cross modal plsticity of tactile perception

8/6/2019 Cross Modal Plsticity of Tactile Perception

1/12

Critical Period for Cross-Modal Plasticity in Blind Humans:A Functional MRI Study

Norihiro Sadat o,* , , Tomohisa Okada ,* Mana bu H onda,* and Yoshihar u Yonekur a* National Institute for Physiological Sciences, Okazaki; Biom edical Im aging R esearch Center, Fuk ui Medical U niversity, Fuku i; and

JS T (Japan S cience and Technology Corporation)/ RIS TEX (Research Institute of S cience and Technology for S ociety), Kaw aguchi, J apan

Received November 20, 2001

The pr imary v isua l cor tex (V1) in congeni ta l ly b l indu m a n s h a s b e e n s h o w n t o b e i n v o l v e d i n t a c t i l e d i s -r i m i n a t i o n t a s k s , i n d i c a t i n g t h a t t h e r e i s a s h i f t i nu n c t i o n o f t h i s a rea o f co r t ex , b u t t h e ag e d ep en d en cyf t h e r e o rg a n i z a t i o n i s n o t f u l l y k n o w n . To i n v e s t i -

ate t he re org an iz ed n e tw o rk , w e m e as ure d t heh an g e o f reg i o n a l c e reb ra l b lo o d o w u s i n g 3 .0 Tes l aunct ional MRI during pass ive tac t i le tasks performedy 1 5 b li n d a n d 8 s ig h te d s u bje c ts . T he re w a s i n -rea sed ac t i v i t y i n t h e p o s t cen t ra l g y ru s t o p o s t e r i o ra r ie t a l c o r te x a n d d e c r e a s e d a c t i v it y i n t h e s e c o n d -ry s o ma to s en s or y a re a i n b li nd c o mp a re d w i thg h t ed su b j ec t s d u r i n g a t a c t i l e d i sc r i mi n a t i o n t a sk .h i s su g g es t s t h a t t h e re i s a g rea t e r d eman d fo r sh ap eiscr iminat ion process ing in b l ind sub jec ts . Bl ind sub-e c t s , i rr e s p e c ti v e o f t h e a g e a t o n s e t o f b l i n d n e s s ,x h i b i t ed h i g h e r ac t i v i ty i n t h e v i su a l a s so c i a t i o n co r-e x t h a n d i d s i g h t ed s u b je c t s. V 1 w a s a c ti v at e d i nli n d s u b j e c ts w h o l o s t t h e i r s i g h t b e f o re 1 6 y e a r s o f g e , w h e r e a s i t w a s s u p p r e s s e d i n b l in d s u b j e c ts w h oo s t t h e i r s i g h t a f te r 1 6 y e a r s o f a g e d u r i n g a t a c ti lei sc ri m in a t io n t a sk . T h i s s u g g e st s t h a t t h e r st 1 6ears o f l i fe repres en t a c r i t i ca l per iod for a funct ion alh i f t o f V1 from process in g v isua l s t imul i to p rocess in gac t i l e s t i mu l i . Becau se o f t h e ag e -d ep en d en cy, V1 i sn l i k e l y t o b e t h e e n t r y n o d e o f t h e c o r t e x f o r t h eed i rec t i o n o f t a c t i le s i g n a l s i n t o v i su a l co r t i c e s a f t e rl i n d i n g . I n s t e a d , t h e v i s u a l a s s o c i a t i o n c o r t e x m a yed i a t e t h e c i rcu i t ry b y w h i ch V1 i s a c t i v a ted d u r i n g

ac t i le s t i mu l a t i o n . 2002 El sev i e r Sc i ence (USA)

INTRODUCTION

Braille reading requires converting simple tacti lenforma tion int o meaningful pat tern s th at ha ve lexicalnd semantic properties (Sadato et al., 1998). Wher easisua l letter identicat ion is routinely accomplished

within t he visual system, t he per ceptua l processing of raille ma y be mediated by the somat osensory system.lectr ophysiological and neur oimaging stu dies have

ndicated tha t, in blind su bjects, t he visual system is

used for tasks other than processing visual informa-tion. Tactile imagery or Braille reading in blind subjects causes task-related activation in the occipitalleads of th e electr oencephalogram (EE G; Uhl et al . ,1991), which suggests that somatosensory input is re-directed into the occipital area. It is known that the

primary visual cortex (V1) is activated when subjectswith congenital and early-onset blindness ( 13 yearsold) read Braille and carry out other tactile discrimi-nation tasks (Sadato et al., 1996, 1998; Lan zenberger et al., 2001). Transcranial magnetic stimulation (TMS)induces transient disruption of cortical function duringident icat ion of Bra ille letter s in ear ly-onset blind subjects ( 10 year s old) but not in sight ed subjects r eadingembossed Roman letters (Cohen et al., 1997), demon-strating the functionality of the visual cortex of blindsubjects. A woman blind from birth who was a pro-cient Braille reader sustained bilateral occipital dam-age from an ischemic stroke (Hamilton et al., 2000).Following the str oke, she wa s n ot able to rea d Bra ille,yet her somatosensory perception appeared to be oth-erwise unchanged. This case supports emerging evi-dence of the recruitment of striate and prestriate cor-tex for Braille reading in early-onset blind subjects.Different neural networks representing different mo-dalities are activat ed dur ing th e performa nce of tactilediscriminat ion ta sks by blind and sighted subjects: th etacti le processing pathways generally l inked to thesecondary somatosensory area (SII) are rerouted inblind subjects to th e vent ral occipita l cort ical regionsgenerally reserved for visual shape discrimination (Sa-dato et al., 1998). These ndings suggest a rema rka bleplasticity of the brain, potentially permitt ing a ddi-tional processing of tactile information in the visualcort ical ar eas.

Reorgan ization of bra in function ma y differ in ear ly-onset a nd lat e-onset blind subjects (Kujala et al., 2000).There is evidence of a critical period for involvement of the visual cortex: Braille reading activates V1 in early-onset but not lat e-onset blind subjects. St imulation of the visual cortex using TMS causes errors in Brail lerea ding only in ear ly-onset blind su bjects (Cohen et al.,

euroImage 16 , 389400 (2002)oi:10.1006/nimg.2002.1111, ava ilable online at ht tp://www.idealibrar y.com on

389 1053-8119/02 $35.00 2002 Elsevier Science (USA)

All rights reserved.


2/12

999), alth ough some au th ors report a ctivat ion of V1 inat e-onset bu t n ot congenita lly blind su bjects (Bu chel et l., 1998). Thu s, th e effect of th e a ge at onset of blind-ess on plast ici ty in neural substrates for tact i le dis-rimination is not fully known.

To clar ify th e crit ical per iod of th e plast ic cha nge du eo visua l deprivation, we used fun ctional MRI an d pa s-ive ta ctile tasks with a nd with out discrimina tion com-onents. Discrimination task with Brai l le characters

was adopted instead of Brai l le words (Sadato et al . ,996, 1998; Bu chel et al., 1998) because Braille wordead ing may induce recal l o f v isual features of theb jects indicated by th e words , which in tu rn mayctivat e V1 in late-onset blind subjects whereas not inarly or congenital blind. Blind subjects with variousges at onset of bl indness and sighted subjects werencluded. Sight ed su bjects exhibited par ietal an d fron-a l cortical involvemen t wit hout activat ion of th e occip-al cortex. V1 was act ive in early-onset blind ( 16ears old), but not late-onset blind ( 16 years o ld )ubjects during a tact i le discrimination task. On theth er h an d, the visua l association cort ex was activatedrespective of the age at onset of blindness.

