RESEARCH ARTICLE
Oscilla o y and s uc u al signa u es o language plas ici y
in b ain umo pa ien s: A longi udinal s udy
Lucia Amo uso
1,2
| Shuang Geng
1,3
| Nicola Molina o
1,2
| Polina Timo ee a
1,3
|
Sand a Gisbe -Muñoz
1,3
| San iago Gil-Robles
4,5
| Iñigo Pomposo
5
|
Ileana Quiñones
1
| Manuel Ca ei as
1,2,3
1
Basque Cen e on Cogni ion, B ain and
Language (BCBL), San Sebas ian, Spain
2
IKERBASQUE, Basque Founda ion o
Science, Bilbao, Spain
3
Uni e si y o he Basque Coun y, UPV/EHU,
Bilbao, Spain
4
Depa men o Neu osu ge y, Hospi al
Qui on, Mad id, Spain
5
BioC uces Resea ch Ins i u e, Bilbao, Spain
Co espondence
Lucia Amo uso, Basque Cen e on Cogni ion,
B ain and Language (BCBL), 20009 San
Sebas ian, Spain.
Email: [email p o ec ed]
Funding in o ma ion
Basque Go e nmen , G an /Awa d Numbe :
BERC 2018-2021; Spanish Minis y o
Economy and Compe i i eness, G an /Awa d
Numbe : RTI2018-096216-A-I00
(MEGLIOMA) and RTI2018-093547-B-I00
(LangConn); Spanish S a e Resea ch Agency,
G an /Awa d Numbe : SEV-2015-0490 and
IJCI-2017-31373
Abs ac
Recen e idence sugges s ha damage o he language ne wo k igge s i s unc-
ional eo ganiza ion. Ye , he spec o- empo al inge p in s o his plas ic
ea angemen and i s ela ion o ana omical changes is less well unde s ood. He e,
we combined magne oencephalog aphic eco dings wi h a p oxy measu e o whi e
ma e o in es iga e oscilla o y ac i i y suppo ing language plas ici y and i s ela ion
o s uc u al eshaping. Fi s , co ical dynamics we e acqui ed in a g oup o heal hy
con ols du ing objec and ac ion naming. Resul s showed seg ega ed be a
(13–28 Hz) powe dec eases in le en al and do sal pa hways, in a ime-window
associa ed o lexico-seman ic p ocessing (250–500 ms). Six pa ien s wi h le
umo s in ading ei he en al o do sal egions pe o med he same naming ask
be o e and 3 mon hs a e su ge y o umo esec ion. When longi udinally compa -
ing pa ien s' esponses we ound be a compensa ion mimicking he ca ego y-based
seg ega ion showed by con ols, wi h en al and do sal damage leading o selec i e
compensa ion o objec and ac ion naming, espec i ely. A he s uc u al le el, all
pa ien s showed p eope a i e changes in whi e ma e ac s possibly linked o plas-
ici y igge ed by umo g ow h. Fu he mo e, in some pa ien s, s uc u al changes
we e also e iden a e su ge y and showed associa ions wi h longi udinal changes in
be a powe la e aliza ion owa d he con alesional hemisphe e. O e all, ou indings
suppo he exis ence o ana omo- unc ional dependencies in language eo ganiza-
ion and highligh he po en ial ole o oscilla o y ma ke s in acking longi udinal
plas ici y in b ain umo pa ien s. By doing so, hey p o ide aluable in o ma ion o
mapping p eope a i e and pos ope a i e neu al eshaping and plan su gical s a egies
o p ese e language unc ion and pa ien 's quali y o li e.
KEYWORDS
b ain hy hms, b ain umo s, language, magne oencephalog aphy, neu oplas ici y
Recei ed: 11 June 2020 Re ised: 12 Decembe 2020 Accep ed: 14 Decembe 2020
DOI: 10.1002/hbm.25328
This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion-NonComme cial-NoDe i s License, which pe mi s use and dis ibu ion in any
medium, p o ided he o iginal wo k is p ope ly ci ed, he use is non-comme cial and no modi ica ions o adap a ions a e made.
© 2020 The Au ho s. Human B ain Mapping published by Wiley Pe iodicals LLC.
Hum B ain Mapp. 2021;42:1777–1793. wileyonlinelib a y.com/jou nal/hbm 1777
1|INTRODUCTION
Neu oplas ici y e e s o he b ain's abili y o modi y i s s uc u e and
unc ion h oughou he li espan, allowing he acquisi ion o new skills
(Ca ei as e al., 2009; Magui e e al., 2000) bu also coping wi h b ain
damage and disease (Payne & Lombe , 2001). When conside ing his
la e aspec , e idence om human s udies in s oke (Bu e isch
e al., 2005; Shimizu e al., 2002) and b ain umo pa ien s
(Du au, 2005; Robles, Ga ignol, Lehe icy, & Du au, 2008) unde -
sco es he exis ence o di e en plas ici y pa e ns, including unc ion
pe sis ence wi hin he umo , unc ion edis ibu ion in pe ilesional
a eas, ipsilesional ac i a ion o mo e dis an a eas and ec ui men o
con alesional homologs. In line wi h a hodo opical unde s anding o
b ain o ganiza ion (Ca ani, 2007; De Benedic is & Du au, 2011;
Du au, Mo i z-Gasse , & Mandonne , 2014), unc ional ealloca ion
would be possible hanks o he exis ence o edundan co ico–
subco ical pa allel ne wo ks po en ially unmasked by he lesion. I
has been sugges ed (Ius, Angelini, Thiebau de Scho en, Man-
donne , & Du au, 2011) ha his high po en ial o eo ganiza ion
would be almos con ined o he co ical le el, wi h subco ical whi e
ma e showing limi ed o null plas ici y. None heless, e idence om
s oke (Schlaug, Ma china, & No on, 2009) and epilep ic pa ien s ol-
lowing empo al lobec omy (Jeong, Asano, Juhasz, Behen, &
Chugani, 2016; Li e al., 2019), sugges s ha whi e ma e plas ici y in
he con alesional hemisphe e is somehow possible.
B ain unc ion and i s eshaping in he damaged b ain has been
classically s udied by means o unc ional magne ic esonance imaging
( MRI). Howe e , hemodynamic esponses a e slow (one olume e e y
2 s) and unc ions as language, which occu on he subsecond ime-
scale, need also o be examined wi h high- empo al esolu ion ech-
niques capable o acking linguis ic p ocessing in eal- ime. Elec o-
and magne o-encephalog aphy (M/EEG) mee his equi emen as
hey can cap u e neu onal ac i i y and i s oscilla o y dynamics wi h
millisecond ime esolu ion, o e ing a new pe spec i e o s udy b ain
plas ici y (Reid e al., 2016).
Oscilla ions a di e en equency-bands and hei synch oniza-
ion a e hough o e lec communica ion wi hin and be ween egions
(F ies, 2005), ele an o beha io and disease (Uhlhaas e al., 2017).
Recen ly, M/EEG s udies ha e been success ul in iden i ying oscilla-
o y ma ke s o b ain damage and language eco e y, unde sco ing
he in ol emen o low- equency ac i i y in unc ional compensa ion.
Fo ins ance, using MEG, Kiela , Deschamps, Jokel, and Mel ze (2016)
epo ed he in ol emen o con alesional igh alpha-be a ac i i y
du ing seman ic p ocessing in s oke pa ien s. Simila ly, T au
e al. (2019) ound ha b ain umo pa ien s exhibi ed a unc ional
shi in be a language la e aliza ion owa d he igh hemisphe e a e
le umo esec ion. Using EEG, Spi onelli, Man edi, and
Ang illi (2013) e alua ed pos -s oke language eo ganiza ion and
ound bila e al pa e ns o be a ac i i y in ipsilesional on al a eas
and con alesional homologs du ing seman ic p ocessing. Piai, Meye ,
D onke s, and Knigh (2017) epo ed alpha–be a powe dec eases
associa ed o lexico-seman ic e ie al in s oke pa ien s wi h le
hemisphe ic lesions. In e es ingly, while con ols showed a le
la e alized e ec , pa ien s exhibi ed a igh la e alized one, which was
ul ima ely p edic ed by he p obabili y o splenium damage.
