Joana Ma ga ida de Mo ais Sayal Ab eu Campos
CROSS-MODAL PLASTICITY IN THE AUDITORY
CORTEX OF THE CONGENITALLY DEAF
AN MRI STUDY USING POPULATION
RECEPTIVE FIELD ANALYSIS
Disse ação no âmbi o do Mes ado In e uni e si á io em Neu opsicologia
Clínica e Expe imen al, o ien ada pelo P o esso Dou o Jo ge Manuel
Cas elo B anco de Albuque que Almeida e pela Dou o a Zoha Tal e
ap esen ada à Faculdade de Psicologia e Ciências da Educação da
Uni e sidade de Coimb a.
Janei o de 2022
Facul y o Psychology and Educa ional Sciences o Uni e si y o Coimb a
CROSS-MODAL PLASTICITY IN THE
AUDITORY CORTEX OF THE
CONGENITALLY DEAF
An MRI s udy using
popula ion ecep i e ield analysis
Joana Ma ga ida de Mo ais Sayal Ab eu Campos
Thesis submi ed o he In e uni e si y Mas e in Clinical and Expe imen al Neu opsychology,
supe ised by P o esso Jo ge Almeida and Doc o Zoha Tal and p esen ed o he Facul y o
Psychology and Educa ional Sciences o Uni e si y o Coimb a.
Janua y 2022
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 2
Resumo
Tí ulo: Plas icidade In e modal no Có ex Audi i o de Su dos Congéni os: Análise de
Campos Rece i os Populacionais a a és de um es udo de IRM
Pala as-cha e: Neu oplas icidade; Imagem po Ressonância Magné ica Funcional; Análise
de Campos Rece i os Populacionais; Su dez Congéni a; O ganização Topog á ica
A neu oplas icidade é a capacidade mani es ada pelo cé eb o humano em eo ganiza -se e
modi ica a sua a i idade ao longo da ida. En ol e mudanças na es u u a, unção, e ligações
do cé eb o, habi ualmen e de i adas da exposição a es ímulos ex e nos ou in e nos (como
ap ende uma no a capacidade), ou como esul ado de lesões aumá icas ou p i ação senso ial.
Es a úl ima es á p esen e no caso da pe da de isão ou su dez p o unda, onde o có ex senso ial
p i ado pode se ec u ado pa a ep esen a in o mações senso iais pe encen es a ou as
modalidades. Es e p ocesso, conhecido como plas icidade in e modal, é o ema cen al des a
ese. Em es udos an e io es, e i icou-se que o có ex audi i o de su dos congéni os, mas não
de ou in es, é ec u ado du an e a e as isuais. No en an o, não é cla o se e a é que pon o essas
espos as in e modais no có ex audi i o p i ado ep esen am in o mações espaciais isuais ou
mapeiam o campo isual. Es e abalho abo da essa ques ão di e amen e a a és de es udos de
caso de IRM , com o obje i o de pesquisa e mapea ca ac e ís icas de e ino opia no có ex
audi i o, simila men e à o ma de o ganização das ep esen ações neu ais no sis ema isual.
Dois su dos congéni os e um ou in e pa icipa am numa expe iência de IRM com es ímulos
adicionalmen e u ilizados pa a mapea o p ocessamen o isual. Foi aplicado um mé odo de
e ino opia po ondas p og essi as (TWR) e de seguida uma écnica de análise de campos
ece i os populacionais (pRF), que em sido subs ancialmen e usada pa a mapea g adien es
opog á icos no cé eb o, incluindo e ino opia. Os esul ados e elam espos as isuais no
có ex audi i o dos su dos congéni os, associadas à ap esen ação dos es ímulos, mas não no
pa icipan e ou in e. Es as espos as, p edominan emen e la e izadas no hemis é io di ei o,
ep esen am o campo isual con ala e al e são ca ac e izadas po g andes campos ece i os,
cen ados nas p oximidades da ó ea. Cu iosamen e, descob imos que essas espos as a
es ímulos isuais e le iam p incipalmen e sinais BOLD nega i os no có ex audi i o dos
su dos, suge indo uma ep esen ação da in o mação isual a a és de sinais de desa i ação
in e modal. Se á discu ida a in e p e ação e ca ac e ís icas dessas ep esen ações neu oplás icas
e desa i ações neu onais, assim como os e ei os compo amen ais, uncionais e ana ómicos da
plas icidade in e modal em su dos congéni os.
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 3
Summa y
Ti le: C oss-Modal Plas ici y in he Audi o y Co ex o he Congeni ally Dea : an MRI
s udy using popula ion ecep i e ield analysis
Keywo ds: Neu oplas ici y; Func ional Magne ic Resonance Imaging; Popula ion Recep i e
Field Analysis; Congeni al Dea ness; Topog aphic O ganiza ion
Neu oplas ici y is he abili y o he human b ain o eo ganize and modi y i s ac i i y
h oughou li e. I in ol es changes in he s uc u e, unc ion, and connec ions wi hin he b ain,
ypically acqui ed ollowing ex e nal o in e nal s imuli (such as lea ning a new skill), o as a
esul o auma ic lesions o senso ial dep i a ion. The la e is p esen in he case o p o ound
blindness o dea ness, whe e he senso y-dep i ed co ex can be ec ui ed o ep esen senso y
in o ma ion belonging o o he modali ies. This p ocess, known as c oss-modal plas ici y, is
he co e subjec o his hesis. Speci ically, p e ious s udies indica ed ha he audi o y co ex
o congeni ally dea , bu no o hea ing indi iduals, is ec ui ed du ing isual asks. Howe e ,
i is no clea i and o wha ex en hese c oss-modal esponses in he dep i ed audi o y co ex
ep esen isual spa ial in o ma ion o map he isual ield. In his wo k, hese ques ions we e
add essed di ec ly in an MRI se o case-s udies, aiming o sea ch and map e ino opy ea u es
in he audi o y co ex, simila ly o he well-known o ganiza ion o neu al ep esen a ions in he
isual sys em. Two congeni ally dea and one hea ing pa icipan wen h ough a con en ional
e ino opy MRI expe imen wi h isual s imuli designed o map he isual sys em. We applied
bo h adi ional a eling-wa e analysis and a biologically inspi ed popula ion ecep i e ield
(pRF) echnique, which ha e been subs an ially used o map opog aphic g adien s in he b ain,
including e ino opy. Resul s e eal e ino opic- ela ed esponses in he audi o y co ex o he
dea , bu no in he hea ing, locked o he isual p esen a ion o s imuli. These esponses, ha
we e mos ly la e alized o he igh hemisphe e, ep esen ed he con ala e al isual ield and
we e cha ac e ized by la ge ecep i e ields, cen e ed o nea o eal a eas. In e es ingly, we
ound ha hese esponses o isual s imuli p edominan ly e lec ed nega i e BOLD signals in
he audi o y co ex o he dea , sugges ing ha isual in o ma ion migh be ep esen ed h ough
c oss-modal deac i a ion signals. The meaning and ea u es o hese neu oplas ic
ep esen a ions and neu onal deac i a ions will be discussed, as well he beha iou al,
unc ional, and ana omical e ec s o c oss-modal plas ici y in he congeni ally dea .
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 4
Acknowledgemen s
Comple ing his mas e ’s and hesis was possible due o he academic and pe sonal
con ibu ion o se e al people du ing hese challenging wo and a hal yea s. I would like o
exp ess my deepes hanks:
To D Zoha Tal, o he ime we spen mee ing and wo king oge he , always wi h a
kind-hea ed spi i . Fo he en husiasm and encou agemen . I has been a pleasu e ha ing D
Zoha as my supe iso , and I ha e lea ned so much wi h he .
To P o esso Jo ge Almeida, also my supe iso , o making i so in e es ing o discuss
neu oscience, and o always ha ing clea insigh s and ideas. Fo always wan ing he bes
possible and pushing me o wa d. Fo he oppo uni y o be pa o P oac ion Lab o almos 5
yea s now.
To P o esso Ósca Gonçal es, o he suppo h oughou his mas e on many
occasions. And o belie ing in me.