SUBJECTS AND METHODS

We studied 15 blind subjects, 4 women and 11 men,ged 43.9 12.3 years (mean SD). Nine of th em lostheir sight before the age of 16 years, and the othersfter a ge 16 years. All blind su bjects were blind due t o

dysfun ction at th e level of th e eye or ear ly optic nerve.Their clinical characteristics are summarized in Table1. Control subjects were 8 sighted volunteers, 5 womenan d 3 men , aged 29.0 4.5 years. The subjects were allr ight-handed according to the Edinburgh handednessinventory (Old eld 1971). There was n o history of neu-rological or psychiatric illness in any of the subjects,and except for the blindness, none had any neurologi-cal de cits. The protocol was approved by the ethicalcommittee of Fukui Medical Universi ty, and al l subjects gave their written informed consent for the study.

MRI

A time-course series of 126 vol was acquired usingT2*-weigh ted , g rad ien t echo, echo p lanar imaging(EP I) sequences with a 3.0 tesla MR ima ger (VP, Gen-eral Elect r ic, Milwaukee, WI). The raw data weretr an sferred t o a para llel supercomputer (ORIGIN2000,SGI, Mount ain View, CA) for reconstr uct ion of theconsecutive t wo-dimensiona l im ages using a two-di-mensional fast Fourier t ran sform (General Electr ic).Each volume consisted of 36 slices, with a slice thick-ness of 3.5 mm an d a 0.5-mm gap, to include t he ent irecerebral and cerebellar cortex. The t ime-interval be-tween two successive a cquisit ions of th e same imagewas 3000 ms, and echo t ime was 30 ms. The eld ofview (FOV) was 22 cm. The in-plane matrix size was64 64 pixels with a pixel dimension of 3.44 3.44mm. The magnet ic sh im was op t imized such that a

TAB L E 1

Characteristics of the Subjects

ubjectno.

Age(yea r ) Sex Ca use of blindness

Age at onsetof blindness

(year)

Visualacuity a

(L/R)

Age a t s ta r tof Braille

reading (year)Tra ining(h/day)

Readingh a n d

Early-onset blind subjects ( 16 years of age)

1 58 F Microph t h a lmos 1 0/0 7 3 1/2 L/R2 39 F Microph t h a lmos 1 0/HM 6 1/2 L3 55 M Microph t h a lmos 1 0/0 6 1 L4 56 M E ye in fect ion 2 0/0 6 1/4 L5 48 F Gla u coma (post mea sles) 3 LP/0 7 1/2 L6 51 M Gla u coma (post mea sles) 9 0/0 6 1/4 L7 39 M Opt ic n erve a t roph y 10 0/0 6 1/4 L8 21 M Ret in it is pigment osa 12 HM/HM 19 6 L9 25 M Ret in it is pigment osa 15 LP/LP 7 3 L/R

Late-onset blind subjects ( 16 years of age)

10 30 M Ret rolen t a l bropla sia 20 H M/HM 20 6 L11 41 M Gla u coma 24 0/0 24 1/6 R12 42 F At rophy of r et ina a nd ch oroid plexus 27 HM/LP 14 1/2 L

13 39 M Ret in it is pigment osa 37 ND/ND 36 1/4 R14 52 M Ret in it is pigment osa 40 LP/LP 33 3 L15 63 M Ret in it is pigment osa 51 LP/LP 45 1 L

a Visual a cuity categories: the subject had no rema ining vision (0), could see only han d m ovements (HM), ha d only l ight per ception (LP),r could only see digit ( nger) number at 1 meter (ND). L, left ; R, right.

9 0 SADATO ET AL.


3/12

ue in-plane resolution of 3.44 3.44 mm was real-zed. Tight but comfortable foam padding was placedroun d th e subject s head t o minimize head movemen t.

For anatomical reference, T2-weighted fast spin echomages were obtained from each subject with locationariables identical to those of the EPIs. In addit ion,igh-resolution whole-brain MRIs were obtained withconventional T2-weighted, fast spin echo sequence. A

otal of 112 tra nsaxial images were obtained. The in-lane matrix size was 256 256, and slice thickness

was 1.5 mm, and pixel size was 0.859 0.859 mm.

actile Tasks

Braille tactile d iscrimin ation task. A session con-is ted of s ix task and s ix res t per iods , each 30 s inurat ion , a l ternat ing task and res t per iods . Brai lletimu li were pr esented pa ssively using a plastic rail on

which different pairs of two-dot sta nda rd Braille cha r-cters (center-to-center distance, 5 mm) were printed.

he ra i l was 1 .7 m long . The sk id (1 m in leng th) ,h rou g h wh ich th e r a il was m ov ed man u a l ly b y anxaminer from th e out side of the MRI gant ry, was xedn th e left s ide of the su bject s body. The su bject placedheir r ight arm across their chest , rested their thumbnd four n g er s a t a xed posi t ion on the skid, andlaced th eir r ight index nger wi th the nger pad onh e ra il (Fig. 1). The position of the rail was rst set sohat the sub ject s r igh t index nger was located be-ween two consecutive pairs of Braille cha ra cters. Theubject s left hand was p laced on a bu t ton box con-ected to a microcomputer for recording the response.

For s igh ted sub jects , a pacemaking cue was p ro-ected on to a semit ransparen t screen hung approx i-

ma tely 1.5 m from t he su bject s eyes. For th is, an LCDrojector (Epson ELP-7200L, Tokyo, Japan) was con-ected to a personal computer (Toshiba Dynabook with

Windows95, Tokyo, J apa n) in which in-house softwar ewas us ed to genera te a visua l cue which is a sma ll lled

ircle. To x the eye posit ion , the sub ject were re-uested t o gaze th e cue circle th roughout the session.his was to control eye movement, because saccadicye movement is known to suppress the activity of thetr iate cortex even in th e darkn ess (Pau s et al., 1995).or 18 s before a session, a yellow cue was presented tol low the subject t ime to posi t ion both hands. Then,uring the tact i le discrimination task, red and greenues, each 3 s in dura t ion, were given al ternat ely for0 s. When the red cue was on, the examiner slowly

moved the rai l to present passively a pair of two-dotraille characters to the subject s nger pad. The rai l

was moved three t imes in 3 s: 30 mm in the head-to-oot direction in 1 s, 30 mm in the foot-to-head directionn the next second, and 30 mm again in the head-to-footirection in 1 s. Th e speed of present at ion was a pprox-mat ely 30 mm /s. The rail moved quietly without m ak-ng a ny task-re la ted sound . The examiner a l so con-

rmed that the sub ject d id no t move the r igh t indexnger for explorat ion. When th e green cue was on, th erai l s topped moving , a nd the sub ject responded bypush ing a bu t ton with their left index n g er i f t h epair-wise characters were the same, or with their mid-

F I G. 1 . Experimenta l se tup. (Top) During each session, withth eir heads positioned in t he MRI magnet , subjects placed th eir rightarm across their chest, rested their thumb and four ngers a t a xedposition on a skid, and placed their r ight index nger with t he ngerpad on a plast ic ra i l on which Brai lle charac ters were pr inted. Thesubject s left hand was placed on a button box connected to a micro-computer for r ecording th e response . The skid (1 m in length) ,through which the ra i l was moved manual ly by an examiner f romoutside the MRI gantry, was xed on the left side of the subject sbody. The ra il was 1.7 m long. (Bott om) Details of the r ail an d skid.The upper part of the skid was open, allowing access with the indexnger to the at portion of the plast ic rail on wh ich different pairs of two-dot standard Brai lle chara c ters (center- to-center distance , 5mm) were printed. Pair-to-pair distance was 30 mm.

39 1CORTICAL PLASTICITY IN THE BLIND


4/12

le nger i f the char acters were different . Reactionmes were not measured. A 30-s rest condit ion fol-

owed, in which red and green cues were given al ter-at ely, as in t he t ask condit ion. When t he r ed cue wasn, no tactile stimulus was presented. When the greenue was on, th e subject pushed buttons with their leftndex and m iddle nger a lterna tely. The compa rison of mages collected du ring t he discrimination ta sk versu s

hose during rest periods enable correction for the ef-ects of the cue and response movement.