He e, we acked oscilla o y dynamics subse ing language plas-
ici y in b ain umo pa ien s be o e and 3 mon hs a e su ge y o
umo esec ion. We eco ded MEG ac i i y while heal hy con ols
and pa ien s o e ly named objec and ac ion pic u es in Spanish. O
no e, b ain umo s could a ec ei he en al o do sal a eas wi hin
he le hemisphe e which a e known o play di e en oles in ep-
esen ing objec and ac ion ca ego ies. B ie ly, p e ious e idence
(Gleichge ch e al., 2016; Lub ano, Fille on, Demone , & Roux, 2014;
Vigliocco, Vinson, D uks, Ba be , & Cappa, 2011) sugges s ha he
seman ic p ocessing o objec and ac ion knowledge is unde pinned
by pa ially dis inc ne wo ks p e e en ially in ol ing in e io - empo al
and on o-pa ie al nodes, espec i ely. Thus, we capi alized on his
dissocia ion o e alua e language unc ion in he heal hy and he
lesioned b ain. Speci ically, we expec ed di e en alpha-be a compen-
sa ion pa e ns depending on umo loca ion and seman ic ca ego y,
wi h en al and do sal lesions mainly comp omising objec and ac ion
p ocessing, espec i ely. We also expec ed ha unc ional changes
would be ela ed o s uc u al ones. Thus, we calcula ed a p oxy mea-
su e o whi e ma e in ol emen in language- ela ed ac s and
assessed i po en ial p eope a i e and/o pos ope a i e s uc u al es-
haping was associa ed wi h unc ional longi udinal changes.
2|MATERIALS AND METHODS
2.1 |Pa icipan s
A o al o 26 pa icipan s ook pa in his s udy. Twen y heal hy
adul s (6 men, age mean = 25.04; SD = 3.94) we e ec ui ed h ough
he BCBL da abase and ecei ed economical compensa ion o hei
pa icipa ion. Six pa ien s (3 men, age ange 24–59; mean = 40;
SD = 12.89) wi h b ain umo s in he le hemisphe e mainly in ol ing
empo al (n= 3), on o-pa ie al (n= 2) o pa ie al egions (n= 1) we e
ec ui ed a he C uces Hospi al whe e hey ecei ed hei diagnosis
and pe o med he awake c anio omy o umo esec ion (see
Figu e 1 o lesion p o ile). One ou o 6 pa ien s had ca e nous angi-
omas, while he o he 5 exhibi ed as ocy omas G ade I and
II. Indi idual pa ien demog aphics, lesion and clinical cha ac e is ics
a e summa ized in Table 1. All pa icipan s we e igh handed as mea-
su ed by Edinbu gh Handedness In en o y (Old ield, 1971). They all
had no mal hea ing and no mal o co ec ed o no mal ision. All
pa ien s and con ols epo ed Spanish as hei i s language and he
a e age naming BEST sco e (de B uin, Ca ei as, & Duñabei ia, 2017)
in Spanish was 63.17/65 o pa ien s and 64.81/65 o con ols. I
should be no ed, howe e , ha pa ien s and con ols also epo ed
knowing some Basque (40/65 o pa ien s and 49/65 o con ols), as
is common in he popula ion o Donos ia-San Sebas ian. The s udy
p o ocol was app o ed by he E hics Boa d o he Euskadi Commi ee
and he E hics and Scien i ic Commi ee o he BCBL, ollowing he
decla a ion o Helsinki. All pa icipan s ga e hei w i en consen
p io o he s udy.
1778 AMORUSO ET AL.
2.2 |S imuli and ask
Seman ic p ocessing was assessed wi h a pic u e naming ask. Pic-
u es we e selec ed om a s anda dized ba e y de eloped by NEURE
clinic®(h ps://www.neu e.eu/). The ask included wo sepa a e se s
o 30 colo ed images wi h line d awings ei he depic ing objec s o a
pe son pe o ming an ac ion, espec i ely. Objec and ac ion s imuli
we e ma ched as close as possible o di e en linguis ic a iables and
di e ences be ween s imuli we e calcula ed using S uden es s o
no mally dis ibu ed a iables and Mann–Whi ney o non-no mally
dis ibu ed ones. Mo e speci ically, s imuli we e ma ched o e-
quency (Objec s: mean = 25.94, SD = 29.19; Ac ions: mean = 14.79,
FIGURE 1 Lesion p o ile wi h espec o majo do sal and en al whi e ma e ac s. Tumo s a e shown in blue. P obabilis ic loca ion o
supe io longi udinal asciculus (SLF, I, II), a cua e asciculus (AF) and in e io on o-occipi al asciculus (IFOF) a e shown in o ange. Pa ien s 1, 2
and 3 (on op) exhibi umo s in ading he le empo al lobe, howe e , none o hem comp omises he IFOF. Pa ien s 4, 5, and 6 (on bo om)
exhibi umo s in ading on o-pa ie al egions a ec ing, in all cases, he SLF and in cases 5 and 6 also he AF
TABLE 1 Demog aphics and clinical cha ac e is ics o pa ien s
Pa ien ID Age (yea s) Gende Educa ion (yea s) Handedness Type o umo Tumo olume (cm
3
) Ex en o esec ion (%)
P1 33 F 21 R Ca e nous
Angioma
2.53 84.48
P2 59 M 12 R As ocy oma
G ade II
5.8 100
P3 24 M 16 R As ocy oma
G ade I
1.85 100
P4 46 F 14 R As ocy oma
G ade II
0.55 75.83
P5 31 F 21 R As ocy oma
G ade II
16.44 100
P6 47 M 19 R As ocy oma
G ade II
74.25 100
AMORUSO ET AL.1779
SD = 17.06, W= 562, p= .055), wo d leng h in e ms o numbe o le -
e s (Objec s: mean = 5, SD = 1; Ac ions: mean = 6, SD =1,
=−1.519, p= .13), and amilia i y (Objec s: mean = 6.26, SD = 0.5;
Ac ions: mean = 6.17, SD = 0.56, = 0.68, p= .49). Name ag eemen
was ≥85% in bo h condi ions.
In sepa a e blocks, pa icipan s we e eques ed o obse e he
pic u es and name hem o e ly in Spanish. P oduc ion o nouns and
e bs was eques ed in he con ex o sho sen ences, which is a
mo e ecological o m o speech han isola ed naming. Mo e speci i-
cally, on op o he objec - ela ed images we added he ex “Es o
es…”[“This is…”in Spanish] o o ce he p oduc ion o a sho sen-
ence ha had o ag ee in numbe wi h he a ge noun (e.g., “This is a
bi d”,“This is an apple”). Simila ly, on op o he ac ion- ela ed pic-
u es, we included he p onouns “El…”o “Ella…”[“He…”o “She…”,
in Spanish]. This in oduc o y ex was used as a cue o he p oduc-
ion o a sen ence ha s a ed wi h he gi en subjec and had a ini e
e b o m in hi d pe son singula (e.g., “She sings,”“He w i es”).
Impo an ly, he use o hese pic u es led o pa icipan s elici ing sen-
ences comp ising conc e e nouns (i.e., as opposed o abs ac con-
cep s like “lo e”) and dynamic mo o ac ions (i.e., as opposed o e bs
e e ing o s a ic s a es like “ hinking”). We used Ma labR2012B and
Cogen Toolbox o pic u e p esen a ion. T ials s a ed wi h a ixa ion
c oss las ing o 500 ms, ollowed by he s imulus displayed o 1 s. ISI
andomly a ied be ween 2 and 4 s. Each pic u e was p esen ed
3 imes o a o al o 90 ials pe block. Each block las ed 10 min,
and pa icipan s we e allowed o ake a sho b eak be ween hem.