To P oac ion Lab: om pos -doc o al esea che s, PhD s uden s, esea ch assis an s,
mas e ’s colleagues, and collabo a o s. I would like o hank hei a ailabili y, eadiness o
help, in e es in my wo k, and iendship.
To he eam ha collec ed his da ase in Beijing, o he pa icipan s, and o D Alessio
F acasso, ha I un o una ely ne e me in pe son, o his expe ise in pRF models ha was so
help ul in ou analysis.
To my in e nship supe iso s, P o esso Manuela Vila , P o esso Má io Simões and
D Diana Du o, as my clinical in e nship o e lapped wi h my hesis o 9 mon hs. I’d like o
hank hem o hei unde s anding and mo i a ion du ing his ime.
To my iends: in music (OAUC, TAUC, Ci ha a) and o a e y speci ic g oup o
“always he e” iends: Ri a, Diana, Ma hilde, Simão, Inês and Chico.
To my amily, my mo he and my a he , o being my home and uncondi ional lo e
and suppo . To my b o he , ha is no only close emo ionally, bu is my pa ne in music and
neu oscience.
Aos meus a ós ma e nos e pa e nos, pa a os quais as pala as são di íceis de escolhe .
Dedico es a ese à minha a ó ma e na e ao meu a ô pa e no, que pa i am em 2021. Sei que
ica iam mui o elizes e o gulhosos com a conclusão des a e apa.
To Uni e sidade de Coimb a, Uni e sidade de Lisboa and Uni e sidade do Minho.
This in e uni e si y mas e ’s expe ience was e y ewa ding, e en in hese pandemic imes
demanding unp eceden ed esilience.

CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 5
This hesis was suppo ed by a Founda ion o Science and Technology (FCT) p ojec
g an POCI-01-0145-FEDER-030757 | PTDC/PSI-GER/30757/2017 - “SeeingEa s - Vendo
com os eus ou idos: como é que a neu oplas icidade de ido a su dez congéni a in e e e com
o p ocessamen o neu onal e com es a égias de eabili ação audi i a”, and P og ama
COMPETE.
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 6
Table o Con en s
1. In oduc ion ............................................................................................................................ 9
1.1. The b ain and i s opog aphic o ganiza ion ............................................................................... 9
1.1.1. The isual sys em and e ino opic o ganiza ion o V1 ............................................................ 9
1.1.2. The audi o y sys em and i s ono opic o ganiza ion .............................................................. 11
1.2. Neu oplas ici y and c oss-modal plas ici y ............................................................................. 13
1.2.1. Dea ness in humans: beha iou al and neu oimaging indings ............................................... 14
1.2.1.1. Dea ness in humans: beha iou al indings ........................................................................ 14
1.2.1.2. Dea ness in humans: ana omical MRI indings ................................................................. 16
1.2.2. Animal models o dea ness ................................................................................................ 18
1.3. Func ional MRI and i s use in neu oscience ........................................................................... 19
1.4. Techniques o measu ing isual ield maps ........................................................................... 20
1.4.1. Phase encoded e ino opy ( a elling wa e e ino opy) ......................................................... 21
1.4.2. Popula ion ecep i e ield (pRF) modelling ......................................................................... 23
1.4.2.1. pRF analysis: ad an ages and applica ions ....................................................................... 24
1.5. Cu en s udy ....................................................................................................................... 25
2. Me hods................................................................................................................................ 26
2.1. Pa icipan s .......................................................................................................................... 26
2.2. S imuli and p ocedu e........................................................................................................... 26
2.3. Ana omical and Func ional Imaging ...................................................................................... 28
2.4. MRI p e-p ocessing ............................................................................................................ 28
2.4.1. Ana omical da a p e-p ocessing .......................................................................................... 28
2.4.2. Func ional da a p e-p ocessing ........................................................................................... 28
2.5. T a elling wa e e ino opy ................................................................................................... 29
2.6. Popula ion ecep i e ield modelling ..................................................................................... 29
3. Resul s .................................................................................................................................. 30
4. Discussion and Conclusion .................................................................................................. 36
Re e ences ................................................................................................................................ 41
Annexes.................................................................................................................................... 47
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 7
Lis o Figu es
Figu e 1 Visual Field Maps in V1, V2 and V3 ...................................................................... 10
Figu e 2 Tono opic Map Layou and In e p e a ions ............................................................. 12
Figu e 3 Visual Field Landma ks .......................................................................................... 21
Figu e 4 T a elling Wa e Re ino opy Maps ......................................................................... 22
Figu e 5 pRF Size Va ia ion Wi h Eccen ici y ..................................................................... 24
Figu e 6 Expe imen al S imuli and P ocedu e ....................................................................... 27
Figu e 7 Re ino opic O ganiza ion o he Visual Co ex in Hea ing and Dea Pa icipan s .. 31
Figu e 8 Explained Va iance in Hea ing and Dea Pa icipan s – Medial View ................... 32
Figu e 9 Explained Va iance in Hea ing and Dea Pa icipan s – La e al View ................... 33
Figu e 10 Be a Values (Slope) in Hea ing and Dea Pa icipan s ........................................... 34
Figu e 11 Pola Angle, Eccen ici y, Ho izon al Loca ion and pRF Wid h in he Dea .......... 35
Figu e A1 TWR Analysis – Phase maps……………………………………………………..40
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 8
Lis o Abb e ia ions
A1
AC
BOLD
CIs
DICOM
DMN
MRI
GLM
GM
HG
HRF
LGN
LH
MRI
NI TI
PAC
pRF
RF
RH
STG
STS
TR
TWR
V1
WM
A ea A1; P ima y Audi o y Co ex
Audi o y Co ex
Blood Oxygen Le el Dependen
Cochlea Implan s
Digi al Imaging and Communica ions in Medicine
De aul Mode Ne wo k
Func ional Magne ic Resonance Imaging
Gene al Linea Model
G ay Ma e
Heschl’s Gy us
Hemodynamic Response Func ion
La e al Genicula e Nuclei
Le Hemisphe e
Magne ic Resonance Imaging
Neu oimaging In o ma ics Technology Ini ia i e
P ima y Audi o y Co ex
Popula ion Recep i e Field
Recep i e Field
Righ Hemisphe e
Supe io Tempo al Gy us
Supe io Tempo al Sulcus
Repe i ion Time
T a elling Wa e Re ino opy
A ea V1; P ima y Visual Co ex
Whi e Ma e
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 15
e y impo an e olu iona ily. Bu how does ha wo k in he dea ? They ely mo e hea ily on
hei emaining senses, since mos o hei inpu om he wo ld comes om he binocula
isual ield (F asnelli e al., 2011). This lack o audi o y- isual con e gence is one o he
easons why Ba elie e al. (2006) p oposed ha hea ing loss leads o changes in highe -le el
a en ional p ocessing, wi h a edis ibu ion o a en ional esou ces o he pe iphe al isual
ield. Mo e ecen s udies like Bo a i e al. (2010) ha e p o ided complemen a y knowledge,
s a ing ha hese di e ences a ise no only om a en ional edis ibu ion, bu also eo ganized
senso y p ocessing.
Di e ences be ween dea and hea ing subjec s a e seen in de ec ing mo ion pe cep ion
changes: dea pa icipan s a e supe io in de ec ing small de ia ions om ho izon al
mo emen s (Almeida e al., 2018; Hau hal e al., 2013). Howe e , heigh ened isual abili ies
as he la e a e no widesp ead in all he a eas o isual cogni ion (F asnelli e al., 2011). In
ac , dea indi iduals can show bo h be e and wo se isual skills han hea ing con ols
(Ba elie e al., 2006). Basic senso y h esholds such as con as sensi i i y (Finney e al.,
2001), mo ion eloci y (B ozinsky & Ba elie , 2004), mo ion sensi i i y (Boswo h &
Dobkins, 1999), b igh ness disc imina ion (B oss, 1979), and empo al esolu ion (Na a e al.,
2008; Poizne & Tallal, 1987) do no seem o be enhanced in dea indi iduals, which is no he
case in mo e complex asks, whe e isual a en ion and p ocessing o he pe iphe al isual ield
a e manipula ed (Almeida e al., 2018). In he absence o audi o y inpu , in o de o moni o
ex ape sonal space, dea indi iduals de o e g ea e p ocessing esou ces o he moni o ing o
he pe iphe al isual ield (Ba elie e al., 2000) and enhanced neu al esponses ha e been
epo ed unde condi ions o pe iphe al compa ed wi h cen al a en ion in congeni ally dea
e sus hea ing con ols (Ne ille & Lawson, 1987). Also, dea signe s appea o be as e a
eo ien ing hei a en ion compa ed wi h hea ing con ols (Pa asnis and Sama , 1985). I has
been sugges ed ha sign language usage (and consequen analysis o hand mo ion) migh al e
mo ion p ocessing (Ba elie e al., 2001). Bo a i e al. (2010) ha e inclusi ely shown ha he
as e eac i i y o isual e en s in he dea happens ega dless o eccen ici y – i.e., bo h
cen ally and pe iphe ally on he isual ield.