For blind su bjects, th e cue for a response was a touch tohe subject s left toe given every 6 s by the examiner. Aotal of 30 pairs of Braille cha ra cters were presen ted, ha lf f which were different and half of which were the same. Braille tactile nondiscrimination task. In t h e t ac -le nondiscrimination task, which was used to control

or sensorimotor effects, six-dot, instead of two-dot,ra i l le characters were p resen ted when the red cue

was g iven . When the g reen cue was on for s igh tedubjects or the touch cue was given for blind subjects,he sub ject pushed bu t tons with the left index a nd

middle nger alternately. Other variables were identi-al to th ose in the Brai lle tact i le discrimination task.ach subject underwent two different sessions (dis-rimination and nondiscrimination).

ata Analysis

Th e rst 6 vol of each fMRI session were discardedue to unst eady magnetizat ion, and the r emaining 120ol per ea ch session, 240 vol per ea ch subject were u sed

or ana lysis. The data were ana lyzed using st at ist ical

ar am etric mappin g (SPM99, Wellcome Departm ent of ognit ive Neurology, London, UK) implemented in

Mat lab (Mathworks , Sherborn , MA; Fr is ton et al .,994, 1995a,b). Following realignment, all images wereoregistered to th e high-resolution, 3-dimensional, T2-

weighted MRI, with use of th e anat omical MRI with2-weighted spin echo sequences from locations identical

o th ose of the fMRI images. The para meters for af nend nonlinear t ransformation into a template of T2-

weighted images that was already t for a standard ste-eotaxic space (MNI template; Evans et al., 1994) werestimated using the high-resolution, 3-dimensional, T2-

weighted MRI using least squares means (Friston et al.,995b). The parameters were applied to the coregisteredMRI data. The anatomically normalized fMRI data wereltered using a Gaussian kernel of 10 mm (full-width atalf-maximum ) in the x, y, and z axes.

S tatistical an alysis. Stat ist ical ana lysis was con-ucted at two levels. First, individual task-related ac-vat ion was evaluated. Second , so that in ferencesould be made at a populat ion level , individual data

were summarized and incorporated into a random ef-ect model (Friston et al., 1999).

Individual analysis. The signa l was proport iona llycaled by set t ing the whole-brain mean value to 100

ar bitrar y un its. The signal t ime-cour se of each su bject,with 240 t ime points, was modeled with two box-carfunctions convolved with a hemodynam ic responsefunction, high-pass lterin g (120 s), an d s ession effect.To test hypotheses about regionally speci c conditioneffects, the est ima tes for ea ch condition were comp ar edby mean s of linear cont ras ts of (1) discrimina tion ta sk versus rest period and (2) nondiscrimination task ver-

sus rest period. The result ing set of voxel values foreach comparison const i tut ed a stat ist ical pa ram etricmap (SPM) of the t sta tistic (SPM{ t }). The SP M{ t }wastran sformed t o the u nit n orma l distr ibution (SPM{ Z }).The threshold for the SPM{ Z } of individual analysesw a s s e t a t P 0.05 with a correct ion for mult iplecompar isons a t the voxel level for the en t ire b ra in(Friston et al., 1996).

To test for the dependency of V1 activation on age atonset of blindness, as suggested by Cohen et al., (1999),th e percent age of cha nge in t he MR signal (percenta gesignal chan ge) measur ed in V1, relat ive to t he globalmean signal, was measured on a region-of-interest ba-sis. In each subject s SPM{ Z }, compa ring discrimina -t ion task versus res t per iod , a spher ical volume of interest with a diameter of 20 mm was placed with thecenter (0, 90, 0) based on Talairach s at las. Withinth is sphere, we searched for a local maximum (in -crease) or local minimum (decrease) of signal changerela ted to the tact i le d iscr iminat ion task comparedwith t he rest condition. The sta tistical th reshold was P

0.05 corrected for multiple comparisons within thesearch volume (Fr is ton , 1997). The percen t s ignal

change evoked during the tact i le discrimination task compa red with th e rest condition relative to th e globalmean signal ( 100) was calculated with the regressor(the box-car function for the discrimination sessionconvolved with the hemodynamic input function) andthe es t imated parameters , and were p lo t ted agains tage a t onset of blindness (Fig. 2). Becau se t he percentsigna l cha nge reversed from positive to negative as th eage at onset of blindness increased past 16 years of age,we categorized the 9 subjects who lost their sight be-fore age 16 a s ear ly-onset blind su bjects and th e other6 who lost their sight after age 16 as late-onset blind

subjects. Group ana lysi s wi th random effect m odel . Th e

weighted sum of the parameter est imates in the indi-vidual analysis const i tuted contrast images, whichwere used for the group analysis (Friston et al., 1999).The contrast images obtained by individual analysisrepresent th e normalized task -related increment of th eMR signal of each subject, that is, discrimination task versus rest period and nondiscrimination task versusrest period. To exam ine t he effect of the t actile discrim -inat ion task by each group (early-onset blind, late-onset bl ind, and sighted) and the task group inter-act ion, the contr ast images of th e early-onset blind,

9 2 SADATO ET AL.


5/12

ate-onset blind, and sighted groups were entered intorandom effect model. Signi cant signal changes for

ach cont ras t were ass essed by means of t sta tistics onvoxel-by-voxel basis (Friston et al., 1995a). The re-

ulting set of voxel values for each contrast constitutedsta t i s t ica l parametr ic map (SPM) of the t stat ist ic

SPM{ t }). The SP M{ t }was t ran sformed to the un it nor-mal d is t r ibu t ion (SPM{ Z }). Th e th resh old for t h e

PM{ Z } wa s s et a t Z 3 .09 and P 0 .05 wi th aorrection for multiple comparisons at the cluster levelor the entire brain (Friston et al., 1996).

RESULTS

ask Perform ance

Task performan ce (percent age correct response) of h e ea rly-onset blind group (80.7 12.4%) wa s signif-cantly better than that of the late-onset blind (57.84.9%) an d sigh ted (59.2 12.6%) groups ( P 0.0002,ne-way ANOVA).

ndividual AnalysisIndividual SPM{ z}showed that during the discrimi-

a tion ta sk, compa red with th e rest period, V1 in ear ly-nset blind subjects was activated, whereas it was in-ib it ed in la t e -on set b lin d su bject s (F ig . 2 ). Th e

ocation of the local maximum (increase) in the early-nset blind group was x 2.2 5.6 mm , y 91.8.4 mm , and z 1.3 3.5 mm ( n 9), overlapping

with the local minimum (decrease) in the late-onsetlind group, x 2.0 4.9 mm, y 87.7 5.3 mm,n d z 2.1 2 . 0 m m ( n 6). Task performa nceigni cantly positively correlated with the activity of

V1 ( y 21.2 x 68.3, r 2 0.3902, F 1,13 8.319, P

0.0128; Fig. 3). F igure 4 shows representa t ive exam-ples from each group.