2.3 |Beha io al assessmen
Vocal esponses we e eco ded and moni o ed online by he expe i-
men e while pa icipan s pe o med he ask. Naming la encies we e
calcula ed using he Ch onse ool (Roux, A ms ong, &
Ca ei as, 2017) which enables he au oma ic de ec ion o speech
onse . Responses con aining dis luencies o e o s we e coded as
in alid and excluded om MEG analysis. In addi ion, esponse la en-
cies sho e han 200 ms and de ia ing om pa icipan 's mean
la ency by >2.5 SD in each condi ion (Miozzo, Pul e mulle , &
Hauk, 2015) we e also emo ed (in o al, 5.9% o he ials we e
elimina ed). Reac ion imes (RTs) and co ec naming esponses we e
compa ed be ween g oups using nonpa ame ic Welch's - es s and
Wilcoxon signed- ank o p e- e sus pos -su ge y s ages wi hin he
pa ien 's g oup (see Table 2). Fu he mo e, we also an C aw o d-
Howell (1998) equen is - es s o single-case analysis using he
psycho Package (Makowski, 2018) on RS udio (Ve sion 1.2.5019) o
compa e each pa ien o he con ol g oup (see Table 3).
2.4 |MEG and MRI acquisi ion
MEG da a we e acqui ed in a magne ically shielded oom using a
360-channel Elek a-Neu omag sys em (Helsinki, Finland). Eye-
mo emen s we e moni o ed wi h wo pai s o elec odes in a bipola
mon age placed on he ex e nal chan i o each eye (ho izon al EOG) and
abo e and below igh eye ( e ical EOG). Elec oca diog aphic (ECG)
ac i i y was also eco ded wi h wo elec odes, one posi ioned jus
below he igh cla icle and he o he below he le ib bone. MEG sig-
nals we e con inuously eco ded a a 1 kHz sampling a e and on-line il-
e ed o 0.1–330 Hz. The head posi ion inside he helme was
con inuously moni o ed using i e head posi ion indica o (HPI) coils.
The loca ion o each coil ela i e o he ana omical iducials (i.e., he
nasion, and le and igh p eau icula poin s) was de ined wi h a 3D dig-
i ize (Fas akPolhemus, Colches e , VA). Digi aliza ion o he iducials
plus 200 addi ional poin s dis ibu ed o e he pa icipan 's scalp we e
used du ing subsequen da a analysis o spa ially align he MEG senso
coo dina es o he na i e T1 high- esolu ion 3D s uc u al MRI. S uc-
u al images we e acqui ed be o e and 3 mon hs a e su ge y o each
pa icipan wi h a Siemens 3T MAGNETOM PRISMA i MR scanne
(Siemens, Munich, Ge many) in a sepa a e session (i.e., 1 day be o e
he MEG session). T1-weigh ed MPRAGE ana omical olumes
we e acqui ed wi h he ollowing pa ame e s: echo ime = 2.97 ms,
epe i ion ime = 2,530 ms, lip angle = 7and ield o iew = 256 ×
256 ×176 mm
3
, numbe o axial slices = 176, slice hickness = 1 mm,
in-plane esolu ion = 1 mm ×1 mm. The T2-weigh ed luid-a enua ed
in e sion eco e y (FLAIR) sequence used he ollowing pa ame e s:
echo ime = 394 ms, epe i ion ime = 5,000 ms, lip angle = 7and ield
TABLE 2 Beha io al esul s. Mean (M) and SD o accu acy and eac ion ime (RT) in each condi ion o each g oup, wi h p- alues om
Welch's - es s compa ing pe o mance be ween pa ien s and con ols and Wilcoxon signed- ank o p e- e sus pos -su ge y s ages wi hin he
pa ien 's g oup
Con ols M (SD) Pa ien s (PRE) M (SD) Pa ien s (POST) M (SD)
Con ols e sus
pa ien s (PRE)
Con ols e sus
pa ien s (POST)
Pa ien s PRE
e sus POST
p- alue p- alue p- alue
Accu acy (%)
Objec naming 98.75 (2.06) 98.27 (2.83) 98.27 (2.83) .71 .71 1.0
Ac ion naming 97.5 (3.33) 97.6 (3.63) 95.43 (3.54) .93 .24 .34
Reac ion ime (ms)
Objec naming 946.6 (275.5) 963.8 (144.5) 958.8 (219.2) .36 .74 .68
Ac ion naming 1,059.2 (263.7) 1,105.4 (191.3) 1,096.2 (196) .36 .44 .68
1780 AMORUSO ET AL.
o iew = 256 ×256 ×176 mm
3
, numbe o axial slices = 192, 1 mm
iso opic esolu ion.
2.5 |MEG da a p e-p ocessing
Con inuous da a we e ini ially p e-p ocessed o -line using he empo-
al ex ension o he signal space sepa a ion me hod (Taulu and Simola,
2006) implemen ed in Max il e 2.2 (Elek a-Neu omag), which sub-
ac s ex e nal magne ic noise om he MEG eco dings, co ec s o
head mo emen s and in e pola es bad channels wi h algo i hms
implemen ed in he so wa e. Subsequen analyses we e pe o med
using he FieldT ip oolbox e sion 20170911(Oos en eld, F ies,
Ma is, & Scho elen, 2011) in Ma labR2014B. Reco dings we e down-
sampled o 500 Hz and segmen ed in o epochs ime-locked o s imu-
lus p esen a ion (i.e., pic u e o be named) om 500 ms be o e image
onse o 1,000 ms a e image onse .
Da a we e il e ed wi h a DFT il e o emo e line noise. A semi-
au oma ic p ocedu e was hen employed o emo e epochs wi h elec-
omyog aphic a i ac s, SQUID jumps and la signal. A as indepen-
den componen analysis (ICA) was used o iden i y eye mo emen s,
blinks and elec oca diog aphic a i ac s (Jung e al., 2000). The
da ase s o ou heal hy pa icipan s we e excluded om he analysis
due o excessi e blinking and/o muscula a i ac s esul ing in he
loss o a la ge numbe o ials (70%). Thus, subsequen analyses
we e pe o med on a o al o 16 heal hy pa icipan s.
TABLE 3 Compa ison o indi idual pa ien sco es o con ol g oup pe o mance du ing naming. Mean (M), - alues, and p- alues om
C aw o d-Howell - es s compa ing accu acy and eac ion ime (RT) du ing objec and ac ion naming be o e and a e su ge y o umo esec ion
P e-su ge y Pos -su ge y
Mean - alue p- alue Mean - alue p- alue
Objec naming
Reac ion imes (ms)
P1 958.76 0.04 .96 1,014.09 0.23 .81
P2 1,198.29 0.88 .38 1,357.76 1.45 .16
P3 868.91 −0.27 .78 789.25 −0.55 .58
P4 818.81 −0.45 .65 762.78 −0.64 .52
P5 1,067.23 0.42 .67 973.62 0.09 .92
P6 870.76 −0.26 .79 855.33 −0.32 .75
Accu acy (%)
P1 96.29 −1.16 .26 96.29 −1.16 .26
P2 93.33 −1.21 .24 93.33 −1.21 .24
P3 100 0.58 .56 100 0.58 .56
P4 100 0.58 .56 100 0.58 .56
P5 100 0.58 .56 100 0.58 .56
P6 100 0.58 .56 100 0.58 .56
Ac ion naming
Reac ion imes (ms)
P1 1,004.07 −0.20 .84 1,087.4 0.10 .91
P2 1,444.15 1.42 .17 1,467.46 1.5 .15
P3 890.12 −0.62 .54 882.57 −0.65 .52
P4 1,027.25 −0.11 .9 1,022.13 −0.13 .89
P5 1,177.37 0.43 .67 1,071.77 0.04 .96
P6 1,089.62 0.11 .91 1,046.11 −0.04 .96
Accu acy (%)
P1 100 0.72 .47 100 0.72 .47
P2 92.6 −1.43 .17 92.6 −1.43 .17
P3 100 0.72 .47 100 0.72 .47
P4 93.33 −1.21 .24 93.33 −1.21 .24
P5 100 0.72 .47 93.33 −1.21 .24
P6 100 0.72 .47 93.33 −1.21 .24
AMORUSO ET AL.1781
2.6 |Senso le el analysis
Time- equency ep esen a ions (TFR) we e calcula ed om he
a i ac - ee MEG segmen s o equencies anging om 1 o 30 Hz.