In addi ion o hei isual abili ies, dea indi iduals also appea o ou pe o m con ols
in ac ile sensi i i y (Le änen & Hamdo , 2001), p esumably e lec ing neu oplas ici y and/o
a en ion inc ease di ec ed o he s imuli (a mono onous sequence o ib a o y s imuli).
Le änen & Hamdo (2001) es ed o equency disc imina ion and de ec ion o andom
sup a h eshold equency changes, and disco e ed ha congeni al dea ness enhances he
accu acy o sup a h eshold ac ile change de ec ion.

CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 16
O e all, bu pa icula ly wi h ision, he s a ed di e ences could be easily unde s ood
as a compensa ion o he audi o y loss, aking ad an age o he audi o y- isual con e gence
ypically seen in hea ing indi iduals (bu jus using ision). The ield con e ges in a common
inding: di e ences in he dea a e ocused on abili ies ha can ha e a compensa o y ole
ega ding ea ly audi o y dep i a ion, such as pe iphe al isual a en ion (a en ion o mo ion)
and o ien ing mechanisms ha ha e esul ed om he eo ganiza ion o senso y p ocessing.
Dea indi iduals show c oss modal-plas ici y e ec s pa icula ly on he isual unc ions ha in
hea ing indi iduals wo k in andem wi h audi o y inpu . As ision and audi ion a e wo
impo an senses needed o na iga e h ough space and ime, hese compensa o y e ec s can
be use ul in he pe cep ion o daily li e dange and spa ial o ien ing. In sum, hese beha iou al
di e ences a e seen mos ly in he aspec s in which bo h he dep i ed and he o e aking sense
would p o ide use ul inpu oge he (Bell e al., 2019), in o de o p o ide unc ional
ad an ages. To be e unde s and how hese unc ional and beha iou al di e ences can be
ep esen ed s uc u ally in he b ain, he nex sec ion will ocus on ana omical s udies and how
hey co ela e wi h unc ional di e ences.
1.2.1.2. Dea ness in humans: ana omical MRI indings
Mo phome ic s udies o he human b ain ha e a emp ed o in es iga e ana omical
changes be ween hea ing and dea indi iduals. Emmo ey e al. (2003), using MRI and
olume ic analysis, concluded ha dea and hea ing indi iduals do no di e in g ay ma e
(GM) olume o he HG, bu dea indi iduals show signi ican ly la ge g ay ma e -whi e
ma e a ios han hea ing subjec s in he HG, ha ing less whi e ma e amoun in bila e al HG
and o he a eas o he Supe io Tempo al Gy us (STG). Howe e , au ho s such as Tae (2015)
and Ama al e al. (2016) con adic his by showing ha he dea ha e dec eased egional GM
olume in he le an e io HG, bo h in e io colliculi, lingual gy i, nuclei accumbens, and le
pos e io halamic e icula nucleus in he midb ain compa ed o hea ing, leading o he
hypo hesis o an unde de elopmen o he AC in he dea .
Addi ionally, hemisphe ic asymme ies ha e been ound in he subco ical isual and
audi o y b ains o he dea . Ama al e al. (2016) showed ha he igh halamus, he igh LGN,
and he igh in e io colliculus a e la ge han hei le coun e pa s. As he igh AC o he
dea has shown neu oplas ici y by ep esen ing isual in o ma ion (Almeida e al., 2015;
Finney e al., 2001; Nishimu a e al., 1999), hese asymme ies sugges ha he subco ical
a eas could be e ou ing isual in o ma ion o he AC. The e a e also s udies wi h di usion-
weigh ed imaging (Shiell & Za o e, 2017) and measu es o co ical hickness (Shiell e al.,
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 17
2016) ha co ela e he s uc u e o he igh planum empo ale, a ypically audi o y egion,
wi h isual abili y in he dea . Shiell & Za o e (2017) ound changes in ac ional aniso opy,
adial di usi i y, and mean di usi i y in he igh planum empo ale o he dea , sugges ing
al e ed myelina ion densi y o c ossing ib es. This co ela ed wi h he enhanced abili y o he
dea o de ec isual mo ion (explo ed in he nex sec ion), e idence o whi e ma e
eo ganiza ion in a ou o isual enhancemen . The same egion showed inc eased co ical
hickness in he dea (Shiell e al., 2016).
O e all, hese di e ences sugges ha he absence o gene al audi o y pe cep ion leads
o a less de eloped AC, pa icula ly when i comes o g ay ma e in he le hemisphe e (LH).
Howe e , we obse e whi e ma e changes in he igh AC, as well as di e ences in co ical
hickness and co ela ions wi h unc ional esul s, which is consis en wi h seeing mo e
neu oplas ic ep esen a ions in he igh hemisphe e (Almeida e al., 2015).
1.2.1.3. Dea ness in humans: Func ional MRI indings
Mul iple unc ional MRI s udies ha e indica ed ha people wi h senso y dep i a ion
exhibi neu oplas ici y and ec ui hei senso y co ices o p ocess in o ma ion ela ed o hei
emaining senses. The neu onal eo ganiza ion o he b ains om people wi h congeni al
dea ness migh lead o he enhancemen o isual pe cep ion, and is conside ed a compensa o y
mechanism, a emp ing o p o ide some subs i u e o he los modali y (Heimle e al., 2014).
Following he indings o beha iou al s udies, MRI s udies ha e been showing ha he
AC o congeni ally dea indi iduals is ec ui ed o isual asks. Finney e al. (2001) used a
mo ing do pa e n as he isual s imulus and showed ha a eas A1 and B odmann’s a eas 42
and 22 (audi o y associa ion co ex) we e p ocessing ha s imuli in he igh hemisphe e. Bola
e al. (2017) used empo ally complex sequences o s imuli ( hy hms) p esen ed in a isual o
audi o y modali y ( lashes o isual and beeps o audi o y) in dea and hea ing pa icipan s,
whe e bo h g oups pe o med he isual ask and he hea ing g oup also pe o med he audi o y
equi alen ask. Resul s sugges ed ha he isual ask ac i a es he AC (peaking in he pos e io
la e al pa o he high-le el AC) in dea pa icipan s, unlike hea ing subjec s, and ha his
ac i a ion pa e n is simila o he one ha he hea ing g oup expe iences while in he audi o y
modali y o he ask. The au ho s also disclosed inc eased unc ional connec i i y in he dea
be ween he AC and do sal isual co ex (a ea V5/MT, associa ed wi h p ocessing o dynamic
isual s imuli), which was no p esen o he hea ing subjec s. Along he same line o esea ch,
Almeida e al. (2015) implemen ed an MRI expe imen ypically designed o map isual
p e e ences in he isual co ex and showed ha he loca ion o a isual s imulus can be
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 18
decoded om he pa e ns o neu al ac i i y in he AC o dea , bu no hea ing indi iduals,
especially in loca ions wi hin he ho izon al plane and he pe iphe y. S udies wi h isual mo ion
(Re e e al., 2018) ha e ound s onge di ec ion-selec i e esponses in he STS egion in dea
pa icipan s and less signi ican ly in he PAC.
As p e iously men ioned, ision is no he only al e ed sense in he dea popula ion.
