ORIGINAL PAPER
Eu opean A chi es o Psychia y and Clinical Neu oscience
h ps://doi.o g/10.1007/s00406-024-01859-z
Gema Mijancos-Ma ínez
[email p o ec ed]
Alejand o Bachille
[email p o ec ed]
Inés Fe nández-Linsenba h
[email p o ec ed]
Se gio Rome o
[email p o ec ed]
Leidy Y. Se na
leidy[email p o ec ed]
Vicen e Molina
[email p o ec ed]
Miguel Ángel Mañanas
[email p o ec ed]
1 Biomedical Enginee ing Resea ch Cen e (CREB),
Depa men o Au oma ic Con ol (ESAII), Uni e si a
Poli ècnica de Ca alunya - Ba celonaTech (UPC), Ba celona,
Spain
2 Ins i u e o Resea ch San Joan de Déu, Ba celona, Spain
3 Psychia y Depa men , School o Medicine, Uni e si y o
Valladolid, Valladolid, Spain
4 CIBER o Bioenginee ing, Bioma e ials and Nanomedicine
(CIBER-BBN), Mad id, Spain
5 Psychia y Se ice, Clinical Hospi al o Valladolid,
Valladolid, Spain
6 Neu osciences Ins i u e o Cas illa y Léon (INCYL),
Uni e si y o Salamanca, Salamanca, Spain
Abs ac
T ansc anial magne ic s imula ion and elec oencephalog aphy (TMS-EEG) eco dings a e c ucial o di ec ly assess co i-
cal exci abili y and inhibi ion in a non-in asi e and ask- ee manne . TMS-EEG signals a e cha ac e ized by TMS-e oked
po en ials (TEPs), which a e employed o e alua e co ical unc ion. None heless, di e en ime windows (TW) ha e
been used o compu e hem o e he yea s. Mo eo e , hese TWs end o be he same o all pa icipan s omi ing he
in e subjec a iabili y. The e o e, he objec i e o his s udy is o assess he e ec o using di e en TWs o compu e he
TEPs, mo ing om a common ixed TW o mo e adap i e indi idualized TWs. Twen y-nine heal hy (HC) con ols and
wen y schizoph enia pa ien s (SCZ) unde wen single-pulse (SP) TMS-EEG p o ocol. Fi s ly, only he HC we e consid-
e ed o e alua e he TEPs o h ee di e en TWs in e ms o ampli ude and opog aphical dis ibu ion. Secondly, he SCZ
pa ien s we e included o de e mine which TW is be e o cha ac e ize he b ain al e a ions o SCZ. The esul s indica e
ha a mo e indi idualized TW p o ides a be e cha ac e iza ion o he SP TMS-EEG signals, al hough all o hem show
he same endency. Rega ding he compa ison be ween g oups, he indi idualized TW is he one ha p o ides a be e
di e en ia ion be ween popula ions. They also p o ide u he suppo o he possible imbalance o co ical exci abili y/
inhibi ion in he SCZ popula ion due o i s educed ac i i y in he N45 TEP and g ea e ampli ude alues in he N100.
Resul s also sugges ha he SCZ b ain has a baseline hype ac i e s a e since he TEPs o he SCZ appea ea lie han
hose o he HC.