TRFs we e ob ained using Hanning ape s and a ixed window leng h
o 500 ms ad ancing in 10 ms s eps, gi ing ise o a 2 Hz equency
esolu ion. Powe es ima es we e calcula ed sepa a ely o each
o hogonal di ec ion o a g adiome e pai and hen combined,
esul ing in 102 measu emen channels. Powe was exp essed as ela-
i e change wi h espec o a 500 ms p e-s imulus baseline. On
a e age, condi ions comp ised 42.33 (SD = 2.02) a i ac - and e o -
ee ials o pa ien s (no di e ences in ial numbe be ween objec
and ac ion naming condi ions o p e- and pos -su ge y sessions,
Wilcoxon signed ank, all ps > .11) and 46.22 (SD = 7.22) o heal hy
con ols (no di e ence be ween objec and ac ion condi ions, p= .22).
Impo an ly, no di e ences in he numbe o ials be ween pa ien s
and heal hy adul s we e obse ed o objec and ac ion naming ei he
be o e o a e su ge y (Welch's - es s, all ps > .18).
2.7 |Selec ion o equency-band and ime-
windows
P e ious M/EEG s udies indica e ha powe changes in he alpha and
be a equency-bands (Piai, Roelo s, Romme s, & Ma is, 2015) e lec
he e ie al o lexical-seman ic in o ma ion. Fu he mo e, e idence
om s udies on s oke (Kiela e al., 2016; Piai e al., 2017; Spi onelli
e al., 2013) and b ain umo pa ien s (Liza azu e al., 2020; T au
e al., 2019) poin s o an in ol emen o alpha and be a ac i i y in
unc ional compensa ion. Thus, we ocused ou analysis on low-
equency ac i i y including alpha (8–12 Hz) and be a (13–28 Hz)
oscilla ions. The ime window was p ima ily chosen based on me hod-
ological cons ain s imposed by ou ask. Indeed, p e ious s udies
show ha in o e p oduc ion asks a i ac - ee b ain eco dings can
be measu ed up o app oxima ely 400 ms pos -s imulus p esen a-
ion (Ganushchak, Ch is o els, & Schille , 2011). Based on his e i-
dence and isual inspec ion o he onse o speech p oduc ion in ou
da a, we ocused ou TFR analysis o he 0–500 ms ime window a e
pic u e onse . Mo e speci ically, we selec ed wo ime-windows cap-
u ing ea ly (0–200 ms) and la e (200–500 ms) pic u e-naming ela ed
p ocesses, including isual ecogni ion, and concep ualiza ion and lexi-
cal selec ion, espec i ely (Inde ey, 2011; Inde ey & Le el , 2004;
Liljes om, Kujala, S e enson, & Salmelin, 2015).
2.8 |S a is ical analysis
In o de o e alua e unc ional compensa ion wi hin pa ien s' lan-
guage ne wo k we i s explo ed he spec o- empo al pa e n o
esponses igge ed by ou pic u e naming ask in a g oup o heal hy
con ols. Mo e speci ically, we calcula ed TFRs o objec and ac ion
naming in ea ly and la e ime-windows ela i e o pic u e onse and
compa ed each o hem ela i e o p e-s imulus baseline ac i i y.
Then, once main oscilla o y pa e ns igge ed by he ask we e iden i-
ied in con ols, we assessed longi udinal changes in he g oup o
pa ien s. Speci ically, we calcula ed TFRs o objec and ac ion naming
be o e umo esec ion and con as ed hem wi h hose ob ained
3 mon hs a e he su ge y.
In all cases, di e ences in spec al powe be ween condi ions a
he senso le el we e assessed using clus e -based pe mu a ion es s
(Ma is & Oos en eld, 2007). This es con ols o mul iple compa i-
sons using a clus e -based co ec ion while main aining sensi i i y
based on empo al, spa ial and equency dependency o neighbo ing
samples. The pe mu a ion p- alue was calcula ed using he Mon e
Ca lo me hod wi h 1,000 andom pe mu a ions. The h eshold o sig-
ni icance es ing was a p- alue below 5% ( wo- ailed). Please no e
ha he inding o a signi ican clus e implies ha he e is a signi i-
can di e ence be ween condi ions. Howe e , he clus e does no
p o ide exac in o ma ion abou he iming and he spa ial loca ion o
he e ec . In o he wo ds, no s a emen s abou he onse /o se o
he e ec a he millisecond le el o abou i s spa ial ex en can be
made (Sassenhagen & D aschkow, 2019). While we had clea hypo h-
eses abou he equency-bands po en ially in ol ed in he language
e ec s (i.e., alpha-be a), no speci ic a p io i hypo heses abou iming
and/o loca ion we e held. Thus, we a e aged o e equency bins
(alpha cen al equency = 10.13 Hz and be a cen al
equency = 20.66 Hz;) bu conside ed all senso s (i.e., combined g a-
diome e s) and ime-poin s wi hin ea ly and la e ime-windows in he
analysis.
2.9 |Sou ce localiza ion
Pa icipan s' high- esolu ion 3D s uc u al MRIs we e segmen ed
using F eesu e so wa e (Dale & Se eno, 1993). Co- egis a ion
be ween he MEG senso coo dina es and he pa icipan 's MRI coo -
dina es was done by manually aligning he digi ized head-su ace and
iducial poin s o he ou e scalp su ace. The o wa d model was
compu ed using he Bounda y Elemen Me hod (BEM) implemen ed
in he MNE so wa e sui e (G am o e al., 2014); RRID:SCR_005972)
o h ee o hogonal angen ial cu en dipoles (one o each spa ial
dimension) placed on a homogeneous 5-mm g id sou ce space co e -
ing he whole b ain. Fo each sou ce, he o wa d model was hen
educed o i s wo p incipal componen s o highes singula alue,
which closely co espond o sou ces angen ial o he skull. We used
bo h g adiome e s and magne ome e s in he sou ce es ima ion, no -
malizing each senso signal by i s noise a iance (500-ms baseline
pe iod p io o pic u e onse ). B ain sou ce ac i i y was calcula ed o
each pa icipan using Linea ly Cons ained Minimum Va iance
(LCMV) beam o me app oach (Van Veen, an D ongelen,
Yuch man, & Suzuki, 1997). A common il e was compu ed by com-
bining he c oss-spec al densi y (CSD) ma ices om he ime-
equency window o he signi ican senso -le el e ec s and an
equally-sized baseline pe iod p io o pic u e onse . The common il e
was hen applied sepa a ely o each condi ion o es ima e sou ce
powe . Since we ocused ou analysis on he local sou ce powe , we
1782 AMORUSO ET AL.
only used eal- alued il e coe icien s (G u zne e al., 2010). To no -
malize sou ce ac i i y, he neu al ac i i y index (NAI) was calcula ed
as a ce ain a io be ween he powe in he expe imen al condi ions
and he p e-s imulus baseline (Pcond-Pbase./Pbase). Fo each session
(p e- and pos -su ge y), he MEG maps we e i s co- egis e ed wi h
hei co esponding indi idual MRIs and hen no malized o he s an-
da d MNI o un g oup le el analyses. This was done by applying a
non-linea ans o ma ion using he spa ial-no maliza ion algo i hm
implemen ed in SPM8 and i was checked by one o he au ho s (LA).
G oup analyses we e pe o med wi h he loca ion-compa ison
me hod desc ibed in (Bou guignon, Molina o, & Wens, 2018). B ie ly,
his me hod gene a es boo s ap g oup-a e aged maps o build a pe -
mu a ion dis ibu ion o loca ion di e ence be ween local maxima in
he wo condi ions being compa ed, and es he null hypo hesis ha
his dis ance is ze o. Local maxima is de ined as se s o con iguous
oxels displaying highe powe han all o he neighbo ing oxels. The
h eshold o s a is ical es ing a p< .05 was compu ed as he
95-pe cen ile o he pe mu a ion dis ibu ion. All sup a- h eshold local
MEG peaks we e in e p e ed as indica i e o b ain egions likely ig-
ge ing he senso -le el e ec s. This obus me hod has shown o deal
well wi h he spec al leakage o he sou ce-p ojec ed MEG da a
which can esul om di ec ly con as ing b ain maps o di e en
condi ions.