Congeni al dea ness also a ec s how he b ain p ocesses soma osensa ion. Fo ins ance, he
dea p esen g ea e signal change in he os ola e al HG han hea ing indi iduals while
p ocessing soma osenso y and bimodal s imuli (double- lash isual illusion induced by wo
ouches o he ace) (Ka ns e al., 2012).
Neu oplas ici y migh lead o neu al eo ganiza ion, bu i does no necessa ily lead o
losing i s o iginal o ganizing p inciples. Li e a u e has shown ha despi e c oss-modal
plas ici y and senso y dep i a ion, he AC o he dea s ill shows opog aphic o ganiza ion:
S iem-Ami e al. (2016) used MRI o look a unc ional connec i i y pa e ns in he dea and
iden i ied opog aphic ono opy-based unc ional connec i i y s uc u es in he AC and
ex ending ono opic g adien s (in audi o y co e and bel , pa abel , ex ending o language and
speech/ oice sensi i e egions). This sugges s ha opog aphical o ganiza ion does no need
senso y expe ience o de elop and is no a ec ed by b ain plas ici y.
O e all, MRI s udies ha e shown ha p ope ies such as isual mo ion (Re e e al.,
2018), posi ion in he isual ield (Almeida e al., 2015; Finney e al., 2001) and
hy hm/ equency disc imina ion (Bola e al., 2017) a e ep esen ed in ea ly and/o associa i e
AC o dea humans due o mechanisms o c oss-modal plas ici y.
1.2.2. Animal models o dea ness
Animal models enable us o examine beha iou wi hou a iables o es ains ha a e
a ibu ed o humans (e.g. language). In humans, c oss-modal and neu oplas ic e ec s obse ed
in congeni ally dea humans may esul om senso y dep i a ion o om he al e ed linguis ic
expe ience, such as sign language (Ba elie & Ne ille, 2002). The use o animal models - e.g.,
ca s udies - show us ha co ical eo ganiza ion happens independen ly o language
acquisi ion.
Congeni ally dea ca s ha e been conside ed a sui able model o human congeni al
dea ness (Heid e al., 1998). They show supe io pe o mance in isual localiza ion in
pe iphe al isual ields and lowe isual mo emen de ec ion h esholds, compa ed o hea ing
ca s (Lombe e al., 2010). Lombe and colleagues (2010) used psychophysical asks and
g aded cooling, con i ming a c oss-modal eo ganiza ion o he dea AC. They we e able o
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 19
localize indi idual isual p ocesses in po ions o he AC, and pa icula ly demons a e he ole
o he supe icial laye s o he do sal zone o he AC in enhanced isual mo ion de ec ion in
dea ca s. Mo eo e , in ca s, as in humans, c oss-modal plas ici y is no es ic ed o ision.
Me edi h & Lombe (2011) showed ha soma osenso y and isual modali ies pa icipa e in
c oss-modal einne a ion ou side A1, in he an e io audi o y ield o ea ly-dea ca s. S udies
in mice using elec ophysiological eco ding echniques combined wi h co ical
myeloa chi ec u e also showed ha he AC o dea mice con ained neu ons ha esponded o
soma osenso y and isual in o ma ion, and also ha hei p ima y isual a ea had an inc ease
in size (Hun e al., 2006).
In sum, animal s udies co obo a e and sha e he majo indings seen in dea humans:
supe io i y in localizing s imuli in he pe iphe al isual ield, supe io i y in mo ion de ec ion,
and neu al di e ences in he p ocessing o bo h isual and soma osenso y in o ma ion
( ec ui men o he AC). These indings p o ide a backg ound o ou expe imen : he
beha iou al and unc ional da a clea ly show ha he AC o he dea is ec ui ed o isual
asks. Mo e speci ically, i p ocesses spa ial in o ma ion ega ding he loca ion o isual
s imuli. He e, we aim o u he explo e his c oss-modal p ocessing, by in es iga ing i he AC
o dea indi iduals ep esen s low-le el spa ial ea u es o isual s imuli, and i hese
ep esen a ions a e o ganized in a e ino opic o ma . We will use MRI as ou neu oimaging
me hod and pRF analysis as ou modelling echnique. These me hods will be desc ibed below.
1.3. Func ional MRI and i s use in neu oscience
Magne ic esonance imaging (MRI) is a echnique ha uses a s ong magne ic ield o
c ea e images o biological issue - in ou case, he b ain. The MR scanne uses a se ies o
changing magne ic g adien s and oscilla ing elec omagne ic ields, known as he pulse
sequence, causing ene gy o be abso bed and emi ed by he a om’s nuclei. Adjus men s in
pa ame e s o he adio equency exci a ion pulses and he magne ic ield g adien s make i
possible o acqui e in o ma ion abou s uc u e (ana omical imaging), low (pe usion
imaging), o neu al ac i i y ( unc ional imaging) (Logo he is & Wandell, 2004).
In he case o unc ional MRI ( MRI), i allows o he iden i ica ion o whe e in he
b ain pa icula men al p ocesses occu , and o cha ac e ize hose pa e ns o b ain ac i a ion.
This is possible by measu ing changes in blood oxygena ion o e ime, as i s le els change
apidly ollowing he ac i i y o neu ons in speci ic b ain egions (Hue el e al., 2008). Neu al
ac i i y leads o he consump ion o oxygen, ATP, and glucose, ollowed by an inc ease in
ce eb al blood low. The e is a misma ch be ween ce eb al blood low and oxygen consump ion
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 20
(leading o an o e supply o oxygen), which is one o he bases o he blood-oxygen-le el-
dependen (BOLD) signal. This BOLD signal is sensi i e o al e a ions in deoxygena ed
haemoglobin le els and ce eb al blood olume, and elies on he phenomenon o
neu o ascula coupling: he link be ween changes in neu onal ac i i y and he cons ic ion o
dila ion o mic o essels in he b ain ( o mo e on how o in e p e he BOLD signal o make
in e ences abou he neu al signal see Logo he is & Wandell, 2004).
Besides MRI, se e al o he echniques a e used in neu oimaging, such as
elec oencephalog aphy (EEG), single-uni eco ding, posi on emission omog aphy (PET),
and magne oencephalog aphy (MEG). All o hese di e in hei spa ial and empo al
esolu ion, how hey acqui e in o ma ion, whe he hei measu emen s a e di ec o indi ec ,
and in asi eness. MRI is a non-in asi e echnique ha has a high spa ial esolu ion, and
because o ha i is inc easingly used in basic and applied neu oscience.
The MRI BOLD signal ha is measu ed om each oxel ( he basic unc ional uni ,
usually a 3x3x3 millime e olume ic uni ) ep esen s he a e age ac i i y o all he neu ons
wi hin ha oxel. Based on he opog aphic o ganiza ion o di e en co ical and subco ical
a eas, neighbou ing cells a e cha ac e ized by simila ecep i e ields p ope ies, and hus, hei
a e aged esponse o s imula ion o hei ecep i e ield will esul in an inc ease o he BOLD
signal. This, as well as he abili y o MRI o ob ain signal om he whole b ain, make MRI
imaging a use ul and e icien me hod o map he opog aphic o ganiza ion o he b ain. Today
i is one o he mos impo an me hods o measu e spa ial o ganiza ion, which is wha we aim
o do. Engel e al. (1994, 1997) used MRI 27 yea s ago o measu e he e ino opic o ganiza ion
wi hin he human isual co ex, using isual s imuli o map eccen ici y and pola angle
p e e ences: speci ically, hey used expanding ings and o a ing wedges espec i ely. These
s imuli we e used o c ea e a elling wa es o neu al ac i i y in he e ino opically o ganized
co ex (see s imuli in Figu e 4).
1.4. Techniques o measu ing isual ield maps
De ailed measu emen s o isual ield maps in indi idual subjec s wi h MRI ha e been
pe o med wi h se e al echniques, including he a elling wa e e ino opy (TWR) echnique
and mo e ecen ly popula ion ecep i e ield (pRF) modelling. TWR - o phase-encoded
e ino opy - has been he gold s anda d pa adigm o he las wo decades. pRF is an inno a i e
app oach ha allows o mo e accu a e and comple e mapping, and sol es some conce ns
p esen ed by con en ional mapping me hods.

CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 21
1.4.1. Phase encoded e ino opy ( a elling wa e e ino opy)
TWR is a me hod used in MRI expe imen s o measu e e ino opic o ganiza ion in he
human co ex and is based on a a eling wa e o neu al ac i i y wi hin e ino opically
o ganized isual a eas, c ea ed by he p esen a ion o a isual s imulus. Because o his
e ino opic o ganiza ion, he p esen a ion o s imuli as he o a ing wedges and expanding ings
(Engel e al., 1994) can elici a con inuous a eling wa e o neu al ac i i y in he isual co ex
(Engel e al., 1997).
The con inuous and pe iodic a ia ion o he isual s imuli loca ion gene a es MRI
esponses ha a y sys ema ically in hei delay, which is measu ed as he phase o he sinusoid
ha bes i s he da a. While measu ing eccen ici y p e e ences ( ing expanding om he o ea
o pe iphe y), we see ha oxels ha espond la e in ime ep esen mo e pe iphe al pa s o
he isual ield, p oducing a a eling wa e o ac i i y mo ing om he pos e io o he an e io
pa o he calca ine sulcus. This means ha he isual ield loca ion ep esen a ions a each
oxel a e es ima ed om he ela i e iming o ha oxel's MRI esponse. This iming is
quan i ied as he phase o he MRI signal, calcula ed by aking he Fou ie ans o m o he
oxel's ime se ies (Engel e al., 1994; Engel, 2012).
The wo o hogonal dimensions needed o iden i y a unique loca ion in he isual space
(x, y) a e pola angle (s imulus is o a ing wedge) and eccen ici y (s imulus is expanding ing),
ep esen ed by wo pola coo dina es: an angle ( he a, di ec ion om he cen e o he poin ),
and a adius (dis ance om he cen e o he poin ). Taken oge he , hese wo measu emen s
speci y he mos e ec i e isual ield posi ion (Engel e al., 1997; Engel, 2012; Se eno e al.,
1995; Wandell e al., 2007) (see Figu e 3).
Figu e 3
Visual Field Landma ks
No e. Visual ield landma ks and ep esen a ion o eccen ici y ( adius) and pola angle ( he a)
meaning. F – ixa ion poin ; HM – ho izon al me idian; UVM – uppe e ical me idian; LVM
– lowe e ical me idian. F om “Re ino opic O ganiza ion in Human Visual Co ex and he
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 22
Spa ial P ecision o Func ional MRI”, by S. Engel, G. Glo e and B. Wandell, 1997, Ce eb al
Co ex, 7(2):181-92. Adap ed wi h pe mission om Ox o d Uni e si y P ess.
These s imuli we e designed o maximally s imula e he p ima y isual co ex and elici
an MRI signal modula ion on he o de o 1%-3%, o 15-20 s anda d de ia ions abo e
backg ound noise (B ewe & Ba on, 2012). An example o e ino opic maps ob ained by he
TWR me hod is shown in Figu e 4, whe e we see a selec ion o oxels ha ha e a powe ul
esponse abo e a de ined h eshold o cohe ence, ha is he co ela ion be ween he esponse
o ha oxel and he sinusoidal model, a he equency o s imulus p esen a ion (B ewe &
Ba on, 2012).
Figu e 4
T a elling Wa e Re ino opy Maps – Medial iew o he Visual Co ex
No e. Colo map shows he esponse phase a each loca ion o pola angle (A) and eccen ici y
expe imen s (B). The solid whi e lines indica e he bounda ies be ween isual a eas. Only
oxels wi h a powe ul esponse a a cohe ence ≥ 0.25 a e colou ed. Rega ding pola angle
p e e ences, we can see ha he lowe e ical me idian is ep esen ed along he uppe bank o
he sulcus (A). We also see ha o eal p e e ences all along he pos e io end o he sulcus,
and pe iphe y p e e ences a e ep esen ed in mo e an e io loca ions in he calca ine (B). PO-
S, pa ie al-occipi al sulcus; Cal-S, calca ine sulcus. F om “Visual Field Map O ganiza ion in
Human Visual Co ex”, by A. B ewe and B. Ba on, in Visual Co ex- Cu en S a us and
Pe spec i es, 2012, In echOpen. Rep in ed wi h pe mission.
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 23
These phase-encoded designs using he TWM a e in o ma i e abou neu al uning
p ope ies and allow o an es ima ion o he en i e isual ield layou , measu ing co ical ield
maps. The applica ion o phase-encoded me hods was pa icula ly e ec i e in de ining ield
maps in ea ly isual a eas, which a e cha ac e ized by neu ons wi h small ecep i e ields ha
a e mos ly con ined o one hemi ield. Howe e , his me hod is less e icien o mapping a eas
wi h la ge ecep i e ields ha include he o ea (B ewe & Ba on, 2012; Wandell e al., 2007).
As we a e no only analysing ea ly isual a eas bu also he AC ( ha migh ha e se e al
ecep i e ield sizes and isual p e e ences), pRF analysis seemed o be a mo e ad an ageous
me hod. Below is a desc ip ion o his echnique and why i is p e e able in ou s udy.
1.4.2. Popula ion ecep i e ield (pRF) modelling
pRF modelling is a compu a ional imaging app oach ha builds models p edic ing he
neu al esponses o a s imulus, and can be used in a wide ange o condi ions (in ou case,
econs uc ing co ical isual ield maps). I couples MRI signals (a millime e scale) wi h
ecep i e ield p ope ies o isual neu ons (a mic on scale). pRF es ima es no only he isual
ield map, bu p ope ies such as ecep i e ield size, la e ali y, and su ound supp ession
(Dumoulin & Wandell, 2008), and is es ima ed o indi idual subjec s. These RF p ope ies
change sys ema ically ac oss eccen ici y and be ween isual ield maps (F acasso e al., 2016).
The key pRF pa ame e s a e ecep i e ield posi ion and ecep i e ield size, uni s ha
can be compa ed o e di e en ins umen s and di e en subjec s. The shape o he ecep i e
ield is a wo-dimensional ci cula ly symme ic (iso opic) Gaussian in he isual ield,
desc ibed by ield posi ion (x,y) and sp ead (s), bo h in isual deg ees. Fo each oxel, he pRF
pa ame e s a e adjus ed o ma ch p edic ed and measu ed MRI ime se ies (Wandell &
Winawe , 2015).
pRF analysis can be used in expe imen s whe e he adi ional mapping s imuli a e
used, bu also wi h a se ies o ba pa e ns ha sweep h ough he isual ield in di e en
di ec ions, o a se ies o s imuli placed a di e en isual ield posi ions. Figu e 5 demons a es
ha pRF size inc eases along he isual hie a chy ( om V1 o V3) and, simul aneously, wi hin
a gi en isual a ea he e is a posi i e co ela ion be ween pRF eccen ici y and size.
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 24
Figu e 5
pRF Size Va ia ion Wi h Eccen ici y
No e. pRF size inc eases wi h eccen ici y and along he maps in isual a eas, wi h smalle pRF
sizes in V1 and la ge in en al and la e al isual a eas (A). The inc easing adius o each
ci cle, ep esen ing pRF size, in each eccen ici y posi ion and isual a ea (B). F om
“Compu a ional neu oimaging and popula ion ecep i e ields”, by B. Wandell and J. Winawe ,
2015, T ends in Cogni i e Sciences, 19 (6), 349-357. Rep in ed wi h pe mission om Else ie .
1.4.2.1. pRF analysis: ad an ages and applica ions
Al hough he pRF me hod was de eloped o e ino opic mapping in he isual co ex
(Dumoulin & Wandell, 2008), i has had o he applica ions such as measu ing ono opic maps
and es ima ing bandwid h o oxels in he human AC (Thomas e al., 2015). Applica ions on
he s udy o isual co ical esponses using pRF analysis include esea ch on clinical condi ions
and cogni i e ask demands (Wandell & Winawe , 2015). Examples a e s udies on a en ion
(Sp ague & Se ences, 2013), plas ici y o he adul isual co ex (Papanikolaou e al., 2014),
de elopmen al plas ici y (Haak e al., 2014), psychia ic and neu ological diso de s such as
demen ia (B ewe & Ba on, 2012) and au ism (Hadjikhani e al., 2004; Schwa zkop e al.,
2014).