Group Analysis with the Random Effect Model

Tactile discrimination task. The bilateral inferiorand superior parietal lobules, superior and inferioroccipita l gyri, fus iform gyri, cerebellum , an d pr efront al

areas; the left primary sensorimotor area and postcen-tral gyrus; and the r ight dorsal premotor area (PMd)and supplementa ry m otor a rea (SMA) were act ivatedin both early-onset and late-onset blind subjects (Fig.5). In sighted subjects, the bilateral inferior and supe-rior parietal lobules, postcentral gyri, and PMd; the leftventral premotor cortex, thalamus, and inferior pre-frontal regions; and the r ight dorsolateral prefrontalarea were active during the tactile discrimination task (Fig. 5). Activation of V1 during the tactile discrimina-t ion task was more prominent in early-onset than inlate-onset blind or sight ed su bjects (Fig. 6). The bilat -

eral dorsal to ventral visual cortices; and the left infe-rior frontal gyrus, postcentral gyrus, and superior pa-r ie t a l lob u le were more act ive d u r in g th e t act ilediscrimination task in blind subjects compared withsighted subjects, regardless of the age at onset of blind-ness (Table 2 and Fig. 7). The bilateral a nterior pari-etal operculum and secondary somat osensory cortex(SII) were more active during the tactile discriminationtask in sighted subjects compared with blind subjects(Table 2 a nd Fig. 7).

Tactile nondiscrimination task. The bilateral infe-rior frontal gyrus; t he left primary sensorimotor area(extending an teriorly to the PMd a nd SMA and poste-r io r ly to the postcen t ra l gyrus and super io r par ie ta llobule); and the r ight inferior and superior parietallobule, PMd, and cerebellum were active in both early-onset and late-onset bl ind subjects during the tact i le

FI G . 3 . Task performance (percentage correct) plotted againstpercent MR signal chan ge in V1 during a tacti le discriminat ion t ask.There was a signi cantly posit ive correlation between task perfor-

man ce and activity in V1 ( y 21.2 x 68.3, r 2

0.3902, F 1,13 8.319,P 0.0128).

FI G . 2 . Adjusted MR signal change recorded in V1 in 15 blindubjects plotted against the age at onset of blindness. V1 was morective during a tacti le discrimination task in early-onset blind sub-ects ( 16 yea r s of age) bu t was l ess act i ve i n l a t e-onse t b li ndubjects ( 16 years of age). For each su bject, a spherical volume of

te rest , with a diameter of 20 mm and a center of (0 , 90, 0), waselec ted based on the a t las of Tala i rach. The local maximum (in-ease) or local minimum (decrease) signal change r elated to ta cti lescr iminat ion was measur ed in the volume wi thin t he sphere . Theat ist ica l threshold was P 0.05, corrected for multiple compar i-

ons within the spherical volume.



6/12

FI G . 4 . Statistical parametric maps of individual analysis of neural activity in early-onset blind (left), late-onset blind (middle), andghted (right) subjects du ring th e Braille discrimination task compar ed with th at during t he rest period. A represent at ive case is shown for

ach group. The ta sk-related increase in MR signal (activation; shown in red) an d th e ta sk-relat ed decrease (inhibition; shown in l ight blue)ere superimposed on sagittal (upper row) and transaxial sections of the T2-weighted high-resolution MRI of each individual. Only pixel

eas th a t were signi cantly ( P 0.05) differen t bet ween conditions, with a corr ection for mult iple compar isons at th e voxel level, are sh own.MRI dat a were norma lized into t he st ereotaxic space. The blue l ines indicate t he pr ojections of each section tha t cross in the center of V1.he Talairach coordinates are: x 4 mm, y 90, and z 0. The yellow arrows indicate the calcarine ssure .

FIG. 5 . Statistical parametric maps of average neural activity within each group during the Braille discrimination task compared withat during t he rest per iod. In ear ly-onset blind (top) and la te-onset blind (second row) subjects , t ask-rela ted increases in MR signalctivation) were superimposed on t hree orthogonal sections of T1-weighted high-resolution MRIs u nrelat ed t o the subjects of the presentudy. fMRI da ta were norma lized into t he st ereotaxic space. The blue l ines indicat e th e projections of each section t hat cross in the cent erf V1. The Talairach coordinates are: x 8 m m , y 90, and z 0. Z score is as indicated by the color bar; statist ical signi cancecreasing a s red proceeds to whi te . Sta t i s t ica l parametr ic maps of average neura l ac t ivity combining ear ly-onset and la te-onset blind

ubjects (third row) and that of sighted subjects (bottom) were superimposed on surface-rendered, high-resolution MRIs unrelated to theubjects of the present study. The statist ical threshold was P 0.05 with a correction for multiple compar isons at the cluster level.

FI G . 6 . There was more prominent ac t iva t ion in ear ly-onset than in la te-onset bl ind subjects during t he tac t i le discr iminat ion task.pper row, focus of activation in pseudocolor functional MRIs superimposed on high-resolution anatomical MRIs in sagittal and coronalanes, as indicated by the blue l ines that cross at ( 8, 92, 2). The statist ical threshold was P 0.05, corrected for multiple comparisons the cluster level. Z score is as indicated by the color bar; sta tist ical signi cance increasing as red proceeds to white. Lower left , stat ist ical

ara metric ma ps ar e shown in st anda rd a nat omical space. The 3-dimensional informa tion was collapsed into 2-dimensional sa gittal , coronal,nd tra nsverse images (i .e., maximum int ensity projections viewed from t he r ight, back, and t op of the br ain). Lower right, t he per cent signalhange in V1 ( 8, 92, 2) in early-onset blind (E), lat e-onset blind (L), and sighted (S) groups. In this group an alysis, the increase in signa lhange was smaller than that in the individual analysis of local maximum and local minimum foci (see Fig. 2).


7/12

nondiscrimination task. The bi lateral postcentral gy-ru s an d SII; the left SM1; and t he right inferior front algyrus and cerebellum were act ive in sighted subjectsduring the tactile nondiscrimination task (Fig. 8).

The left SI I of s igh ted sub jects was more p romi-nently a ct ive during t he nondiscrimination task com-par ed with blind subjects (Table 2, Fig. 9).

DISCUSSION

The results of the present st udy using fMRI show thatthere is a critical period from birth to approximately 16years of age for reorganization of the V1 to functionduring tactile discrimination tasks. People who lost theirsight before age 16 years exhibited increased activity inthe V1 during a tactile discrimination task. In contrast,people who lost their sight after age 16 years and sightedcontrols exhibited decreased activity in the visual cortexduring the same t actile ta sk.

Task Design

Previous studies (Sadato et al., 1996, 1998; Cohen et al., 1999) utilized tactile discrimination tasks that in-volve active exploration, which include inherent vari-ability that cannot be completely controlled for usingsubtra ct ion paradigms (Sadato et al., 2000). It is dif -cult , t herefore, to determine wheth er the task-relatedact ivi ty measured dur ing th ese tasks i s sensory ormotor, or both (Kujala et al., 2000), esp ecially becau seenhan ced movement-related potentials have been ob-served in bl ind subjects, compared with sighted sub-

jects (Leht okoski et al., 1998). To exclude the effect of motor control, the present study eliminated active n -ger movement for both the discrimination and nondis-crimination tasks. The results of the present stu dy areconsistent with those of previous studies that utilizedactive explorat ory tasks (Sadato et al . , 1996, 1998;Cohen et al., 1999), con rming that posterior act iva-t ion in b l ind sub jects i s due to sensory ra ther thanmotor processes.

Perform an ce Differen ces

Performa nce on the Brai lle tact i le discriminationtask was signi cantly better in the early-onset than inthe late-onset blind and sighted groups. P erforma nceof the last two groups did not differ signi can tly. Be-cau se perform an ce of the discrimination t ask involvinge x p lor a t or y m o ve m e n t i s e xp e r i en c e d e p e n d e n t(Heller, 1989), th e elimina tion of th e explora tory m ove-

FIG. 7 . Areas more prominently activated dur ing the tactile discrimination ta sk in all blind subjects compared with sighted subjects (red) and inghted subjects compared with blind subjects (blue). The areas are superimposed on surface-rendered high-resolution MRIs viewed from the back, top,ght, and left. All blind subjects exhibited activation of the bilateral occipital cortex (middle and inferior occipital gyri and fugiform gyrus), left inferior

ontal gyrus, postcentral gyrus, and superior parietal lobule, more prominent than sighted subjects. The SII and parietal operculum was suppressedlaterally in blind subjects. Statistical threshold is P 0.05, corrected for multiple comparisons at cluster level, Z 3.09.