2.10 |3D lesion econs uc ion
Lesions we e manually d awn on he na i e space o pa icipan s'
T1-weigh ed MPRAGE image by a ained echnician using he
MRIc on so wa e (Ro den, Ka na h, & Bonilha, 2007) and u he
supe ised by he neu osu geons in cha ge o he pa ien s' awake c a-
nio omy (SGR and IPG). The econs uc ion was pe o med also using
in o ma ion om T2 images when lesion bounda ies we e no clea in
he T1 MRI. The lesion was hen no malized o he MNI empla e and
alignmen be ween he econs uc ed lesion and he lesion in he na i e
space was checked by one o he au ho s (IQ). A olume o in e es
(VOI) was c ea ed o each pa ien each ime poin . F om each p e- and
pos -su ge y 3D econs uc ion, he umo olume (cm
3
) was calcu-
la ed. Ex en o esec ion (cm
3
) was measu ed on pos ope a i e imaging
as: (Volume o (p eope a i e 3D Tumo Recons uc ion pos ope a i e
Resec ion)*100/p eope a i e umo olume).
2.11 |S uc u al measu e o whi e ma e changes
Fo he s uc u al analysis, p eope a i e and pos ope a i e T1 and T2
images we e p e-p ocessed and analyzed using he Voxel-Based Mo -
phome y (VBM) oolbox and he SPM12 so wa e package. Images
we e co ec ed o bias- ield inhomogenei y; classi ied in o g ay, whi e
ma e and ce eb ospinal luid; egis e ed o a s anda d MNI space using
high-dimensional DARTEL no maliza ion (Ashbu ne , 2007) and u he
smoo hed wi h a 6 mm ull wid h hal maximum (FWHM) Gaussian ke -
nel. We used a segmen a ion app oach based on an adap i e maximum,
a pos e io echnique which does no need a p io i in o ma ion abou is-
sue p obabili ies (Rajapakse, Giedd, & Rapopo , 1997). We u he
e ined his p ocedu e, by accoun ing o pa ial olume e ec s and by
applying a hidden Ma ko andom ield model which inco po a es spa ial
p io in o ma ion o he adjacen oxels in o he segmen a ion es ima-
ion (Tohka, Zijdenbos, & E ans, 2004).
To assess po en ial di e ences in whi e ma e in ol emen we
used a egion o in e es (ROI) app oach. ROIs we e de ined using a
p obabilis ic ac og aphy a las (Rojko a e al., 2016). The selec ed
ac s we e he supe io longi udinal asciculus (SLF I), he a cua e as-
ciculus (AF, long b anch) and he in e io - on o-occipi al asciculus
(IFOF), which cons i u e key bundles wi hin do sal and en al lan-
guage pa hways and hei damage is known o a ec language
p ocessing (Agos a e al., 2013; Almai ac, He be , Mo i z-Gasse , de
Champ leu , & Du au, 2015; Ca ani & Mesulam, 2008; Mandelli
e al., 2014). Fo each o hese ROIs, we ex ac ed p eope a i e and
pos ope a i e mean olumes in le and igh hemisphe es and
co ec ed i o b ain size using he o al in ac anial olume (TIV). By
doing so, we ob ained a p oxy ma ke o whi e ma e in ol emen
based on lesion dis ibu ion in ela ion o whi e-ma e p obabilis ic
dis ibu ion de i ed om he ac og aphy a las (Rojko a e al., 2016).
Compa isons be ween pa ien 's mo phome ic alues and con ols
we e pe o med using C aw o d-Howell - es s.
2.12 |Co ela ional analysis be ween s uc u e
and unc ion
Fi s , we calcula ed a Language la e ali y index (LI) using he ollowing
o mula:
LI = R−LðÞ=R+LðÞ:
whe e “R”and “L” ep esen powe a e aged ac oss senso s o nam-
ing condi ions (objec and ac ion pooled oge he ) in igh and le
hemisphe es, espec i ely; hus yielding posi i e alues o igh -
la e alized and nega i e alues o le -la e alized language- ela ed
ac i i y. Gi en he le -la e alized pa e n o oscilla o y be a esponses
obse ed in he heal hy con ol g oup as well as p e ious s udies
using be a powe o calcula ing LI in speech p oduc ion asks (T au
e al., 2019), we easoned ha be a ac i i y (13–28 Hz) was be e
sui ed han alpha o cap u e a po en ial shi owa d he igh hemi-
sphe e igge ed by umo p esence and/o esec ion. Thus, he index
was calcula ed o each pa ien and session (i.e., be o e and a e su -
ge y) only in he be a band. Fu he mo e, gi en he common le -
la e alized pa e n obse ed in bo h objec and ac ion naming, we
combined hem in o a unique naming condi ion o educe dimension-
ali y and ob ain a highe signal- o-noise a io in he da a. Please no e,
ha his me hodological choice (i.e., ocus on be a esponses) was u -
he suppo ed by he longi udinal con as in pa ien s, showing oscil-
la o y e ec s ci cumsc ibed o he be a equency-band (see below).
LI was es ed sepa a ely in con ols and pa ien s wi h Wilcoxon signed
ank es s agains ze o and be ween g oups using Welch's - es s.
AMORUSO ET AL.1783
In addi ion, he same index was used o calcula e la e aliza ion o
whi e ma e ac s in con ols and pa ien s. Mo e speci ically, in he
case o pa ien s, his index was calcula ed sepa a ely in p eope a i e
and pos ope a i e s ages and, in each case, indi idually compa ed
agains he con ol g oup using C aw o d-Howell - es s.
Finally, Pea son co ela ions be ween p eope a i e and pos ope -
a i e whi e ma e ROIs LI and be a longi udinal changes (pos −p e-
su ge y be a LI) we e un o es ablish whe he po en ial eshaping a
he s uc u al le el was associa ed wi h unc ional one.
3|RESULTS
3.1 |Beha io al esul s
Table 2 shows mean accu acy and eac ion ime alues (RT) o
heal hy con ols and pa ien s, as well as con as s be ween g oups
and su ge y s ages. O e all, no di e ences in pe o mance
(i.e., accu acy and RTs) we e obse ed be ween g oups. These esul s
we e u he con i med a he indi idual pa ien le el wi h C aw o d-
Howell - es s, which suppo ed he absence o signi ican beha io al
e ec s (see Table 3). In addi ion, no di e ences we e obse ed wi hin
pa ien s when compa ing pe o mance be o e and a e su ge y. This
inding was well expec ed, gi en ha pa ien s wi h slow-g owing
b ain umo s ypically exhibi a no mal neu ological and beha io al
explo a ion, a leas when conside ing ela i ely easy low-le el asks
(DeAngelis, 2001). Fu he mo e, he main enance o his beha io al
pa e n a e su ge y speaks in a o o success ul language
compensa ion.
3.2 |Oscilla o y signa u es o pic u e naming in
heal hy con ols
Figu e 2 shows he ime- equency ep esen a ions (TFRs), opo-
g aphical dis ibu ions and sou ce localiza ion plo s o he naming
condi ions as compa ed o baseline in he alpha (8–12 Hz) and be a
(13–28 Hz) equency-bands.
Ea ly ime-window e ec s (0–200 ms): Objec and ac ion naming
showed ea ly alpha powe inc eases as compa ed o baseline. This
e ec was highligh ed by signi ican posi i e clus e s (bo h Mon e
Ca lo ps = .004, wo- ailed), o e bila e al pos e io and le middle
senso s, in he case o objec s; and bila e al pos e io senso s in he
case o ac ions. Sou ce localiza ion o ea ly alpha e ec s iden i ied
he ela ed local maxima in occipi o-pa ie al egions. In addi ion, simi-
la be a powe inc eases we e obse ed o objec and ac ion naming
condi ions as indica ed by signi ican posi i e clus e s o e le pos e-
io and middle senso s (bo h Mon e Ca lo ps = .01, wo- ailed).