Keywo ds TMS-EEG · Cha ac e iza ion · Indi idualiza ion · TEP · Time window · Schizoph enia
Recei ed: 15 Feb ua y 2024 / Accep ed: 20 June 2024
© The Au ho (s) 2024
Indi idualized ime windows enhance TMS-EEG signal cha ac e iza ion
and imp o e assessmen o co ical unc ion in schizoph enia
GemaMijancos-Ma ínez1,2 · Alejand oBachille 1,2· InésFe nández-Linsenba h3· Se gioRome o1,2,4·
Leidy Y.Se na1,4· Vicen eMolina3,5,6· Miguel ÁngelMañanas1,2,4
1 3
Eu opean A chi es o Psychia y and Clinical Neu oscience
In oduc ion
T ansc anial Magne ic S imula ion (TMS) is a no el, non-
in asi e echnique ha al e s he co ical neu onal exci -
abili y h ough a magne ic pulse [1]. I allows a be e
unde s anding o he co ical ci cui y [1, 2]. TMS s imula-
ion igge s cha ac e is ics e en ela ed po en ials known
as TMS-e oked po en ials (TEPs) ha can be eco ded
simul aneously using elec oencephalog aphy (EEG). In he
EEG signals, TEPs a e cha ac e ized by changes in ol age
ac oss ime sp eading h ough he scalp. They a e o igina ed
by he pos -synap ic po en ials o he py amidal neu ons and
in e neu ons s imula ed by he TMS-pulse; hus, hey a e
ime-locked o he TMS-pulse [3]. TEPs a e de ined ega d-
ing hei pola i y (nega i e o posi i e; N o P) and hei
la ency (in ms, ime a e TMS-pulse onse ) [4]. The e alu-
a ion o he TMS-EEG signals, ocusing on TEP ampli ude
and i s spa ial dis ibu ion, has been used by esea che s o
s udy di e en b ain diso de s, such as schizoph enia, bipo-
la and mood diso de s o subs ance use diso de s. In his
con ex , mos s udies ha e adi ionally a ge ed he mo o
co ex al hough he e is an inc easing end in esea ch
ha in ol es s imula ing he do sola e al p e on al co ex
(DLPFC) [3].
Schizoph enia is a ch onic psycho ic b ain diso de
a ec ing 24 million people wo ldwide, acco ding o he
Wo ld Heal h O ganiza ion [5]. This diso de is cha ac e -
ized by s uc u al and unc ional al e a ions in he b ain,
mani es ing in symp oms such as hallucina ions, delusions
and diso de ed speech o hough s. Pa ien s o en p esen
blun a ec and a ypical beha io s, including social wi h-
d awal [6, 7]. The e o e, esea ches ha e used TMS-EEG
da a o in es iga e his complex diso de . S udies in ol ing
schizoph enic (SCZ) popula ion ha e e ealed an al e a-
ion in he exci a ion-inhibi ion balance [3, 8]. Speci ically,
esea ch sugges s al e a ions in co ical inhibi ion mecha-
nisms, hough o be media ed by he GABAe gic neu ons in
he DLPFC [3, 9]. Neu opa hological and in i o e idences
suppo ha a co ical exci a o y/inhibi o y disequilib ium
plays a ole in he pa hophysiology o his diso de [10, 11].
Conc e ely, he mos consis en ly eplicable TEPs in cu en
esea ch, a ising om a single TMS-pulse a he DLFPC o
mo o co ex (M1) s imula ion, include P30, N45, P60, N100
and P180 [3, 4, 12, 13]. Among hese, he N100 de lec ion is
p obably he mos ep oducible o he TMS-e oked po en-
ials componen s [14, 15]. Since he N100 ampli ude o he
TMS-e oked po en ial has been epo ed o index he glu a-
ma e/GABA balance in i o [16], i is necessa y o imp o e
he me hods o compa ing his TEP be ween pa ien s and
con ols o inc ease ou unde s anding o he unde pinnings
o psycho ic diso de s.
Despi e an ex ensi e lis o cha ac e ized TEP epo ed
in schizoph enia li e a u e, he e has been a consis en a i-
abili y in he compu a ion o TEP ampli ude and opog a-
phy due o he use o he e ogeneous ime windows (TW).
A TW is de ined by he la ency and he window leng h con-
side ed o compu e he desi ed pa ame e s used o signal
cha ac e iza ion. While some esea che s ha e used a ixed
TW, employing a ime-cons an , common window o all
pa icipan s, ei he based on da a om he s udy o no [12,
17–20], o he s ha e employed a mo e indi idualized TW
cen e ed a ound he manually selec ed TMS-peak o each
pa icipan [21–23]. The di e se p ocedu es o c ea ing
he TWs, which will de ine he TEPs, in oduce a a iabil-
i y in signal cha ac e iza ion, he eby in luencing he inal
TEP analysis. Mo eo e , in cases o ixed TW o lack o
subjec indi idualiza ion, ce ain TEPs migh be missed and
excluded om he analysis.