The ad an ages o pRF o e TWR include he ac ha pRF analysis is mo e p ecise in
he case o isual ield maps wi h la ge ecep i e ields. Mo eo e , pRF analysis p o ides no
only he p e e ed cen e o each oxel’s pRF bu also pRF size/sp ead. Fu he mo e, i also
p o ides in o ma ion on la e ali y. These a e he easons why pRF was chosen o his s udy,
as i e eals opog aphical o ganiza ion mo e clea ly han con en ional me hods and gi es us
access o addi ional in o ma ion (B ewe & Ba on, 2012).
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 31
Figu e 7
Re ino opic O ganiza ion in he Visual Co ex o Hea ing and Dea Pa icipan s
No e. Eccen ici y and pola angle ep esen a ions along he ea ly isual co ex. A. eccen ici y.
The colou blue ep esen s he cen e o he isual ield (0 isual deg ees), up o ed (pe iphe y,
6 o mo e isual deg ees) B. pola angle ( anging om -π ( ed) o π). RH- igh hemisphe e.
LH- le hemisphe e. Th eshold is explained a iance ≥ 0.20. Dashed lines delinea e he
calca ine sulcus and possible bo de s be ween V1 and V2. pRF scaled model including only
posi i e be a alues.

CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 32
In his analysis, we expec ed o see s ong isual p e e ences and o ganiza ion in ea ly
isual a eas. We clea ly see eccen ici y p e e ences anging om he o ea o he pe iphe y
along he axis o he calca ine sulcus in he ea ly isual co ex. Rega ding pola angle, we see
a clea con ala e al ep esen a ion o he isual ield (RH ep esen s le isual ield), and he
expec ed mapping o he s imulus ega ding he y/ e ical axis. In ac , pola angle is one o
he clea e measu es o de ine he bo de s be ween V1, V2, and V3. This is pa icula ly
disce nible o he pa icipan s dea 01 and hea ing.
Nex , we u ned o explo e he nega i e pRF model, by compa ing he ex ac ed
pa ame e s be ween he nega i e and posi i e models. Fi s , we looked a he a iance
explained by each one o he models, ocusing on ea ly isual a eas (see Figu e 8). The maps
show ha bo h models pe o m simila ly and p o ide good i ing o he da a in he ea ly isual
co ex. The nega i e pRF model co e s addi ional a eas compa ed o he posi i e model,
mainly in medial pa ie al a eas (see Szin e and Knapen (2020) o a simila esul ). Figu e 9
shows he same a iance explained maps in a la e al iew including he AC.
Figu e 8
Va iance Explained in Hea ing and Dea Pa icipan s – Medial View o he Visual Co ex
No e. Va iance explained in a medial iew o ea ly isual a eas o he 3 subjec s. Th eshold is
2 ≥ 0.08. pRF scaled model including posi i e and nega i e be a alues (A) and only posi i e
be a alues (B). RH – igh hemisphe e; LH- le hemisphe e.
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 33
Figu e 9
Va iance Explained in Hea ing and Dea Pa icipan s – La e al iew o he B ain
No e. Va iance explained in a la e al iew o he b ain, including he AC. Th eshold is 2 ≥
0.08. pRF scaled model including posi i e and nega i e be a alues (A) and only posi i e be a
alues (B). The hea ing subjec does no show di e ences be ween models, bu he dea
subjec s show mo e clus e s o signi ican explained a iance in he model ha includes
nega i e be a alues (A). RH – igh hemisphe e; LH- le hemisphe e.
O e all, we see ha he hea ing subjec does no appea o show clea di e ences
be ween he wo models, bu he dea show some mo e a iance explained in he model ha
includes nega i e be a alues. Thus, i seems ha po en ial e ino opic esponses in he AC a e
be e cap u ed by he nega i e pRF model, which led us o ocus on he model ha includes
bo h posi i e and nega i e pRFs in he ollowing analysis s eps.
To u he cha ac e ize he esponses o he isual s imuli we plo ed he be a alues o
he bes i ed p edic o a each co ical loca ion, colou coded by i s sign (Figu e 10A and
10B). This igu e shows ha in ea ly isual a eas he be a alues a e p edomina ely posi i e
ac oss he pa icipan s, wi h nega i e alues a he medial pa ie al co ex (in a eas o he DMN,
simila o Szin e and Knapen, 2020). On a la e al iew (Figu e 10A), dea (bu no he hea ing)
indi iduals show mo e ep esen a ions ela ed o nega i e be as alues, mainly in he p e on al
and empo al lobes. The di e ence be ween he hea ing and dea pa icipan s is mo e e iden
when plo ing he nega i e and posi i e alues in sepa a e maps (Figu e 10C and 10D). This is
pa icula ly obse able in dea 01, which shows pe inen clus e s in he STS, STG, and HG,
bu also in dea 02.
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 34
Figu e 10
Be a Values (Slope) in Dea and Hea ing Pa icipan s
No e. Slope (be a alues om ≤ -4 o ≥4, wa m colou s code posi i e alues and cold colou s
code nega i e alues), in he la e al (A) and medial (B) iew in he model wi h posi i e and
nega i e be a alues, and la e al iews o he masked model wi h jus posi i e be a alues (C)
and wi h jus nega i e be a alues (D). Th eshold is 2 ≥ 0.08.
The indings o Figu e 10 lead us o explo e which isual ea u es a e ep esen ed in
he audi o y a eas delinea ed by he nega i e model. Since isual esponses we e seen only in
he AC o he dea and only o nega i e pRF uning, we ocused on hese subjec s in he las
analysis, aiming o explo e which isual ea u es a e ep esen ed in he AC o he dea . Fo
ha , we ex ac ed he pa ame e s ep esen ing pola angle, eccen ici y, x loca ion, and pRF
size (see Figu e 11) only in oxels wi h nega i e be a alues, as i is he p edominan
ep esen a ion in he dea (bu no in he hea ing).
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 35
Figu e 11
Pola Angle, Eccen ici y, Ho izon al Loca ion and pRF Wid h in he Dea
No e. Visual ep esen a ions in he AC o he dea pa icipan s. These pa ame e s we e es ed
wi h he masked model wi h jus nega i e be a alues. A. Pola angle anges om -π ( ed) o
π. B. In eccen ici y p e e ences, he colou blue ep esen s he cen e o he isual ield (0
isual deg ees), ex ending un il ed (pe iphe y, 6 o mo e isual deg ees). C. The ho izon al
isual loca ion was measu ed on a scale o ≤ -6 o ≥ 6. D. pRF size alues a e be ween 0 and ≥
5. Th eshold is 2 ≥0.08. Do ed lines delinea e he STS and HG.
We can d aw se e al conclusions ega ding hese maps: o e all, explained a iance
alues a e no high in he AC, and al hough he RH shows s onge esponses han he le ,
he e a e no clea e ino opic maps. Then, in pola angle, we see ep esen a ions o he le
isual ield in he RH, which is con ala e al and in acco dance wi h wha happens in he isual
co ex. The LH does no show clus e s la ge enough o d aw any conclusion. Rega ding
eccen ici y, he maps sugges ha he p e e ed a ea ep esen ed in he AC is mos ly cen e o
he isual ield, which is no wha we expec ed gi en he end owa ds pe iphe y
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 36
ep esen a ions seen in he li e a u e. X posi ion shows mo e clea ly wha we had al eady seen
in pola angle: he e is a no able p e e ence o ep esen ing he le isual ield, pa icula ly in
he RH. Wid h does no seem o show a consis en g adien , bu i shows a endency o high
alues, meaning a la ge ecep i e ield size. I also does no seem o be posi i ely co ela ed
wi h eccen ici y p e e ences.