FI G . 8 . Stat ist ica l parametr ic maps of average neura l ac t ivi tyithin each group during th e nondiscriminat ion t ask compar ed witha t dur ing the rest period. A combined stat ist ical par amet ric map of l blind su bjects (top) and one of all sighted su bjects (bottom) were

uperimposed on surface rendered high-resolution MRIs unrelated the su bjects of th e present study viewed from th e right, back, andft . The sta t i s t ica l threshold was P 0.05 with a correction forultiple comparisons at the cluster level.FI G . 9 . There was more prominent ac t iva t ion in SII in sighted

ubjects than in early-onset or late-onset blind subjects during thectile n ondiscriminat ion task . Focus of activat ion on a pseudocolor

unctional MRI superimposed on a high-resolution anatomical MRI

the sagit ta l (upper le ft ) , coronal (upper r ight ), an d t ran saxia lower left) planes, as indicated by the blue l ines that cross at ( 46 ,30, 24) corresponding to SII. Activity level is as indicated by the

olor bar; a ctivity increasing as red pr oceeds to white. The stat ist icalreshold was P 0.05, corrected for multiple comparisons at theuster level. Lower right, th e percent signal chan ge in th e SII ( 46 ,30, 24) in ea rly-onset blind (E), late-onset blind (L), and sighted (S)

roups.



8/12

ment aspect in the present study makes i t l ikely thaterforma nce differences are related to the perceptualomponent. The absence of performance differences in

ight ed an d lat e-onset blind subjects is consisten t withrecent study showing that early-onset ( 5 years old)ut not late-onset ( 5 years old) blind Braille readersan detect a signi cantly ner offset in the alignmentf a row of three embossed dots, compared with sightedubjects (Gran t et al., 2000). This result illustrates thempact of th e age tha t visual deprivation begins and it selation to a critical period for tactual acuity, at leastor discrimination of passively pr esented Braille cha r-cters.

Age Dependency of V1 Activation

The resu l t s o f the p resen t s tudy suppor t the ideahat involvement of V1 during tact i le discriminationepends on the age at onset of visual deprivation. The

MRI signa l increased in early-onset blind subjects an decreased in lat e-onset blind subjects, reversing a t ap-roximately 16 years of age, which is consistent withesults of TMS and PET studies (Cohen et al., 1999).

Although task performance was positively correlatedo th e act iv it y of V1 , two la t e b lin d su bject s wh ohowed equivalent performance to the early blind dideveal n egat ive response in the V1. And hence theivergent ndings of V1 are dif cult to be explained byerforma nce alone. Evidence of the functional rele-

vance of the visual cortex, including the V1, to tactilediscrimination in early-onset blind subjects was r stshown using TMS (Cohen et al., 1997), and later con-

rmed by a case report of infarct ion of the occipi talart ery area (Ham ilton et al., 2000). Taken togeth er, itis concluded that the V1 of early-onset blind subjects isfun ctiona lly relevan t.

There is a previous report t ha t a Bra ille reading ta sk act ivated both str iate and extra str iate visual cortex inlate-onset blind subjects, but only t he extra str iate vi-su al cort ex in congen ita lly blind s ubjects (Bu chel et al.,1998). The authors speculated that activation of V1 inlate-onset blind subjects ma y be due t o visual ima geryin subjects with early visual experience. These seem-ingly contr adictory results have at least two possibleexplanations. First , there may be a difference in thecontrol condit ions used in the studies. In the presents tudy, we u t i l ized a res t per iod as a con t ro l , ra therthan an auditory discrimination task, as used in theirstudy, because there is evidence t hat auditory st imuliact ivates visual areas in bl ind subjects. Roder et al .(1999) found that blind subjects showed sound local-ization abilities that were superior to those of sightedcont rols at far latera l locat ions. Electr ophysiologicalrecordings obtained at the same time suggested poste-rior shift of the early spat ial at tent ion mechan isms inthe blind su bjects. Results from a study using ma gne-toencephalogra phy, showed t ha t th e visual ass ociation

TAB L E 2

Cortical Areas Differentially Activated during a Passive Tactile Discrimination Task and a Nondiscrimination Task in Blind (Both Early-Onset and Late-Onset, n 15) Subjects and Sighted ( n 8) Subjects

Clust er level Voxel level x

(mm) y

(mm) z

(mm)

Location

P size P t Z Side Ar ea

Tactile discrimination: Blind Sighted

0.001 323 0.001 8.92 5.68 32 88 2 R GOm (19)0.001 1761 0.002 6.92 4.94 34 78 4 L GOi (18)

0.005 6.41 4.72 42 64 16 L GF (37)0.001 384 0.006 6.31 4.67 48 60 14 R GF (37)0.002 95 0.034 5.4 4.23 46 14 24 L GFi (44)0.001 261 0.043 5.27 4.16 44 18 32 L GPoC0.003 89 0.135 4.64 3.81 20 46 48 L LPs (7)

Tactile discrimination: Sighted Blind

0.001 184 0.004 6.14 4.59 64 16 30 R SII0.001 94 0.009 5.70 4.38 42 4 18 L Pa r iet a l opercu lum0.002 49 0.020 5.29 4.17 52 6 10 R Pa r iet a l opercu lum0.001 128 0.028 5.11 4.07 58 22 28 L SII

Tactile n ondiscriminat ion: Blind Sighted

0.026 250 0.18 5.40 4.19 46 30 24 L SII

Note. Abbreviations: GF, fusiform gyrus; GFi, inferior fronta l gyrus; GOi, inferior occipital gyrus; GOm, middle occipital gyrus; GPoC,ostcentra l gyrus; GTi, inferior temporal gyrus; LPs, superior pa rietal lobule; SII, secondary somat osensory cortex. Nu mbers in pa rent hesese Brodmann s areas. All P values a re corrected for mult iple comparisons. Height thr eshold, Z 3.09, P 0.001. L, left ; R, right.

9 6 SADATO ET AL.


9/12

ortices of early-onset blind subjects are activated dur-ng auditory discrimination tasks (Kujala et al., 1995),ndicat ing tha t these subjects exhibit auditory-to-vi-ual cross-modal plasticity (i.e. , the occipital associa-on areas, generally involved in dorsal-stream visualrocessing, are act ive during auditory local izat ionasks). This cross-modal plasticity appears to be inde-endent of the age at onset of bl indness, because re-

ordings of event-related-potentials also indicate thatos ter ior b ra in areas are involved in act ive sound-han ge detect ion in both early-onset and late-onsetlind subjects (Kujala et al . , 1997). In total ly blindubjects, auditory st imulat ion act ivates the occipitalrea , which su ggests th at deafferent ed posterior visua lreas are recru i ted to car ry ou t aud i tory funct ionsLeclerc et al., 2000). Finally, a P ET stu dy showed th athe occipital cortex is involved during sound localiza-on ta sks in congenita lly blind s ubjects (Weeks et al.,000). Taken t ogether, th ese results indicate th at V1 inarly-onset blind subjects (Bu chel et al., 1998) may bective during both tactile tasks and auditory tasks, butn different ways, so if auditory-related activity is sub-a cted from act ivity evoked during a Brai lle t act i le

ask, th e results may be skewed. In the pr esent study,n the other h and, th e rai l moved quiet ly without anyask-related auditory st imulat ion. And hence the re-u lts clearly indicate t he effect of ta ctile sha pe discrim-nat ion alone: V1 was active du ring t he t actile discrim-nation task only in early-onset blind subjects, not inate-onset blind subjects.