Fo hese easons, we aimed o analyze he opog aphical
di e ences esul ing om he u iliza ion o di e en TWs
and hei impac when compa ing wo di e en popula ions
(HC and SCZ pa ien s). Fu he mo e, we hypo hesized ha
a be e TMS-EEG signal cha ac e iza ion is achie ed when
pe sonalized la encies a e used a he han using a gene -
alized common TW. In addi ion, we belie e ha a pe son-
alized TW has he po en ial o mo e p ecisely a ge he
desi ed TEP, consequen ly acili a ing he iden i ica ion o
mo e signi ican di e ences be ween popula ions.
In he cu en s udy, we employed single-pulse (SP)
TMS-EEG signals om he HC g oup o assess a ious TW
epo ed in he li e a u e. This in ol ed cha ac e izing he
TEPs in e ms o ampli ude, la ency and spa ial dis ibu ion
o de e mine i he di e en TWs yielded dis inc in o ma-
ion. Las ly, he TEPs o SCZ pa ien s and HC we e e alu-
a ed ega ding hei ampli ude, opog aphy and la encies o
he di e en TWs in o de o de e mine which TW be e
e lec he di e ences be ween he wo g oups.
Ma e ials and me hods
Pa icipan s
Twen y-nine igh -handed heal hy con ols (HC; 13 males,
27 ± 12 yea s) and wen y schizoph enia pa ien s (SCZ; 11
males, 35 ± 13 yea s) pa icipa ed in he s udy. The SCZ
g oup was composed by 11 i s episodes and 9 ch onic
pa ien s. Pa ien s had been diagnosed by a psychia is
acco ding o he c i e ia o he Diagnos ic and S a is ical
Manual o Men al Diso de s 5 h edi ion, conside ing hei
cu en men al s a e, clinical eco ds, and in o ma ion om
ela i es. Pa icipan s mee ing any o he ollowing exclu-
sion c i e ia we e excluded: (a) in elligence quo ien unde
1 3
Eu opean A chi es o Psychia y and Clinical Neu oscience
70; (b) his o y o p esen o pas subs ance dependence
(excluding ca eine and nico ine); (c) head auma wi h
loss o consciousness; (d) neu ological o men al diagnosis
besides schizoph enia ( o pa ien s); (e) any cu en neu-
ological o psychia ic diagnosis (con ols); ( ) ecei ing
addi ional ea men a ec ing cen al ne ous sys em; (g)
po en ially a isk o unde going TMS. P io he inclusion
in he s udy, pa icipan s p o ided in o med w i en consen
a e ull w i en in o ma ion. The s udy was endo sed by he
local e hics commi ee and adhe ed o he e hical s anda ds
in he Helsinki Decla a ion o 1975, as e ised in 2008.
T ansc anial magne ic s imula ion
Se en y- i e monophasic single TMS pulses we e adminis-
e ed o he DLPFC o each pa icipan using a igu e-o -8
coil (MCF-B70) and a MagP oX100 s imula o (MagVen-
u e, Denma k). The adminis a ion o he pulses was
semi- andomized, occu ing be ween 5 and 7 s o p e en
an icipa ion o he nex pulse. The speci ic s imula ion si e
was he midpoin o a line be ween he F3 and F5 elec odes,
wi h a 45º o a ion ela i e o he midline. This posi ioning
was chosen o i s accu a e es ima ion o he le DLPFC
and low in e -subjec a iabili y in he absence o neu o-
na iga ional equipmen , [24, 25]. Pa icipan s we e sea ed
com o ably, looking di ec ly ahead wi h hei eyes open
du ing he pulse adminis a ion. The in ensi y o he pulses
was se o 120% o he Res ing Mo o Th eshold (RMT),
de e mined o e he mo o co ical egion ollowing he
ela i e equency me hod [26].
