hiele et al., 2007),

r results point towards intensied neurophysiologicalhat may reect compensatory efforts (Beste et al., 2007)unctioning that leads to superior cognitiveperformancemains in the manifest stage (Beste, Saft, Gntrkn, &, 2008). In contrast to the study by Beste et al. (2007)error monitoring processes the current results do notcompensatory processes in pre-HD. However, responsend error monitoring processes are most likely medi-ferent basal ganglia-prefrontal loops (e.g. Beste, Baune,Falkenstein, &Konrad, 2010b) andmay therefore be dif-odulated inpre-HDs.Amoredetailed analysis of sucha

ferentialmodulation of different fronto-striatal loops inst HD may prove useful to understand processes occur-manifest HD and may be valuable to track pre-manifestgression in future longitudinal studies.ary, the behavioural results show that pre-HD subjectsroblems when discordant contextual information andut have to be integrated to enable response inhibition.were evident, when response inhibition can be based onual stimulusresponse mappings, i.e., when contextualinformation were congruent. The results suggest that

onto-striatal loops become overstrained by increasingxity of action selection. While habitual action selectiond by changes in striatal structures inuencing reliabilitynchronizationprocesses, efcient controlled processesem to be closely dependent upon highly reliable neuralation processes. The neurophysiological analysis sug-specially pre-motor inhibition processes (Nogo-N2) areis was most strongly reected in a decline in the degreecking in the Nogo-N2 range. Decits in pre-HDs seems a consequence of phase-locking-behavioural decou-

nical interest, declines in the precision of phase-lockingn the amount of the individuals mutant huntingtin

nd predicted the probability of disease manifestationve years. This suggests that phase-locking parametersuseful in future studies evaluating a possible functionker in Huntingtons disease.

gements

arch was supported by a FoRUM grant, University ofZ: K040-2009) to C.S. and partly by a Grant from theation and by to C.B. and C.S. We thank all participants.

zky, S., Kincses, T. Z., Jakab, K., Benedek, G., & Vecsei, L. (2003). Percep-rization is impaired in Huntingtons disease: An electrophysiologicalentia Geriatric Cognitive Disorders, 16, 187192.atkins, L., Sahakian, B. J., Monsell, S., Barker, R. A., & Robbins, T.. Task-set switching decits in early stage Huntingtons disease:ns for basal ganglia function. Journal of Cognitive Neuroscience, 15,

C. Beste et al. / Neuropsychologia 49 (2011) 34843493 3493

Band, G. P., & van Boxtel, G. J. (1999). Inhibitory motor control in stop paradigms:review and reinterpretation of neural mechanisms. Acta Psychologica (Amst.),101, 179211.

Bar-Gad, I.,Morris, G., & Bergman, H. (2003). Information processing, dimensionalityreduction and reinforcement learning in the basal ganglia. Progress in Neurobi-ology, 71, 439473.

Beste, C., Baune,B. T.,Domschke,K., Falkenstein,M.,&Konrad,C. (2010a). Paradoxicalassociation of the brain-derived-neurotrophic factor Val66Met genotype withresponse inhibition. Neuroscience, 166, 178184.

Beste, C., Baune, B. T., Domschke,K., Falkenstein,M.,&Konrad, C. (2010b).Dissociableinuences of NR2B-receptor related neural transmission on functions of distinctassociative basal ganglia circuits. Neuroimage, 52, 309315.

Beste, C., Kolev, V., Yordanova, J., Domschke, K., Falkenstein, M., Baune, B. T., et al.(2010). The role of BDNF Val66Met polymorphism for the synchronization oferror-specic neural networks. Journal of Neuroscience, 30, 1072710733.

Beste, C., Willemssen, R., Saft, C., & Falkenstein, M. (2010). Response inhibitionsubprocesses and dopaminergic pathways: Basal ganglia disease effects. Neu-ropsychologia, 48, 366373.

Beste, C., Dziobek, I., Hielscher, H.,Willemssen, R., & Falkenstein,M. (2009). Effects ofstimulusresponse compatibility on inhibitory processes in Parkinsons disease.European Journal of Neuroscience, 29, 855860.

Beste, C.,Willemssen, R., Saft, C., & Falkenstein,M. (2009). Error processing in normalaging and in basal ganglia disorders. Neuroscience, 159, 143149.

Beste, C., Saft, C., Andrich, J., Gold, R., & Falkenstein, M. (2008a). Response inhibitioninHuntingtons diseaseA study using ERPs and sLORETA. Neuropsychologia, 46,12901297.