The second explanation of the seemingly contradic-

ory ndings i s re la ted to the character i s t ics o f thea ctile ta sk condition. Heller (1989) studied t he cont ri-ution of visual experience to tactile perception in con-enitally blind, late-onset blind, and sighted subjects.ate-onset blind subjects were far better than the con-enital ly bl ind or sighted subjects at tact i le pictur edenti cation. The author concluded that the superior-y of late-onset blind subjects is due to visual exposure

o drawings and the rules of pictorial representat ion,which may be helpful in tact i le picture identi cationwhen combined with tactual experience. Performancen tact i le mat ching accura cy was similar for sightedubjects and both groups of bl ind subjects, however,ndicating that visual experience is clearly not neces-ary for ef cient ta ctile form p erception (Heller, 1989).h is i s consisten t with the ndings of th e p resen ttudy, as i t u t i l ized a sor t o f tac tual fo rm match inga sk in wh ich perform an ce of lat e-onset blind subjects

was not better than that of early-onset blind subjects,nd V1 was not act ivated. Bu chel et al . (1998), how-ver, utilized Braille word reading, which may induceecall of visual featu res of the objects indicat ed by t he

words , which in tu rn may act ivate V1 in la te-onsetlind subjects. We conclude tha t the difference be-ween th e r esu lt s of t h e p resen t s t u d y an d th ose o f

Bu chel et al . (1998) is due to differences in task andcontrol conditions.

Different Neural Activity Recorded from Blind and Sighted Subjects during Tactile Discrimination

Parietal cortex. Compared with sighted subjects,blind subjects, irrespective of age at onset of blindness,

exhibi ted more prominent act ivat ion of the left post-centra l gyrus, bilateral posterior parietal cortex, an dassociation visual cortices during the tactile discrimi-nation task. The sighted subjects, however, exhibitedmore prominent activation of the bilateral SII.

The most pr ominen t a ctivation of the left p ostcentra lgyrus in the present study was close to the postcentralsulcus presumably corresponding to Brodman n s area(BA) 2. In macaque monkeys, lesions of BA 1 affectedonly texture discrimination, and of BA 2, only size orsh ap e ta sks, bu t r emoval of BA 3b, severely affected a lltasks (Randolph and Semmes, 1974; Carlson, 1981). As

sensory information is t ransferred from S1 to BA 5,transformation of that information begins in BA 1 andBA 2 (Pearson and Powell, 1985; Shanks et al., 1985).

The superior parietal lobule (LPs) is designated asBA 7 , and area 7 o f macaque monkeys i s sa id to behomologous to BA 7 of humans (Haxby et al., 1991).Neuronal populations in the LPs posterior to the post-central sulcus in monkeys are responsive to somaes-thet ic st imuli from both the skin a nd joints (Sakata et al., 1973; Hyva r inen and Poranen, 1974; Mount cast leet al., 1975; Robinson and Burton, 1980). Neurons inthis cortical area encode the kinematics (e.g., position,d ir ect ion , an d d isp lacemen t ) of t h e u p p er limb s(Kalaska et al . , 1990). In an act ivat ion s tudy us ingPET, a passive tact i le st imulus on the r ight ngertipactivated the left LPs (Burton et al., 1997).

Another PET st udy (Roland et al., 1998) showed tha tsomatosensory perception of form (length and shape) isrelated t o activat ion of th e ant erior pa rt of th e intra pa-rietal sulcus, and that perception of roughness is re-lated to activation of the SII, which suggests the exis-tence of separa te neur ona l circuitry for processing th edifferent somatosensory submodalities of microgeom-etry and macrogeometry. Other neuroimaging stu diesshowed that the contralateral postcentral gyrus, LPs,and the cortex l ining the anterior part of the intrapa-rietal sulcus were activated speci cally du ring hapt icprocessing of shape and length of objects (Roland andLarsen , 1976; Seitz et al., 1991; O Sullivan et al., 1994;Hadjikhani and Roland, 1998) or during non-Brai l leta ctile shape discrimina tion (Sadat o et al., 1998, 2000).The fact that there was more prominent act ivat ion of the postcentra l gyrus to posterior parietal cortex andless act ivat ion of the SII in b lind sub jects than ins igh ted sub jects in the p resen t s tudy, suggests thatthere is greater dema nd for shape discrimination pro-cessing in blind subjects.



10/12

Occipital cortex. Th e r esu l t s o f t h e p resen t s t u d yon rmed th a t t h e v isu a l a s soci a t ion cor t ex i s i n -olved in tact i le discrimination in bl ind subjects i r-espect ive o f the age a t onset o f v isual depr ivat ion .h is ob serv a t ion su g ges t s t h a t t h e d y n amics of t h e

eorgan izat ion of bra in functions d ue t o visua l depri-a t i on may b e m ed ia t ed b y t h e v isu a l a s soci a t ionortex. Maunsell et al . (1991) have repor ted that in

V4, a v isual area , neuronal act iv i ty can be affectedy ha p t ic in forma t ion of or ien ta t ion u t i lized t o per-orm a v isual o r ien ta t ion-match ing task . They sug-es ted that the in format ion encoded in neuronal ac-vi ty r ep resen t s g en era l or i en t a t i on r egard l ess of

h e sensory modali ty th rough which th e in forma t ionr r i ve s. H u m a n n e u r oim a g in g s t u d ie s com b in e d

wi th TMS a l so su p p o r t t h e i d ea t h a t t h e v i su a l a s -ociat ion cortex is affected by nonvisual sensory in-orma tion. Disru ptin g function of the occipi ta l cort exn sighted subjects by means of focal TMS interferes

with t ac t il e d i scr imin a t ion of g ra t i n g or i en t a t i onZangaladze et al . , 1999) . The act ivated area i s lo -a t e d i n t h e e xt r a s t r i a t e cor t e x n e a r t h e p a r ie t o-cci pi t a l s u lcu s . T h e a u t h o r s s p ecu l a t ed t h a t i n -o lvement o f the v isual cor tex may be bene cial inh e d iscr iminat ion of ma crogeometr ic featu res suchs or ien ta t ion an d sha pe, becau se p r ocess ing of or i -n t a t ion a n d s h a p e d is cr i m in a t i on g en e r a lly in -olves vision. Performance of a tact i le object recog-it ion t a s k b y s igh t e d s u bje ct s m a y in volve aetwork of cor t ical reg ions subserv ing soma tosen-ory, motor, visual , and, at t imes, lexical processing

Deibert et al . , 1 9 99 ). Th e au t h o r s s t r e s sed t h a t t h eisua l cor t ices may be invo lved in topographic spa -al processing of tactile object recognition. Amedi et l . (2001) found that bo th tact i le and v isual ob jectecog n it i on t a sk s act i va t ed t h e v en t r a l v i su a l p a th -

way. They specu lat ed th at th i s b imodal a ct ivat ion inh e occip ito temporal reg ions re ects s to red ob ject -e la ted v isual in format ion that can be accessed v iaues f rom the somatosensory modal i ty. Th is i s con-i s t en t wi th t h e p resen t n d in g t h a t t h e r e wa s n occip i ta l ac t ivat ion in s igh ted sub jects because theask was tact i le discrimination without object recog-i t ion, which pr ima ri ly rel ies on vision (Amedi et al.,001).