These clus e s in he dea appea in he RH along he STS, STG, he in e io ci cula
sulcus o he insula and he e is also a clus e in he T ans e se Tempo al Gy us (TTG), which
co esponds o he Heschl's gy us. This means ha hey a e seen bo h in ea ly and associa i e
audi o y a eas, bu mo e signi ican ly in he associa i e.
4. Discussion and Conclusion
This hesis aimed o in es iga e c oss-modal plas ici y in congeni ally dea pa icipan s:
i s ly, by eplica ing p e ious esul s indica ing ha he e a e ep esen a ions o isual
in o ma ion in he AC o he dea bu no he hea ing; secondly, by a emp ing o map hese
ep esen a ions, and mo e speci ically, unco e e ino opic ea u es o he AC (i any) ha a e
simila o he isual co ex. This se o single cases included an MRI expe imen wi h s imuli
ypically used in classic e ino opic expe imen s designed o map he isual co ex. We
analysed he da a wi h a adi ional a elling-wa e me hod, and hen ocused on ou elec ed
echnique: he pRF analysis. The esul s ha e demons a ed ha he e a e isual- ela ed
esponses in he AC o he dea ollowing a cyclic isual p esen a ion, bu no in he hea ing.
Conce ning he e ino opy ea u es, se e al aspec s a e o ele ance: he e a e no ably
mo e ep esen a ions in he RH/ igh AC o he dea han in he le , and hese clus e s ep esen
in o ma ion om he con ala e al isual ield, he e o e co esponding o he le isual ield
and in acco dance wi h co ical isual p ocessing. This ana omical la e ali y (p edominancy o
he RH) in he dea is cohe en wi h li e a u e on hemisphe ic asymme ies, whe e ana omical
s udies ha e shown inc eased whi e ma e olume in he RH compa ed o he LH (Ama al e
al., 2016; Tae, 2015). I also suppo s MRI s udies s a ing he sup emacy o he igh AC o
he dea in ep esen ing isual in o ma ion (Almeida e al., 2015; Finney e al., 2001), inding
also p esen in connec i i y and co ical hickness s udies: Bola e al. (2017) e ealed unc ional
coupling be ween he AC and a ea V5 in he dea (an a ea esponsible o he p ocessing o
dynamic isual s imuli), and Shiell e al. (2016) ha e sugges ed a beha iou -s uc u e
co ela ion in dea pa icipan s whe e he ones wi h be e pe o mances a isual mo ion
de ec ion ha e inc eased co ical hickness in he igh planum empo ale.

CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 37
Rega ding he maps ep esen ing pola angle and eccen ici y, al hough we do no see
clea e ino opic g adien s ully co e ing he isual ield, he esponses in he AC sha e some
e ino opic ea u es: pRFs in he AC ep esen in o ma ion om la ge pa s o he con ala e al
isual ield, and he pRF cen es a e mos ly localized o he o eal o nea cen al pa o he
isual ield. Func ional la e ali y is ein o ced by he map o ho izon al posi ion, suppo ing
his con ala e al ep esen a ion o he isual ield in he RH. pRF size (wid h) is o e all la ge
in he audi o y a eas bu he e is no g adien o posi i e co ela ion wi h eccen ici y, as i
happens wi h RFs in he isual sys em (Dumoulin & Wandell, 2008). In ac , i he e is a
co ela ion, i ends owa ds a nega i e one: we see la ge wid h pRFs wi h cen es localized in
he o ea, meaning ha hey ep esen la ge pa s o he isual ield.
Al oge he , he alues o explained a iance a e low in he AC o he dea , compa ed
o he alues in he isual sys em, and dea 01 clea ly shows mo e clus e s in he maps han
dea 02. Howe e , he explained a iance alues a e s ill highe han he alues ound in he
hea ing pa icipan , and highe han he a e age alues ob ained a non- isual a eas (whi e
ma e ). Mo eo e , hese clus e s appea in ana omical audi o y egions ha ha e p e iously
been associa ed wi h neu oplas ici y ollowing senso ial dep i a ion: he HG and he STS
(Almeida e al., 2015; Daphne Ba elie e al., 2001; Finney e al., 2001; Re e e al., 2018). To
gi e an example, Re e e al. (2018) used isual mo ion s imuli ( andom-do kinema og ams)
in an MRI expe imen and epo ed ha he PAC o dea (and no he hea ing) shows di ec ion-
selec i e isual mo ion esponses, and ha he igh STS is ex ensi ely ec ui ed o ha
p ocessing. Ba elie e al. (2001) co obo a e his ec ui men o he pos e io STS in a simila
ask. These a eas belong o he ea ly and associa i e AC: while he HG is he i s co ical
s uc u e o p ocess audi o y in o ma ion and is pa o he PAC, he STS has been conside ed
he main egion o audio- isual in eg a ion, al hough i has also been associa ed wi h
biological mo ion pe cep ion, speech p ocessing, p ocessing o aces and heo y o mind (Hein
& Knigh , 2008).
A c i ical inding in ou esul s is ha he isual- ela ed esponses in he dea we e
e ealed h ough he applica ion o he “nega i e” pRF model - i.e., wi h he nega i e be a
alues. We can discuss he meaning o hese nega i e alues, ha a e associa ed wi h nega i e
BOLD signal and he e o e would po en ially ep esen neu al deac i a ion o supp ession.
Nega i e c oss-modal modula ion has been epo ed p e iously in he li e a u e wi h hea ing
popula ion, i.e., isual ac i i y modula ing audi o y ac i i y in he o m o deac i a ion in he
AC (Lau ien i e al., 2002; Mozolic e al., 2008). Al hough some a eas o he co ex (poly-
senso y a eas) can be in ol ed in a a ie y o highe -o de mul isenso y pe cep ions, senso y-
speci ic co ices (such as isual and audi o y) do no unc ion as independen ly as i was
conside ed in he pas . Lau ien i e al. (2002) used MRI in a passi e s imula ion pa adigm
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 38
wi h 3 condi ions (a isual s imulus alone, an audi o y s imulus alone, and a combined isual-
audi o y s imulus) and concluded ha isual s imula ion esul ed in nega i e BOLD signals in
he AC, namely along he supe io and middle empo al gy i wi h a peak in B odmann a ea 22
(loca ed in he PAC). A simila e ec occu ed wi h he audi o y condi ion, as nega i e BOLD
was obse ed in oxels o ex as ia e isual a eas, bu no o he combined s imulus condi ion.
Addi ionally, i has been p oposed ha he di icul y o he ask de e mines he deg ee o
deac i a ion (Hai s on e al., 2008). This is also ue o in a-modal deac i a ions (Wilson e
al., 2019), whe e nega i e BOLD esponses a e elici ed wi hin he s imula ed co ex, as
unila e al s imulus can deac i a e he ipsila e al senso y co ex (e.g., unila e al hand
mo emen s ac i a e he con ala e al senso imo o co ex and deac i a e he ipsila e al co ex
(Allison e al., 2000); ea ly isual a eas ha a e uned o o eal ep esen a ion espond wi h
posi i e BOLD signal o o eal isual s imuli and wi h nega i e BOLD signal when he s imuli
a e p esen ed in he pe iphe y, and ice e sa (Shmuel e al., 2006).
This ask-speci ic deac i a ion howe e is con o e sial and is no epo ed in all b ain
imaging s udies using single senso y s imuli (Lau ien i e al., 2002). This can be due o di e en
expe imen al condi ions in dis inc con ex s and wi h di e en asks, and because mos s udies
ocus on b ain ac i a ion and posi i e modula ion esul s ins ead o nega i e BOLD signal and
deac i a ion. We see his con as in ou da ase , among he dea and hea ing pa icipan s.
Gi en ha all pa icipan s we e p esen ed wi h he same s imuli in an equal con ex , di e ences
be ween dea -hea ing could a ise due o hei own unc ional and ana omical b ain p ope ies.
I would be e oneous o gene alize his hypo hesis since a g oup analysis has no been
pe o med (which would equi e mo e pa icipan s), bu i is an assump ion o u he explo e.