FIGURE 2 Oscilla o y signa u es o speech p oduc ion in heal hy con ols. TFR o alpha and be a powe in he objec ( op panel) and ac ion
(bo om panel) condi ions o e ime. TFRs a e plo ed as ela i e powe change compa ed o he baseline pe iod o e ep esen a i e signi ican
senso s (objec s = M1632 + M1633; highligh ed in o ange; ac ions = M1722 + M1723; highligh ed in blue). Topog aphic dis ibu ion plo s show
pos e io alpha and be a powe inc eases a ea ly s ages (0–200 ms), and le -la e alized an e io and pos e io be a powe dec eases a la e
s ages (200–500 ms)
1784 AMORUSO ET AL.
Sou ce localiza ion showed local maxima peaking in le angula gy us
and in e io on al gy us (IFG) o objec s; and in le sup ama ginal
and o bi o on al a eas o ac ions (Figu e 2, igh panel).
La e ime-window e ec s (200–500 ms): Du ing his pe iod, a sig-
ni ican nega i e clus e (Mon e Ca lo p= .01, wo- ailed)
encompassing bila e al pos e io and middle senso s e ealed alpha
powe dec eases only o he objec naming condi ion. A he sou ce
le el, his e ec showed local minima in bila e al occipi o- empo al
egions.
Finally, bo h condi ions showed be a powe dec eases ha
we e unde sco ed by signi ican nega i e clus e s o e bila e al
pos e io and an e io senso s in he case o objec s (Mon e Ca lo
p= .004, wo- ailed); and bila e al pos e io and le an e io sen-
so s in he case o ac ions (Mon e Ca lo p=.01, wo- ailed).Sou ce
localiza ion o be a e ec s showed local minima in he le IFG,
i espec i ely o he naming condi ion, while objec naming addi-
ionally ec ui ed he le an e io empo al pole and ac ion naming
he le supe io pa ie al and do sal p emo o co ex (Figu e 2, le
panel).
O e all, hese indings p o ide a baseline o in e p e po en ial
language eshaping in pa ien s. In b ie , hey suppo p e ious M/EEG
s udies (Piai e al., 2015, 2017) showing he in ol emen o alpha-be a
oscilla ions in speech p oduc ion and align well wi h e idence indica -
ing he exis ence o pa ially non-o e lapping ne wo ks o he
p ocessing o objec and ac ion knowledge, showing a di e se con i-
bu ion o en al and do sal nodes o he language ne wo k, espec-
i ely (Vigliocco e al., 2011).
3.3 |Func ional plas ici y in b ain umo pa ien s
Figu e 3 shows TFRs, opog aphic dis ibu ions o he objec e ec
ound in pa ien s wi h en al empo al lesions (Figu e 3a) and o he
ac ion e ec , ound in pa ien s wi h do sal on o-pa ie al lesions
(Figu e 3b). O e all, when compa ing oscilla o y ac i i y ac oss ses-
sions (pos - s. p e-su ge y o umo esec ion) wi hin each g oup o
pa ien s, we ound signi ican di e ences be ween sessions in he
be a band (13–28 Hz), wi h powe inc eases a e umo esec ion.
FIGURE 3 Longi udinal unc ional plas ici y in b ain umo pa ien s. (a) TFRs o pa ien s wi h en al lesions showing be a powe inc eases
a e su ge y only o objec s; (b) TFRs o pa ien s wi h do sal lesions showing a simila e ec bu only o ac ions. TFRs a e plo ed as ela i e
powe changes compa ed o he baseline pe iod in a e aged signi ican senso s (shown in black o e he opog aphical dis ibu ion o he
signi ican clus e s)
AMORUSO ET AL.1785
Hauk, O., Johns ude, I., & Pul e mulle , F. (2004). Soma o opic ep esen a-
ion o ac ion wo ds in human mo o and p emo o co ex. Neu on,41
(2), 301–307. h ps://doi.o g/10.1016/s0896-6273(03)00838-9
He be , G., Mo i z-Gasse , S., & Du au, H. (2017). Di ec e idence o he
con ibu i e ole o he igh in e io on o-occipi al asciculus in non-
e bal seman ic cogni ion. B ain S uc u e & Func ion,222(4),
1597–1610. h ps://doi.o g/10.1007/s00429-016-1294-x
Hu enloche , J., & Lui, F. (1979). The seman ic o ganiza ion o some sim-
ple nouns and e bs. Jou nal o Ve bal Lea ning and Ve bal Beha io ,18
(2), 141–162.
Inde ey, P. (2011). The spa ial and empo al signa u es o wo d p oduc-
ion componen s: A c i ical upda e. F on ie s in Psychology,2, 255.
h ps://doi.o g/10.3389/ psyg.2011.00255
Inde ey, P., & Le el , W. J. (2004). The spa ial and empo al signa u es o
wo d p oduc ion componen s. Cogni ion,92(1–2), 101–144. h ps://
doi.o g/10.1016/j.cogni ion.2002.06.001
Ius, T., Angelini, E., Thiebau de Scho en, M., Mandonne , E., & Du au, H.
(2011). E idence o po en ials and limi a ions o b ain plas ici y using
an a las o unc ional esec abili y o WHO g ade II gliomas: owa ds a
“minimal common b ain”.Neu oImage,56(3), 992–1000. h ps://doi.
o g/10.1016/j.neu oimage.2011.03.022
Jeong, J. W., Asano, E., Juhasz, C., Behen, M. E., & Chugani, H. T. (2016).
Pos ope a i e axonal changes in he con ala e al hemisphe e in chil-
d en wi h medically e ac o y epilepsy: A longi udinal di usion enso
imaging connec ome analysis. Human B ain Mapping,37(11),
3946–3956. h ps://doi.o g/10.1002/hbm.23287
Jung, T. P., Makeig, S., Humph ies, C., Lee, T. W., Mckeown, M. J., I agui, V.,
& Sejnowski, T. J. (2000). Remo ing elec oencephalog aphic a i ac s
by blind sou ce sepa a ion. Psychophysiology,37(02), 163–178.
Kamali, A., Flande s, A. E., B ody, J., Hun e , J. V., & Hasan, K. M. (2014).
T acing supe io longi udinal asciculus connec i i y in he human
b ain using high esolu ion di usion enso ac og aphy. B ain S uc-
u e & Func ion,219(1), 269–281. h ps://doi.o g/10.1007/s00429-
012-0498-y
Kemme e , D., Rud au , D., Manzel, K., & T anel, D. (2012). Beha io al pa -
e ns and lesion si es associa ed wi h impai ed p ocessing o lexical
and concep ual knowledge o ac ions. Co ex,48(7), 826–848. h ps://
doi.o g/10.1016/j.co ex.2010.11.001
Kiela , A., Deschamps, T., Jokel, R., & Mel ze , J. A. (2016). Func ional eo -
ganiza ion o language ne wo ks o seman ics and syn ax in ch onic
s oke: E idence om MEG. Human B ain Mapping,37(8), 2869–2893.
h ps://doi.o g/10.1002/hbm.23212
Kopell, N., E men ou , G. B., Whi ing on, M. A., & T aub, R. D. (2000).
Gamma hy hms and be a hy hms ha e di e en synch oniza ion
p ope ies. P oceedings o he Na ional Academy o Sciences o he
Uni ed S a es o Ame ica,97(4), 1867–1872. h ps://doi.o g/10.1073/
pnas.97.4.1867
Laaksonen, H., Kujala, J., Hul en, A., Liljes om, M., & Salmelin, R. (2012).
MEG e oked esponses and hy hmic ac i i y p o ide spa io empo-
ally complemen a y measu es o neu al ac i i y in language p oduc-
ion. Neu oImage,60(1), 29–36. h ps://doi.o g/10.1016/j.neu oimage.