EEG eco ding
Alongside wi h DLPFC s imula ion, EEG was acqui ed
using a 64-channel sys em ampli ie (B ain Vision [B ain
P oduc s GmbH]) ollowing he 10–10 in e na ional sys em
a a sampling equency o 25 kHz. Six y-one channels we e
used o eco ding b ain signals whe eas he emaining 3
we e alloca ed o e ical and ho izon al elec ooculog a-
phy and elec omyog aphy. The impedance o each elec-
ode was kep below 5 kΩ, and channels we e e e enced
o e Cz du ing acquisi ion.
TMS-EEG p e-p ocessing and p ocessing
The p e-p ocessing and p ocessing o TMS-EEG da a we e
pe o med using MATLAB (R2021b; The Ma hWo ks Inc.,
Na ick, MA, USA) and FieldT ip [27].
Fi s , he TMS-EEG da a was segmen ed in o 2-sec-
onds epochs cen e ed a ound he TMS-pulse onse . Due o
he i e ie able na u e o he samples o he TMS-pulse,
wi hin each epoch, samples o he pulse and a ound i ( om
− 1ms o 10ms) we e emo ed and cubic in e pola ed [28].
Then, he da a was e- e e enced o common a e age. Sec-
ondly, independen componen analysis (ICA) was applied
o educe a e ac s. Independen componen s we e manu-
ally and blindly selec ed by h ee di e en expe s, based
on ime- equency maps, ial-a e aged ampli ude, spa ial
dis ibu ion and ac i a ion maps [28–30]. Thi dly, bad chan-
nels in e pola ion and bad ial ejec ion we e au oma ically
pe o med. Finally, a baseline co ec ion was applied using
an in e al o 800 ms be o e he TMS-pulse onse , and he
da a was downsampled o 5 kHz and band-pass il e ed
be ween 0.5 Hz and 70 Hz.
A i ac - ee da a was ial a e aged o each subjec o
ob ain TMS-e oked po en ials (TEPs). I is no ewo hy ha
only signals om HC we e used o TW analysis, whe eas
da a om bo h popula ions was conside ed o he cha ac-
e iza ion o he TMS-EEG signal.
Time window (TW) analysis
Se e al TW p e iously employed in di e en s udies we e
analyzed [17, 18, 29, 31, 32]. The aim was o compa e hese
TWs conce ning he opog aphy o he HC popula ion o
unde s and hei impac on he cha ac e iza ion o he TMS-
EEG signals. The a iables used o desc ibe he TW we e
he wid h o he window’s leng h, and he la ency, as he
ime ins an a ound which he TW is cen e ed. Bo h a i-
ables a e measu ed in milliseconds (ms). Depending on he
combina ion o hese a iables, di e en TW can be de ined:
Fixed TW
Fou di e en TW a e de ined acco ding o he li e a u e
depending on he speci ic TEP unde s udy: N45 (35-60ms),
P70 (60-80ms), N100 (85-140ms) and P180 (150-230ms)
[17, 18, 31].
GMA TW
G and-mean a e age (GMA) TW we e based on he g and-
mean a e age signal o he HC popula ion. The GMA signal
was compu ed by i s a e aging all ials o each subjec
and hen pe o ming an a e age o all subjec s. Subse-
quen ly, he channel displaying he mos p ominen p ojec-
ion o he TEP in he GMA signal was iden i ied, and he
onse ime o he TEP was used as he cen e o he TW
(la ency). Finally, TWs o 10 ms wid h a ound he la ency
we e used o ea ly TEPs (N45 and P60), while 30 ms wid h
TWs we e c ea ed o la e TEPs (N100, P140, and P180)
[29, 32].
The same p ocedu e was eplica ed o he SCZ popula-
ion, conside ing he GMA signal o he SCZ signals.
1 3
Eu opean A chi es o Psychia y and Clinical Neu oscience
S a is ical analysis
To analyze he TW and assess whe he di e ences exis ed
be ween hem, a s a is ical opog aphy s udy was conduc ed
using he pai ed sample Wilcoxon singed- ank es . Simi-
la ly, a Wilcoxon ank sum es was used o pe o m a s a-
is ical opog aphy s udy be ween condi ions (HC and SCZ
pa ien s). The same es was pe o med o TEP ampli ude
and la ency.