Besides, i deac i a ion/supp essing non ele an in o ma ion con ibu es o he maximal
pe cep ual in eg i y and a en ion o senso y in o ma ion (Hai s on e al., 2008), one could
a gue ha his deac i a ion is in ol ed in he supe io pe o mance o he dea in some isual
asks, as p e iously men ioned. They would di ec mo e a en ion o he ask han he hea ing
pa icipan s and he e o e exhibi supe io abili ies. As discussed in he in oduc ion o his
hesis, he answe is pe haps ound in he middle: enhancemen s o he congeni ally dea in
non-dep i ed senses esul om a en ion eo ien ing and neu onal eo ganiza ion. Also, we
could hypo hesize ha gi en ha he pRF cen es a e loca ed nea he o ea in he AC o he
dea (and he peak o deac i a ion happens wi h o eal ep esen a ions), his deac i a ion migh
sha pen hei abili y o dis inguish be ween cen al and pe iphe al isual ep esen a ions.
Ne e heless, in a and c oss-modal deac i a ions be ween senso y co ices a e no he
only occasions whe e nega i e BOLD is p esen . Va ious s udies ega ding he DMN men ion
deac i a ion, nega i e BOLD signal, and inclusi ely use pRF as an analysis echnique. As a
ma e o ac , Szin e & Knapen (2020) coined he e m “nega i e pRFs” o e e o signals
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 39
be e p edic ed by deac i a ion. They used MRI and pRF analysis o p o e ha he DMN
selec i i y deac i a es as a unc ion o he posi ion o a isual s imulus, and ha i ac s as a
nega i ely modula ed high le el isual ne wo k. This sha es some simila i y wi h ou esul s
(bo h in dea and hea ing bu mos ly in dea ) in medial pa ie al a eas, belonging o he DMN
(see igu e 10B) - ha englobes egions in he la e al p e on al, pos e omedial, in e io pa ie al
co ices, as well as egions in he la e al and medial empo al co ex. They also s a ed, as we
did, ha he signal esul ing om isual a eas (bo h high and low le el) is bes explained wi h
posi i e pRF models, so models p edic ing a posi i e ampli ude modula ion (posi i e slope).
Fu he mo e, hey also pe cei e ha hese oxels p e e ably ep esen he con ala e al isual
ield (as he AC o ou dea subjec s, RH) and hei RF sizes inc eased wi h eccen ici y
( e ino opy ea u e ha we could no obse e). DMN a eas ypically show ac i a ion du ing
es (Raichle 2001) and sel - ela ed high-le el cogni i e asks such as episodic and seman ic
memo y o mind wande ing and deac i a ion du ing a en ion-demanding and ex e nally
o ien ed asks (Al es e al., 2019). Fo ins ance, mo o mo emen s such as hand and oo
mo emen s deac i a e he DMN (Naka a e al., 2019).
Ano he aspec we could e lec on is he ma e o con olling a en ion main enance,
by u ning an appa en ly passi e ask in o an ac i e one. In his expe imen , he pa icipan s
iewed he s imuli on he sc een and solely had o ixa e on he cen al ixa ion poin . O he
iden ical expe imen s ha e used beha iou al asks such as clicking on a esponse bu on o
epo mo ion di ec ion changes (see Re e e al., 2018), o moni o he s imuli o epo he
changes in he end o he block (see Ba elie e al., 2001) o e en click on a bu on when he
colou o he ixa ion do changed (see Szin e & Knapen, 2020). I emains undisclosed whe he
adding such asks ha modula e spa ial a en ion would inc ease a en ion ec ui men o
a en ion main enance h oughou he uns o ou da ase , and hei e ec s in he ampli ude o
he BOLD esul s, ei he posi i e o nega i e in ega d o he baseline. We could hypo hesise
ha a la ge a en ion ec ui men would p o ide clea e esul s.
Va ious s eps can be pe o med o u he explo e how he isual co ex o he dea use
addi ional esou ces om he b ain, namely he AC. In u u e esea ch, collec ing a la ge
sample migh s eng hen he esul s and po en ia e a g oup analysis, con as ing he hea ing and
he dea , and di e en pa ame e s. E en i he pRF model was applied a an indi idual le el,
mo e pa icipan s would pe mi compa isons and gene aliza ions. This would be pa icula ly
in o ma i e gi en ha dea 01 and dea 02 do no beha e exac ly he same, and i would allow
o po en ially ein o ce he esul s o dea 01. Then, a Region o In e es (ROI) analysis wi h
selec ed a eas in he isual and audi o y co ices would p o ide a mo e de ailed s a is ical
analysis o he da a. Addi ional expe imen s could include epea ing his s udy wi h o he
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 40
senso y modali ies, such as sea ching o nega i e BOLD in he isual co ex o he blind and
looking o ono opic o ganiza ion in his ec ui men o deac i a ion.
Resea ch on neu oplas ici y is ele an o neu oscience and neu opsychology bo h o
i s con ibu ion o basic science, i.e., o he s a e o he a on he b ain’s opog aphical
o ganiza ion and eo ganiza ion, bu also o applied neu o ehabili a ion pu poses: de ices
such as cochlea implan s (CIs) can be de eloped o compensa e o he senso y loss in he dea
popula ion. These implan s con e audi o y signals in o elec ic impulses deli e ed o he
acous ic ne e, eplacing no mal cochlea unc ions, and equi e a deep comp ehension o he
dep i ed b ain’s abili y o e e se changes. Di e en a eas and unc ions ha e hei own
windows o sensi i i y and suscep ibili y o plas ici y. Some mechanisms may be a ailable
h oughou li e (co ico-co ical plas ici y), bu o he s a e ha dly main ained in adul s
(subco ical plas ici y) (Ba elie & Ne ille, 2002). This aises he ques ion o how c oss-modal
plas ici y will in e ac wi h he p esence o audi o y inpu . I he AC has eo ganized o p ocess
ision o ouch, will neu oplas ici y be de imen al o CIs? Human and animal da a sugges
ha i in e e es wi h he ese lemen o he egained audi o y inpu s (K al e al., 2006; K al &
Sha ma, 2012; Sha ma e al., 2014). Thus, i can be hypo hesised ha c oss-modal plas ici y is
adap i e o senso y dep i a ion bu maladap i e o senso y eco e y. Heimle e al. (2014)
ha e sugges ed a mo e balanced amewo k o his impac , s a ing ha c oss-modal plas ici y
migh also ha e maladap i e ou comes in senso y dep i a ion - being ela ed wi h impai ed
unc ions in he emaining senses, namely empo al ac ile p ocessing (Bolognini e al., 2012)
- and po en ial adap i e ou comes in senso y es o a ion – since unc ionally selec i e plas ici y
migh acili a e he eco e y o speci ic cogni i e unc ions (Hassanzadeh, 2012). Senso y
ehabili a ion p og ammes should conside mul isenso y subs i u ions ainings o he dea , as
hey mo e success ully lead o senso y eco e y ac oss he li espan (Heimle & Amedi, 2020).
This hesis was a se o case-s udies ha aimed a in es iga ing c oss-modal plas ici y
in he AC o congeni ally dea pa icipan s h ough an MRI expe imen using pRF analysis.
Resul s showed e ino opic- ela ed esponses p edominan ly wi h nega i e BOLD signal in he
AC o he dea bu no in he hea ing. Unde s anding how he b ain is o ganized and
eo ganized h ough neu oplas ici y con ibu es o he s a e o he a in neu oscience, and o
a be e and mo e independen expe ience o he wo ld by people a ec ed by senso y loss.
CROSS-MODAL PLASTICITY IN THE CONGENITALLY DEAF 47
Annexes
Figu e A1
TWR analysis – phase maps
No e. Phase maps. Pa ame e s we e he co ela ion coe icien and he phase alue. This
analysis was pe o med sepa a ely o (A) eccen ici y ( ing) and (B) pola angle (wedge)
ep esen a ions. The phase o he esponse is ep esen ed by a colou code, o e layed on he
ana omical da a. The colou scale in A has blue ep esen ing he o ea and ed he pe iphe y o
he isual iled. The colou scale in B anges om 0 o 2π. The h eshold is cohe ence ≥ 0.5.
RH – igh hemisphe e. LH – le hemisphe e.