Beste, C., Saft,compatibiJournal of N

Beste, C., Saft,tioning inevent-relanal of Neur

Beste, C., Saft,Functionalclinical Hu

Bokura, H., Yaresponse i

Cepeda, C., Wcorticostri253271.

DeLong, M. R.ganglia. Ar

Falkenstein, Mtask and th

Gurney, K., Preels of the b453459.

Huntingtons Ding a trinuchromosom

Huntington Stbility and

Kitano, K., & Furing mod267276.

Kolev, V., & Yoing event-

Lago-Fernandedenceof syNetworks,

Langbehn, D.national Hpredictionon CAG len

Mangun, G. R. (1995). Neural mechanisms of visual selective attention. Psychophys-iology, 32, 48.

Mitchell, I. J., Cooper, A. J., & Grifth, M. R. (1999). The selective vulnerability ofstriatopallidal neurons. Progress in Neurobiology, 59, 691719.

Mnte, T. F., Ridao-Alonso, M. E., Preinfalk, J., Wieringa, B. M., Matzke, M., Dengler,R., et al. (1997). An electrophysiological analysis of altered cognitive functionsin Huntingtons disease. Archives of Neurology, 54, 10891098.

Nguyen, L., Bradshaw, J. L., Stout, J. C., Croft, R. J., & Georgiou-Karistianis, N. (2010).Electrophysiological measures as potential biomarkers in Huntingtons disease.Brain Research Reviews, 64, 177194.

Nieuwenhuis, S., Yeung, N., van den Wildenberg, W., & Ridderinkhof, K. R. (2003).Electrophysiological correlates of anterior cingulate function in a go/nogo task:Effects of response conict and trial type frequency. Cognitive Affective andBehavioural Neuroscience, 3, 1726.

Nunez, P. L., Srinivasan, R., Westdorp, A. F., Wjiesinghe, R. S., Tucker, D. M., Silber-stein, R. B., et al. (1997). EEG coherency. I: Statistics, reference electrode, volumeconduction, Laplacians, cortical imaging, and interpretation at multiple scales.Electroencephalography Clinical Neurophysiology, 103, 499515.

Ocklenburg, S., Gntrkn, O., & Beste, C. (2011). Lateralized neural mechanismsunderlying the modulation of response inhibition processes. Neuroimage, 55,17711778.

Penney, J. B., Jr., Vonsattel, J. P., MacDonald, M. E., Gusella, J. F., & Myers, R. H. (1997).CAG repeat number governs the development rate of pathology in Huntingtonsdisease. Annals of Neurology, 41, 689692.

Perrin, F., Pernier, J., Bertrand, O., & Echalier, J. F. (1989). Spherical splines for scalppotential and current density mapping. Electroencephaolgraphy Clinical Neuro-

iologyT., Br3). Gualogrra, P.,rkinse, P.,overine, P.,tion toe, P., Rt al. (2ns for. J., &verviezophr. A., Griminabition. D., Sbral chetero., Pau1). Ne14.S. J., L9). Biodinalologyaudryan exf Neur, E. A.,vo celmutantics, 2Hiele,t al. (2ons dC., Andrich, J., Gold, R., & Falkenstein, M. (2008b). Stimulusresponselity in Huntingtons disease: A cognitive-neurophysiological analysis.europhysiology, 99, 12131223.

C., Gntrkn, O., & Falkenstein, M. (2008). Increased cognitive func-symptomatic Huntingtons disease as revealed by behavioral andted potential indices of auditory sensory memory and attention. Jour-oscience, 28, 1169511702.C., Yordanova, J., Andrich, J., Gold, R., Falkenstein, M., et al. (2007).compensation or pathology in cortico-subcortical interaction in pre-ntingtons disease. Neuropsychologia, 45, 29222930.maguchi, S., & Kobayashi, S. (2005). Event-related potentials fornhibition in Parkinsons disease. Neuropsychologia, 43, 967975.u, N., Andre, V. M., Cummings, D. M., & Levine, M. S. (2007). Theatal pathway in Huntingtons disease. Progress in Neurobiology, 81,