Resul s
TW compa ison
The dis ibu ion o la encies o HC o each TEP is shown
in Fig. 1. GMA la encies o he ea ly TEPs (N45 and P60)
a e sho e han he ixed la encies p oposed in he li e a-
u e, whe eas he la e TEPs (N100 and P180) exhibi highe
la encies. The GMA la ency alues consis en ly all wi hin
he in e qua ile ange o 25–75% o he Pe sonalized TW
la encies, whe eas he la ency o he Fixed TW is ou side o
his ange in he N45 and P140. As o he la e TEP, i is
no commonly epo ed in he li e a u e, hence, he e is no
ixed la ency o his TEP in he s a e-o - he a .
An example o he di e en TW applied o 2 subjec s is
depic ed in Fig. 2. As obse ed, using da a-dependen TWs,
such as he GMA o he Pe sonalized TW, allows o na -
owing he wid h o he TW and being mo e p ecise by cen-
e ing he TW o each peak.
Rema kable di e ences eme ged among he h ee di -
e en opog aphical dis ibu ion o TW (Fig. 3). The i s
ow (Fig. 3a) illus a es he s a is ical di e ences be ween
he GMA and Fixed TWs. These TWs main ain consis ency
ac oss all subjec s, bu hey use di e en wid hs and la en-
cies. S a is ical di e ences a e iden i ied o all TEPs. In
pa icula , he N45 wa e p esen s mo e s a is ically signi i-
can elec odes han he o he TEPs, whe eas P180 shows
he ewes di e ences. The N45 wa e exhibi s al e a ions in
ampli ude ac oss nea ly he en i e ce eb al co ex. Changes
Pe sonalized TW
The hi d ype o TW assessed was he Pe sonalized TW.
Fo each subjec , he la ency o each TEP was de e mined
on he same channel ob ained in he GMA TW me hod.
Simila ly, he onse o he TEP o each subjec was used
as he cen e o he subjec ’s TW and he wid h TW was
adjus ed analogously o he GMA TW. The e o e, o he
Pe sonalized TW, each subjec has a di e en TW la ency.
None heless, he wid h emains he same o all pa icipan s,
co esponding o he wid h o he GMA TW. A summa y o
he 3 TW e alua ed is shown in Table 1.
Signal cha ac e iza ion
To cha ac e ize he esponse o he TMS-pulse, he opo-
g aphical dis ibu ion and he mean ampli ude o he TEPs
we e compu ed o each TW. In addi ion, o he Pe sonal-
ized TW, he la ency o he TEPs was de e mined. Conse-
quen ly, he esponse o he s imulus was e alua ed based on
he magni ude o he TEPs, hei dis ibu ion h oughou he
b ain and hei ime o appea ance.
Fig. 1 La encies o each TEP and ype o TW ( om op o bo om:
Fixed TW, GMA TW and Pe sonalized TW)
TEP N45 P60 N100 P140 P180
Fixed TW
La ency (ms) 47.5 70 112.5 - 190
Wid h (ms) 20 20 55 - 80
Range (ms) 35–60 60–80 85–140 150–230
GMA TW
La ency (ms) 43.6 63.4 118.4 136.8 214.8
Wid h (ms) 10 10 30 30 30
Pe sonalized TW
La ency (ms)a43.8 ± 2.9 64.2 ± 9.4 115.9 ± 11 149.7 ± 15.8 208 ± 25.2
Wid h (ms) 10 10 30 30 30
Table 1 Range, la ency and wid h
o he di e en ype o ime
windows o each TEP. No e ha
o he ixed TW he e is no P140
TEP
aIn he Pe sonalized TW, each
subjec has hei own la ency o
each TEP. The la ency shown in
he able a e he mean and he
s anda d de ia ion
1 3
Eu opean A chi es o Psychia y and Clinical Neu oscience
Fig. 3 S a is ical opog aphy o he compa ison o he TW o each
TEP. (a) GMA s. Fixed TWs; (b) Pe sonalized s. Fixed TWs and
(c) Pe sonalized TW s. GMA TW. Red colo s indica e ha he ampli-
ude on he i s TW is highe han he second g oup and blue colo
he opposi e. Ligh colo s deno e endency (p- alues be ween 0.1 and
0.05), medium in ensi y colo s indica e p- alue be ween 0.05 and 0.01
and da k colo s e e o p- alues below 0.01
Fig. 2 Time windows ma ked
o 2 con ols. Fo each TEP, he
channel ob ained in he GMA
me hod is depic ed. As he Fixed
TW does no e alua e he P140
TEP, i s ep esen a ion has one
channel less. (a) Fixed TW, (b)
GMA TW and (c) Pe sonalized
TW
1 3
Eu opean A chi es o Psychia y and Clinical Neu oscience
opog aphic di e ence be ween popula ions, and he s a is-
ically signi ican di e ences be ween popula ions o each
TW.