Involvement o f the v isual associa t ion cor tex dur-ng performa nce of tact i le ta sks by b l ind sub jects i sonsis ten t wi th th e concep t of am odal p rocess ing of pat ial informa tion (Heller, 1989). The pr esent s tu dyncluded sub jects wi th a re la t ively large var ie ty o f i su a l d i s tu rb an ces . A l l b u t o n e b l in d su b jec t h ad

ost p at ter n vision. Several others ha d defect ive l ightr motion perception, or both. One subject with pig-

ment ary degenerat ion of th e re t ina h ad d ig it ( nger)u mb er p e rcep t io n a t a d i s t an ce o f 1 m. Nev er th e -ess , a ct ivat ion of th e v isua l a ssocia t ion cor tex dur-

ing the tact i le d iscr iminat ion task was consis ten t lyobserved in al l b l ind subjects. This nding suggeststh at l igh t percep t ion speci cally may not be cri t icalfo r t h e ac t i v a t io n , b u t a d ec l in e i n t h e h ig h er p ro -cessing of vision is . Fur th erm ore, involvement of th ev isual associa t ion cor tex i s more p rominent dur ingthe d iscr iminat ion task than dur ing the nondiscr im-inat ion task . Th is ind icates th at cross-modal recru i t -ment of the v isual cor tex i s enh anced by the d iscr im-inat ion process . Thus , the v isual associa t ion cor texma y be relat ed to th e shape discrimina tion process ina modal i ty - independent manner.

Involvem ent of V1. In t h e p resen t s t u d y, t actu a lactivation of V1 depended on age at onset of blindness,but activation of the visual association cortex did not.Thus , V1 is un likely to be the entry n ode of th e cort exfor tactile signals redirected into visual cortices aftervisual deprivation.

V1 is a topograph ically organ ized low-level cort ex

(i.e., early in the visual processing sequence) that re-ceives h igh resolut ion inform at ion dur ing bottom-upperception, en abling effective edge-detection an d re-gion-organ izing processes (Felleman an d Van Essen,1991). A visua l imagery task th at requires one to visu-al ize pat terns that depict information such as length,width, orientat ion, and the amount of space betweenb ar s, act i va t ed V1 in a fu n ct ion a lly r e levan t way(Kosslyn et al., 1999). This nding suggests tha t st oredinformat ion can evoke v isual pat terns in re la t ivelylow-level v isual areas dur ing imagery, p romot ingsha pe pr ocessing. Becau se th e m ajority of visual a reasin t he m acaque m onkey have r eciprocal conn ections toother visual areas, receiving information from the ar-eas t o which th ey send inform at ion (Felleman an d VanEssen, 1991), it is possible that the top-down process-ing dur ing v isual imagery i s mediated by the v isualassociation cortex.

Considering tha t both tact i le and visual processesar e represent ed in t he visua l association cortex, visualand tactile processing may be competitively balancedin the associa t ion cor t ices where the inpu ts ad join(Rauschecker, 1995). Therefore, visual deafferentation

causes less demand on the bot tom-up processing of vision, which may in tur n introduce opportu nity forexpansion of the tact i le representat ion in the visualassociation cortex.

Taken together, th e results of the present stu dy maybe int erpreted as follows. In blind subjects, wh ose bot-tom-up visual processing is int errupted, tact i le shapediscrimination processing expands into t he visual as-sociat ion cort ex. In ea rly-onset, but not lat e-onset blindsubjects, V1 is also recruited in a functionally relevantway, as in t op-down processing dur ing visual ima gery,result ing in bet ter performance on shape discrimina-tion in early-onset blind subjects.

9 8 SADATO ET AL.


11/12

ACKNOWLEDGMENTS

The authors are apprecia t ive of the subjec ts part ic ipa t ion andank Mr. K. Kubota, Fukui Blind School, for advice regarding thesk design, an d B. J . Hessie, E.L.S., for skillful editing. This st udyas supported in par t by a research gran t for t he Research for theu tu re Program from the Japan Society for the Promotion of Sci-nce (JSPS-RFTF97L00203).

REFERENCES

medi, A., Malach, R., Hen dler, T., Peled, S., and Zohar y, E. 2001.Visuo-haptic object-related activation in the ventral visual path-way. Nat. Neurosci. 4: 324 330.

u chel, C., Pr ice, C., Fr ackowiak, R. S., and Friston, K. 1998. Dif-ferent activation patterns in the visual cortex of late and congen-itally blind subjects. Brain 121 : 409 419.

urton, H., MacLeod, A.-M. K., Videen, T. O., and Raichle, M. E.1997. Mul t iple foci in par ie ta l and fronta l cor tex ac t iva ted byrubbing embossed gra t ing pat terns across ngerpads: A positronemission tomography study in humans. Cereb. Cortex 7: 317 .

arlson, M. 1981. Characteristics of sensory de cits following lesionsof Brodmann s areas 1 and 2 in the postcent ra l gyrus of Macacamula t t a . Brain Res. 204 : 424 430.

ohen, L. G., Celnik, P., Pascual-Leone, A., Corwell , B., Faiz, L.,Dambrosia, J. , Honda, M., Sadato, N., Gerloff, C., Catala, M. D.,and Hallett , M. 1997. Functional relevance of cross-modal plast ic-i ty in bl ind h umans. Nature 389 : 18 0 183.

ohen, L. G., Weeks, R. A., Sadato, N., Celnik, P., Ishii , K., andHallett, M. 1999. Period of susceptibility for cross-modal plasticityin th e bl ind. Ann. Neurol. 45 : 45 1 460.

e iber t , E . , Kraut , M., Kremen, S. , and Ha rt , J . 1999. Neura l pa t h-ways in tactile object recognition. Neurology 52 : 1413 1417.

vans, A. C., Kamber, M., Collins, D. L., and MacDona ld, D. 1994.

An MRI-based probal ist ic a t las of n euroanatomy. In Magnetic Resonance Scanning and Epilepsy (S. D. Shorvon, Ed.), pp. 263 274. Plenum Press, New York.

e l leman, D. J . , a nd Van Essen, D. 1991. Dist r ibuted hierarchica lprocessing in the primate cerebral cortex. Cereb. Cortex 1: 1 47 .

riston, K. J. 1997. Testing for anatomically speci ed regional ef-fects. Hum. Brain Mapp. 5: 133 136.

riston, K. J . , Worsley, K. J. , Fra ckowiak, R. S. J . , Mazziott a, J . C.,and Evans, A. C. 1994. Assessing the signi cance of focal activa-t ions u sing the i r spat ia l extent . Hum. Brain Mapp. 1: 210 220.

r iston, K. J ., Holmes, A. P., Worsley, K. J ., Poline, J . B., Frith , C. D.,and Frackowiak, R. S. J . 1995a. Sta t i s t ica l parametr ic maps infunctional imaging: A general l inear approach. Hum . Brain Mapp.2:

18 9 210.riston, K. J. , Ashbur ner, J . , Frith, C. D., Heath er, J. D., and Fr ack-owiak, R. S. J . 1995b. Spat ia l regist ra t ion and normal iza t ion of images. Hum. Brain Mapp. 2: 16 5 189.

riston, K. J., Holmes, A., Poline, J.-B., Price, C. J., and Frith, C. D.1996. Detecting activations in PET and fMRI: Levels of inferenceand power. Neuroimage 4: 223 235.

riston, K. J. , Holmes, A. P., and Worsley, K. J. 1999. How manysubjects constitute a study? Neuroimage 10 : 15.

rant , A. C. , Thiagara jah, M. C. , and Sathian, K. 2000. Tact i lepercept ion in blind Brai l le readers: A psychophysica l s tudy of acui ty and hyperacui ty using gra t ings and dot pa t terns. Percept.Psychophys. 62 : 30 1 312.

adj ikhani , N. , and Roland, P. E . 1998. Cross-modal t ran sfer of information between the tacti le and the visual representations in

the human brain: A positron emission tomographic study. J . N eu -rosci. 18 : 1072 1084.

Ham ilton, R., Keenan, J . P., Catala , M., and P ascual-Leone, A. 2000.Alexia for Braille following bilateral occipital stroke in an earlyblind woman. Neuroreport 11 : 23 7 240.