2011.11.087
Li, W., An, D., Tong, X., Liu, W., Xiao, F., Ren, J., …Zhou, D. (2019). Di e -
en pa e ns o whi e ma e changes a e success ul su ge y o
mesial empo al lobe epilepsy. Neu oimage: Clinical,21, 101631.
h ps://doi.o g/10.1016/j.nicl.2018.101631
Liljes om, M., Kujala, J., S e enson, C., & Salmelin, R. (2015). Dynamic
econ igu a ion o he language ne wo k p eceding onse o speech in
pic u e naming. Human B ain Mapping,36(3), 1202–1216. h ps://doi.
o g/10.1002/hbm.22697
Liza azu, M., Gil-Robles, S., Pomposo, I., Na a, S., Amo uso, L.,
Quiñones, I., & Ca ei as, M. (2020). Spa io empo al dynamics o pos -
ope a i e unc ional plas ici y in pa ien s wi h b ain umo s in lan-
guage a eas. B ain and Language,202, 104741. h ps://doi.o g/10.
1016/j.bandl.2019.104741
Lub ano, V., Fille on, T., Demone , J. F., & Roux, F. E. (2014). Ana omical
co ela es o ca ego y-speci ic naming o objec s and ac ions: A b ain
s imula ion mapping s udy. Human B ain Mapping,35(2), 429–443.
h ps://doi.o g/10.1002/hbm.22189
Magui e, E. A., Gadian, D. G., Johns ude, I. S., Good, C. D., Ashbu ne , J.,
F ackowiak, R. S., & F i h, C. D. (2000). Na iga ion- ela ed s uc u al
change in he hippocampi o axi d i e s. P oceedings o he Na ional
Academy o Sciences o he Uni ed S a es o Ame ica,97(8), 4398–4403.
h ps://doi.o g/10.1073/pnas.070039597
Makowski. (2018). The psycho package: An e icien and publishing-
o ien ed wo k low o psychological science. Jou nal o Open Sou ce
So wa e,3(22), 470. h ps://doi.o g/10.21105/joss.00470
Mak is, N., Kennedy, D. N., McIne ney, S., So ensen, A. G., Wang, R.,
Ca iness, V. S., J ., & Pandya, D. N. (2005). Segmen a ion o subcom-
ponen s wi hin he supe io longi udinal ascicle in humans: A quan i-
a i e, in i o, DT-MRI s udy. Ce eb al Co ex,15(6), 854–869. h ps://
doi.o g/10.1093/ce co /bhh186
Mandelli, M. L., Ca e zasi, E., Binney, R. J., Hen y, M. L., Lobach, I.,
Block, N., …Go no-Tempini, M. L. (2014). F on al whi e ma e ac s
sus aining speech p oduc ion in p ima y p og essi e aphasia. The Jou -
nal o Neu oscience,34(29), 9754–9767. h ps://doi.o g/10.1523/
JNEUROSCI.3464-13.2014
Ma is, E., & Oos en eld, R. (2007). Nonpa ame ic s a is ical es ing o EEG-
and MEG-da a. Jou nal o Neu oscience Me hods,164(1), 177–190.
Miozzo, M., Pul e mulle , F., & Hauk, O. (2015). Ea ly pa allel ac i a ion o
seman ics and phonology in pic u e naming: E idence om a mul iple
linea eg ession MEG s udy. Ce eb al Co ex,25(10), 3343–3355.
h ps://doi.o g/10.1093/ce co /bhu137
Moseley, R. L., Pul e mulle , F., & Sh y o , Y. (2013). Senso imo o seman-
ics on he spo : B ain ac i i y dissocia es be ween concep ual ca ego-
ies wi hin 150 ms. Scien i ic Repo s,3, 1928. h ps://doi.o g/10.
1038/s ep01928
Noonan, K. A., Je e ies, E., Visse , M., & Lambon Ralph, M. A. (2013).
Going beyond in e io p e on al in ol emen in seman ic con ol: E i-
dence o he addi ional con ibu ion o do sal angula gy us and pos-
e io middle empo al co ex. Jou nal o Cogni i e Neu oscience,25
(11), 1824–1850. h ps://doi.o g/10.1162/jocn_a_00442
Old ield, R. C. (1971). The assessmen and analysis o handedness: The
Edinbu gh in en o y. Neu opsychologia,9(1), 97–113.
Oos en eld, R., F ies, P., Ma is, E., & Scho elen, J. M. (2011). FieldT ip:
Open sou ce so wa e o ad anced analysis o MEG, EEG, and in a-
si e elec ophysiological da a. Compu a ional In elligence and Neu osci-
ence,2011, 156869–156869. h ps://doi.o g/10.1155/2011/156869
Payne, B. R., & Lombe , S. G. (2001). Recons uc ing unc ional sys ems
a e lesions o ce eb al co ex. Na u e Re iews. Neu oscience,2(12),
911–919. h ps://doi.o g/10.1038/35104085
Piai, V., De Wi e, E., Sie powska, J., Zheng, X., Hinkley, L. B., Mizui i, D., …
Naga ajan, S. S. (2020). Language neu oplas ici y in b ain umo
pa ien s e ealed by magne oencephalog aphy. Jou nal o Cogni i e
Neu oscience,32,1–11. h ps://doi.o g/10.1162/jocn_a_01561
Piai, V., Meye , L., D onke s, N. F., & Knigh , R. T. (2017). Neu oplas ici y
o language in le -hemisphe e s oke: E idence linking subsecond
elec ophysiology and s uc u al connec ions. Human B ain Mapping,
38(6), 3151–3162. h ps://doi.o g/10.1002/hbm.23581
Piai, V., Roelo s, A., Romme s, J., & Ma is, E. (2015). Be a oscilla ions
e lec memo y and mo o aspec s o spoken wo d p oduc ion. Human
B ain Mapping,36(7), 2767–2780. h ps://doi.o g/10.1002/hbm.
22806
Pisoni, A., Ma a elli, G., Casa o i, A., Comi, A., Ri a, M., Bello, L., &
Papagno, C. (2018). Objec -ac ion dissocia ion: A oxel-based lesion-
symp om mapping s udy on 102 pa ien s a e glioma emo al.
Neu oimage: Clinical,18, 986–995. h ps://doi.o g/10.1016/j.nicl.
2018.03.022
Pul e mulle , F., Lu zenbe ge , W., & P eissl, H. (1999). Nouns and e bs
in he in ac b ain: E idence om e en - ela ed po en ials and high-
1792 AMORUSO ET AL.
equency co ical esponses. Ce eb al Co ex,9(5), 497–506. h ps://
doi.o g/10.1093/ce co /9.5.497
Rajapakse, J. C., Giedd, J. N., & Rapopo , J. L. (1997). S a is ical app oach
o segmen a ion o single-channel ce eb al MR images. IEEE T ansac-
ions on Medical Imaging,16(2), 176–186. h ps://doi.o g/10.1109/42.
563663
Reid, L. B., Boyd, R. N., Cunning on, R., Rose, S. E. (2016). In e p e ing
in e en ion induced neu oplas ici y wi h m i: he case o mul imodal
imaging s a egies. Neu Plas ,1–13.
Robles, S. G., Ga ignol, P., Lehe icy, S., & Du au, H. (2008). Long- e m
b ain plas ici y allowing a mul is age su gical app oach o Wo ld
Heal h O ganiza ion g ade II gliomas in eloquen a eas. Jou nal o Neu-
osu ge y,109(4), 615–624. h ps://doi.o g/10.3171/JNS/2008/109/
10/0615
Rojko a, K., Volle, E., U banski, M., Humbe , F., Dell'Acqua, F., & Thiebau
de Scho en, M. (2016). A lasing he on al lobe connec ions and hei
a iabili y due o age and educa ion: A sphe ical decon olu ion
ac og aphy s udy. B ain S uc u e & Func ion,221(3), 1751–1766.