The spa ial dis ibu ion o HC and SCZ using a Fixed
TW closely esembles each o he o some TEPs (Fig. 4).
Di e ences a e obse able in N45, N100, and P180. In SCZ
pa ien s, N45 ampli ude is highe in he le - on al elec-
odes, al hough i does no each s a is ical signi icance. In
he case o N100, he SCZ g oup exhibi s a g ea e ol age
le el in pa ie al elec odes and a lowe le el in he on al
egion, bo h eaching s a is ical signi icance when com-
pa ed o he HC popula ion. Finally, he P180 wa e o SCZ
pa ien s displays an ampli ude inc ease, pa icula ly in cen-
al elec odes, and hese di e ences also each s a is ical
signi icance.
Simila ly, Fig. 5 illus a es he opog aphical dis ibu-
ion o he GMA TW. The spa ial dis ibu ion o N45 di -
e s be ween he wo g oups, wi h a mo e nega i e pa e n
(lowe ampli ude) in he igh on al a ea o he HC popu-
la ion, al hough only a ew elec odes ha e p- alues lowe
han 0.05. No signi ican di e ences a e obse ed be ween
HC and SCZ g oups in P60. Ne e heless, di e ences in
ampli ude alues become mo e p onounced o la e TEPs.
N100 p esen s s a is ically signi ican di e ences in almos
all b ain egions excep some cen al elec odes, wi h he
SCZ g oup ha ing highe ampli ude alues in he pa ie al
a ea and lowe alues in he on al a ea compa ed o he
o P60 each s a is ical signi icance in cen al elec odes,
whe eas o N100 and P140, di e ences a e obse ed in
he occipi al and igh pa ie al a eas. Finally, he elec odes
ha accoun o mos o he obse ed a iance in P180 a e
loca ed in he on al and cen al egions.
Rega ding he Pe sonalized TW, i exhibi s mo e elec-
odes wi h s a is ically signi ican di e ences compa ed
o he Fixed TW (Fig. 3b). N45 and P180 p esen he la g-
es di e ences, wi h: almos all elec odes yielding p- al-
ues < 0.01. Di e ences in P60 a e e iden in on al and
occipi al elec odes. Di e ences in N100 a e obse ed in
occipi al and igh pa ie al egions. The P140 wa e shows
s a is ically signi ican di e ences in ewe elec odes
loca ed a cen al and on al egions. On he o he hand,
when compa ing he Pe sonalized TW wi h he GMA TW
(Fig. 3c), he e a e ewe di e ences in he ea ly TEPs. S a-
is ically signi ican di e ences in N100 a e obse able only
be ween on al and occipi al elec odes. Con a ily, size
e ec s a e mo e p ominen o la e TEPs.