Haxby, J. V., Grady, C. L., Horwitz, B., Ungerleider, L. G., Mishkin,M., Carson, R. E., Herscovitch, P., Schapiro, M. B., and Rapoport,S. I . 1991. Dissocia t ion of object and spat ia l visual processingpathways in human extrastriate cortex. Proc. Natl. Acad. Sci. USA

88 : 1621 1625.Heller, M. A. 1989. Pictur e and pa tter n perception in t he sighted an d

the blind: The advant age of the late blind. Perception 18 : 379 389.Hyva rinen, J. , and Poranen, A. 1974. Function of the parietal asso-

ciative ar ea 7 a s revealed by cellular discharges in a lert monkeys. Brain 97 : 67 3 692.

Kalaska , J. F ., Cohen, D. A., Pru d homme, M., and Hyde, M. L. 1990.Parie ta l a rea 5 neuronal a ct ivity encodes movement kinemat ics,not movement dynamics. Exp. Brain Res. 80 : 351 364.

Kosslyn, S. M., Pa scual-Leone, A., Felician, O., Camposano, S.,Keenan, J . P. , Thompson, W. L. , Ganis, G. , Sukel , K. E. , andAlpert , N. M. 1999. The r ole of area 17 in visual imagery: Conver-gent evidence from PET and rTMS. Science 284 : 167 170.

Kujala, T., Alho, K., Kekoni, J. , Hamalainen, H., Reinikainen, K.,Salonen, O., Sta ndert skjold-Nordenstam , C.-G., and Na at anen , R.1995. Auditory and somatosensory event-related brain potentialsin ear ly bl ind h umans. Exp. Brain Res. 104 : 519 526.

Kujala, T., Alho, K., H uotilainen, M., Ilmoniemi, R. J . , Lehtokoski,A., Leinonen, A., Rinne, T., Salonen, O., Sinkkonen, J. , Stand-ertskjold-Nordenst am, C. G., and Naa ta nen, R. 1997. Electrophys-iological evidence for cross-modal plasticity in humans with early-and late-onset blindness. Psychophysiology 34 : 213 216.

Kujala, T., Alho, K., an d N aat anen , R. 2000. Cross-modal r eorgani-zation of huma n cortical functions. Trends Neurosci. 23 : 115 120.

Lanzenberger, R., Uhl, F., Windischberger, C., Gartus, A., Streibl,B., Edward, V., Moser, E., Deecke, L., and Beisteiner, R. 2001.

Cross-modal plasticity in congenitally blind subjects. In t l . Soc. Magn. Reson. Med., Vol 9, p. 670. Glasgow, UK.

Leclerc, C., Saint -Amour, D., Lavoie , M. E. , Lassonde, M., andLepore , F . 2000. Bra in funct ional reorganiza t ion in ear ly bl indhumans revealed by audi tory event-re la ted potent ia ls . Neurore- port 11 : 54 5 550.

Lehtokoski , A. , Kuja la , T. , Naatanen, R. , and Alho, K. 1998. En-han ced bra in activity preceding volunta ry movement in ea rly blindhumans . Neurosci. Lett. 253 : 155 158.

Maun sell , J. H . R., Sclar, G., Nealy, T. A., an d DePriest , D. D. 1991.Extrare t ina l r epresenta t ions in area V4 in th e macaque monkey.Visual Neurosci. 7: 561 573.

Mount castle, V. B., Lynch, J . C., Georgopoulos, A., Saka ta , H., a nd

Acun a, C. 1975. Poster ior par ieta l association of th e monkey: Com-mand functions for operations in extrapersonal space. J . Neuro- physiol. 38 : 871 908.

Old eld, R. C. 1971. The assessment and analysis of handedness:The Edinburgh inventory. Neuropsychologia 9: 97 113.

O Sul livan, B. T. , Roland, P. E . , and Kawashima, R. 1994. A PETstudy of somatosensory discr iminat ion in man: Microgeometryversus macrogeometry. Eur. J. Neurosci. 6: 13 7 148.

Paus, T., Marrett , S., Worsley, K. J. , and Evans, A. C. 1995. Extra-re t ina l modula t ion of cerebra l blood- ow i n t h e h u m a n vi su a lcortex: Implications for saccadic suppression. J . N europhysiol. 74 :2179 2183.

Pearson, R. C., and Powell, T. P. 1985. The projection of the primary

somat ic sensory cortex upon area 5 in th e monkey. Brain R es.356 :

89 107.



12/12

andolph, M., an d Semmes, J . 1974. Behaviora l consequence of se lect ive subtota l abla t ions in the postcent ra l gyrus of Macacamula t t a . Brain Res. 70 : 55 70 .

auschecker, J. P. 1995. Compensatory plasticity and sensory sub-sti tution in the cerebral cortex. Trend Neurosci. 18 : 36 43 .

obinson, C. J., and Burton, H. 1980. Organization of somatosensoryreceptive elds in cortical areas 7b, retroinsula, postauditory andgranu lar insula of M. fascicular is. J. Comp. Neurol. 192 : 69 92 .

o der, B., Teder-Sa leja rvi, W., Sterr, A., Ro sler, F., Hillyard, S. A.,

and Neville, H. J. 1999. Improved au ditory spatial tu ning in blindhumans . Nature 400 : 162 166.

oland, E., and Larsen, B. 1976. Focal increase of cerebral blood owduring stereognostic testing in man. Arch. Neurol. 33 : 551 558.

oland, P. E., O Sul livan, B., an d Kawashima, R. 1998. Shape androughness ac t iva te di fferent somatosensory areas in the humanbrain. Proc. Natl. Acad. Sci. USA 95 : 3295 3300.

adato, N., Pascual-Leone, A., Grafman, J., Iban ez, V., Deiber, M.-P.,Dold, G., and Hallett , M. 1996. Activation of the primary visualcortex by Braille reading in blind subjects. Nature 380 : 52 6 528.

ada to, N., Pa scual-Leone, A., Grafman , J., Deiber, M. P., Iban ez, V.,and Hallett , M. 1998. Neural networks for Braille reading by theblind. Brain 121 : 1213 1229.

Sadato, N., Iban ez, V., Deiber, M.-P., an d Hallett , M. 2000. Genderdifference in premotor activity during active tactile discrimination. Neuroimage 5: 53 2 540.

Sakata , H. , Taka oka, Y. , Kawarasaki , A., a nd Shibutan i , H. 1973.Somatosensory properties of neurons in the superior parietal cor-tex (area 5) of the rhesus monkey. Brain Res. 64 : 85 102.

Seitz, R., Roland, P., Bohm, C., Greitz, T., and Stone-Elander, S.1991. Somat osensory discrimination of shape: Tactile explorat ionand cerebral activation. Eur. J . N eurosci. 3: 481 492.

Shanks, M. D. , Pearson, R. C. A. , and Powel l , T. P. S. 1985. Theipsilater al cortico-cortical connexions between the cytoarchitec-tonic subdivisions of the pr imary somat ic sensory cor tex in themonkey. Brain Res. Rev. 9: 67 89 .

Uhl, F., F ran zen, P., Lindinger, G., Lang, W., and Deecke, L. 1991.On the functionality of the visually deprived occipital cortex inearly blind person. Neurosci. Lett. 124 : 256 259.

Weeks, R., Horwitz, B., Aziz-Sultan, A., Tian, B., Wessinger, C. M.,Cohen, L. G., Hallett , M., and Rauschecker, J. P. 2000. A positronemission tomographic study of auditory localization in the congen-itally blind. J . Neurosci. 20 : 2664 2672.

Zangaladze, A., Epstein, C. M., Grafton, S. T., and Sath ian, K. 1999.Involvement of visual cortex in tacti le discrimination of orienta-tion. Nature 401 : 58 7 590.

0 0 SADATO ET AL.

cross modal plsticity of tactile perception

Documents