h ps://doi.o g/10.1007/s00429-015-1001-3
Ro den, C., Ka na h, H. O., & Bonilha, L. (2007). Imp o ing lesion-symp om
mapping. Jou nal o Cogni i e Neu oscience,19(7), 1081–1088. h ps://
doi.o g/10.1162/jocn.2007.19.7.1081
Roux, F., A ms ong, B. C., & Ca ei as, M. (2017). Ch onse : An au oma ed
ool o de ec ing speech onse . Beha io Resea ch Me hods,49(5),
1864–1881. h ps://doi.o g/10.3758/s13428-016-0830-1
Saa inen, T., Laaksonen, H., Pa iainen, T., & Salmelin, R. (2006). Mo o
co ex dynamics in isuomo o p oduc ion o speech and non-speech
mou h mo emen s. Ce eb al Co ex,16(2), 212–222. h ps://doi.o g/
10.1093/ce co /bhi099
Salmelin, R., & Sams, M. (2002). Mo o co ex in ol emen du ing e bal
e sus non- e bal lip and ongue mo emen s. Human B ain Mapping,
16(2), 81–91. h ps://doi.o g/10.1002/hbm.10031
Sa ubbo, S., De Benedic is, A., Me le , S., Mandonne , E., Balbi, S.,
G anie i, E., & Du au, H. (2015). Towa ds a unc ional a las o human
whi e ma e . Human B ain Mapping,36(8), 3117–3136. h ps://doi.
o g/10.1002/hbm.22832
Sassenhagen, J., & D aschkow, D. (2019). Clus e -based pe mu a ion es s
o MEG/EEG da a do no es ablish signi icance o e ec la ency o
loca ion. Psychophysiology,56(6), e13335. h ps://doi.o g/10.1111/
psyp.13335
Schlaug, G., Ma china, S., & No on, A. (2009). E idence o plas ici y in
whi e-ma e ac s o pa ien s wi h ch onic B oca's aphasia unde go-
ing in ense in ona ion-based speech he apy. Annals o he New Yo k
Academy o Sciences,1169, 385–394. h ps://doi.o g/10.1111/j.1749-
6632.2009.04587.x
Shimizu, T., Hosaki, A., Hino, T., Sa o, M., Komo i, T., Hi ai, S., &
Rossini, P. M. (2002). Mo o co ical disinhibi ion in he una ec ed
hemisphe e a e unila e al co ical s oke. B ain,125(P 8),
1896–1907. h ps://doi.o g/10.1093/b ain/aw 183
Spi onelli, C., Man edi, M., & Ang illi, A. (2013). Be a EEG band: A mea-
su e o unc ional b ain damage and language eo ganiza ion in aphasic
pa ien s a e eco e y. Co ex,49(10), 2650–2660. h ps://doi.o g/
10.1016/j.co ex.2013.05.003
Taulu, S, & Simola, J. (2006). Spa io empo al signal space sepa a ion me hod
o ejec ing nea by in e e ence in MEG measu emen s. Physics in Med-
icine & Biology,51(7), 1759.
Thompson-Schill, S. L., D'Esposi o, M., & Kan, I. P. (1999). E ec s o epe i-
ion and compe i ion on ac i i y in le p e on al co ex du ing wo d
gene a ion. Neu on,23(3), 513–522. h ps://doi.o g/10.1016/s0896-
6273(00)80804-1
Tohka, J., Zijdenbos, A., & E ans, A. (2004). Fas and obus pa ame e es i-
ma ion o s a is ical pa ial olume models in b ain MRI. Neu oImage,
23(1), 84–97. h ps://doi.o g/10.1016/j.neu oimage.2004.05.007
T anel, D., Adolphs, R., Damasio, H., & Damasio, A. R. (2001). A neu al
basis o he e ie al o wo ds o ac ions. Cogni i e Neu opsychology,
18(7), 655–674. h ps://doi.o g/10.1080/02643290126377
T au , T., Sa desh, N., Bulubas, L., Findlay, A., Honma, S. M., Mizui i, D., …
Ta apo e, P. E. (2019). MEG imaging o ecu en gliomas e eals unc-
ional plas ici y o hemisphe ic language specializa ion. Human B ain
Mapping,40(4), 1082–1092. h ps://doi.o g/10.1002/hbm.24430
Uhlhaas, P. J., Liddle, P., Linden, D. E. J., Nob e, A. C., Singh, K. D., &
G oss, J. (2017). Magne oencephalog aphy as a ool in psychia ic
esea ch: Cu en s a us and pe spec i e. Biological Psychia y: Cogni-
i e Neu oscience and Neu oimaging,2(3), 235–244. h ps://doi.o g/10.
1016/j.bpsc.2017.01.005
Van Veen, B. D., an D ongelen, W., Yuch man, M., & Suzuki, A. (1997).
Localiza ion o b ain elec ical ac i i y ia linea ly cons ained mini-
mum a iance spa ial il e ing. IEEE T ansac ions on Biomedical Engi-
nee ing,44(9), 867–880. h ps://doi.o g/10.1109/10.623056
Vassal, M., Le Ba s, E., Mo i z-Gasse , S., Menjo , N., & Du au, H. (2010).
C ossed aphasia elici ed by in aope a i e co ical and subco ical
s imula ion in awake pa ien s. Jou nal o Neu osu ge y,113(6),
1251–1258.
Va ela, F., Lachaux, J. P., Rod iguez, E., & Ma ine ie, J. (2001). The b ainweb:
Phase synch oniza ion and la ge-scale in eg a ion. Na u e Re iews. Neu-
oscience,2(4), 229–239. h ps://doi.o g/10.1038/35067550
Vigliocco,G.,Vinson,D.P.,D uks,J.,Ba be ,H.,&Cappa,S.F.(2011).Nouns
and e bs in he b ain: A e iew o beha iou al, elec ophysiological, neu-
opsychological and imaging s udies. Neu oscience & Biobeha io al Re iews,
35(3), 407–426. h ps://doi.o g/10.1016/j.neubio e .2010.04.007
Vinson, D. P., Vigliocco, G., Cappa, S., & Si i, S. (2003). The b eakdown o
seman ic knowledge: Insigh s om a s a is ical model o meaning ep-
esen a ion. B ain and Language,86(3), 347–365. h ps://doi.o g/10.
1016/s0093-934x(03)00144-5
Vo we k, J., Cho, J. H., Rampp, S., Hame , H., Knosche, T. R., &
Wol e s, C. H. (2014). A guideline o head olume conduc o model-
ing in EEG and MEG. Neu oImage,100, 590–607. h ps://doi.o g/10.
1016/j.neu oimage.2014.06.040
Walke , D. G., & Kaye, A. H. (2003). Low g ade glial neoplasms. Jou nal o
Clinical Neu oscience,10(1), 1–13. h ps://doi.o g/10.1016/s0967-
5868(02)00261-8
Wa son, C. E., Ca dillo, E. R., Ianni, G. R., & Cha e jee, A. (2013). Ac ion
concep s in he b ain: An ac i a ion likelihood es ima ion me a-analy-
sis. Jou nal o Cogni i e Neu oscience,25(8), 1191–1205. h ps://doi.
o g/10.1162/jocn_a_00401
Weiss, S., & Muelle , H. M. (2012). “Too many be as do no spoil he
b o h”: The ole o be a b ain oscilla ions in language p ocessing. F on-
ie s in Psychology,3, 201. h ps://doi.o g/10.3389/ psyg.2012.00201
W igh , P., S ama akis, E. A., & Tyle , L. K. (2012). Di e en ia ing hemi-
sphe ic con ibu ions o syn ax and seman ics in pa ien s wi h le -
hemisphe e lesions. The Jou nal o Neu oscience,32(24), 8149–8157.
h ps://doi.o g/10.1523/JNEUROSCI.0485-12.2012
Zhang, N., Xia, M., Qiu, T., Wang, X., Lin, C. P., Guo, Q., …Zhou, L. (2018).
Reo ganiza ion o ce eb o-ce ebella ci cui in pa ien s wi h le hemi-
sphe ic gliomas in ol ing language ne wo k: A combined s uc u al
and es ing-s a e unc ional MRI s udy. Human B ain Mapping,39(12),
4802–4819. h ps://doi.o g/10.1002/hbm.24324
How o ci e his a icle: Amo uso L, Geng S, Molina o N, e al.
Oscilla o y and s uc u al signa u es o language plas ici y in
b ain umo pa ien s: A longi udinal s udy. Hum B ain Mapp.
2021;42:1777–1793. h ps://doi.o g/10.1002/hbm.25328
AMORUSO ET AL.1793