, & Wichmann, T. (2007). Circuits and circuit disorders of the basalchives of Neurology, 64, 2024.., Hoormann, J., & Hohnsbein, J. (1999). ERP components in Go/Nogoeir relation to inhibition. Acta Psychologica, 101, 267291.scott, T. J., Wickens, J. R., & Redgrave, P. (2004). Computational mod-asal ganglia: From robots to membranes. Trends in Neuroscience, 27,

isease Collaborative Research Group. (1993). An novel gene contain-cleotide repeat that is expanded andunstable onHuntingtons disease

es. Cell, 72, 971983.udy Group. (1996). Unied Huntingtons disease rating scale: Relia-consistency. Movement Disorders, 11, 136142.kai, T. (2004). Temporal characteristics of the predictive synchronouseled by spike-timing dependent plasticity. Learning and Memory, 11,

rdanova, J. (1997). Analysis of phase-locking is informative for study-related EEG activity. Biological Cybernetics, 76, 229235.z, L. F., Corbacho, F. J., &Huerta, R. (2001). Connection topologydepen-nchronizationofneural assemblies onclass I and II excitability.Neural14, 687696.R., Brinkman, R. R., Falush, D., Paulsen, J. S., Hayden, M. R., & Inter-untingtons disease collaborative group. (2004). A new model forof the age of onset and penetrance for Huntingtons disease basedgth. Clinical Genetics, 65, 267277.

physPivik, R.

(199ceph

Praamstin Pa

Redgravdisc

Redgravsolu

RedgravH., ecatio

Roach, BAn oschi

Roche, Rdiscinhi

Rosas, HCereand

Stout, J. C(20125, 1

Tabrizi,(200gituNeur

Tallon-Bhumnal o

ThomasIn viingGene

van derA., eingt, 72, 184187.oughton, R. J., Coppola, R., Davidson, R. J., Fox, N., & Nuwer, M. R.idelines for the recording and quantitative analysis of electroen-

aphic activity in research contexts. Psychophysiology, 30, 547558.& Plat, F. M. (2001). Failed suppression of direct visuomotor activationons disease. Journal of Cognitive Neuroscience, 13, 547558.& Gurney, K. (2006). The short-latency dopamine signal: A role ing novel actions? Nature Reviews Neuroscience, 7, 967975.Prescott, T. J., & Gurney, K. (1999). The basal ganglia: A vertebratethe selection problem? Neuroscience, 89, 10091023.odriguez, M., Smith, Y., Rodriguez-Oroz, M. C., Lehericy, S., Bergman,010). Goal-directed and habitual control in the basal ganglia: Impli-Parkinsons disease. Nature Reviews Neuroscience, 11, 760772.Mathalon, D. H. (2008). Event-related EEG timefrequency analysis:w of measures and an analysis of early gamma band phase locking in

enia. Schizophrenia Bulletin, 34, 907926.aravan, H., Foxe, J. J., & OMara, S. M. (2005). Individual differenceste event-related potentials but not performance during response. Experimental Brain Research, 160, 6070.alat, D. H., Lee, S. Y., Zaleta, A. K., Pappu, V., Fischl, B., et al. (2008).ortex and the clinical expression of Huntingtons disease: Complexitygeneity. Brain, 131, 10571068.lsen, J. S., Queller, S., Solomon,A. C.,Whitlock, K. B., Campbell, J. C., et al.urocognitive signs in prodomal Huntington disease. Neuropsychology,

angbehn, D. R., Leavitt, B. R., Roos, R. A., Durr, A., Craufurd, D., et al.logical and clinical manifestation of Huntingtons disease in the lon-

TRACK-HD study: Cross-sectional analysis of baseline data. Lancet, 8, 791801., C., Bertrand, O., & Fischer, C. (2001). Oscillatory synchrony betweentrastraite areas during visual short-term memory maintenance. Jour-oscience, 21, RC177.Coppola, G., Tang, B., Kuhn, A., Kim, S., Geschwind, D. H., et al. (2011).l-autonomous transcriptional abnormalities revealed inmice express-t huntingtin in striatal but not cortical neurons. Human Molecular0, 10491060.K., Jrgens, C. K., Vein, A. A., Reijntjes, R. H., Witjes-Ane, M. N., Roos, R.007). Memory activation reveals abnormal EEG in preclinical Hunt-

isease. Movement Disorders, 22, 690695.

On the role of fronto-striatal neural synchronization processes for response inhibition-Evidence from ERP phase-synchronization analyses in pre-manifest Huntingtons disease gene mutation carriers1 Introduction2 Materials and methods2.1 Participants2.2 Task2.3 EEG recording and analysis2.4 Time-frequency analysis and calculation of the phase-locking factor (PLF)2.5 Statistical analysis

3 Results3.1 Behavioural data3.2 Neurophysiological data: time domain analysis3.3 Neurophysiological data: time-frequency analysis3.4 Neurophysiological data: phase-locking analysis3.5 Regression analyses

4 DiscussionAcknowledgementsReferences