Signal cha ac e iza ion
Topog aphic dis ibu ion o he TEPs
Figu es 4, 5 and 6 ep esen he opog aphical dis ibu-
ion o each TEP o bo h popula ions (SCZ and HC), he
Fig. 4 Topog aphy o each TEP o he Fixed TW. (a) SCZ pa ien s, (b)
HC popula ion, (c) di e ence (SCZ-HC) and (d) s a is ical opog aphy
SCZ s. HC. In a), b) and c) he opoplo s exhibi he ampli ude ac oss
he ce eb al co ex. In d) he p- alue o each elec ode o he compa i-
son SCZ s. HC is depic ed, hose ma ked wi h ligh pu ple indica ed
endency (0.05 < p- alue < 0.1), medium pu ple indica es s a is ical
signi icance wi h a p- alue be ween 0.01 and 0.05 and da k pu ple
indica es p- alue < 0.01
1 3
Eu opean A chi es o Psychia y and Clinical Neu oscience
Fig. 6 Topog aphy o each TEP o he Pe sonalized TW. (a) SCZ
pa ien s, (b) HC popula ion, (c) di e ence (SCZ-HC) and (d) s a is i-
cal opog aphy SCZ s. HC. In a), b) and c) he opoplo s exhibi he
ampli ude ac oss he ce eb al co ex. In d) he p- alue o each elec-
ode o he compa ison SCZ s. HC is depic ed, hose ma ked wi h
ligh pu ple indica ed endency (0.05 < p- alue < 0.1), medium pu ple
indica es s a is ical signi icance wi h a p- alue be ween 0.01 and 0.05
and da k pu ple indica es p- alue < 0.01
Fig. 5 Topog aphy o each TEP o he GMA TW. (a) SCZ pa ien s, (b)
HC popula ion, (c) di e ence (SCZ-HC) and (d) s a is ical opog aphy
SCZ s. HC. In a), b) and c) he opoplo s exhibi he ampli ude ac oss
he ce eb al co ex. In d) he p- alue o each elec ode o he compa i-
son SCZ s. HC is depic ed, hose ma ked wi h ligh pu ple indica ed
endency (0.05 < p- alue < 0.1), medium pu ple indica es s a is ical
signi icance wi h a p- alue be ween 0.01 and 0.05 and da k pu ple
indica es p- alue < 0.01
1 3
Eu opean A chi es o Psychia y and Clinical Neu oscience
When examining he mean ampli ude o he TEPs o he
HC g oup ac oss he di e en TW, no s a is ically signi ican
di e ences a e obse ed be ween Fixed and GMA TWs, no
be ween GMA and Pe sonalized TWs. None heless, he
compa ison o he ampli udes be ween Fixed TW and Pe -
sonalized TW, he P60, N100, and P180 po en ials exhibi
p- alues below 0.01, and he p- alues o he emaining
TEPs ange be ween 0.05 and 0.08.
Conce ning he SCZ popula ion, he e is s a is ical sig-
ni icance in he ampli ude o he P140 when compa ing
Fixed and GMA TWs (p- alue = 0.002) and GMA and Pe -
sonalized TWs (p- alue = 0.0001). Fu he mo e, P180 also
exhibi s s a is ically signi ican di e ences in ampli ude
when e alua ing Fixed and GMA TWs (p- alue = 0.01).
Simila o he heal hy popula ion, he p- alues o P60,
N100, and P140 a e non-signi ican .
The la encies o each TEP o bo h g oups a e ep esen ed
in he iolin plo s o Fig. 8. Fo ea ly TEPs and N100, he
onse o he po en ials occu s ea lie in he SCZ popula ion
han in he heal hy g oup, wi h p- alues lowe han 0.01 o
N45 and P60, and less han 0.05 o N100. On he o he
hand, he onse o he la e TEPs occu s ea lie in he HC
han in he SCZ (p- alue < 0.01 o P140).
In summa y, he TW ha exhibi s ewe di e ences in he
dis ibu ion o he TEPs be ween popula ions is he Fixed
TW. Bo h, he GMA and Pe sonalized TWs display simila
scalp ol age dis ibu ion. N100 and P140 po en ials show
mo e p onounced di e ences ac oss popula ions o all h ee
heal hy g oup. Fo he po en ials P140 and P180 TEPs he
di e ences in ampli ude a e p ima ily in he cen al and
pa ie al egions, wi h la ge dispa i ies in he cen al a ea
whe e he SCZ g oup has g ea e ampli ude alues han he
HC popula ion.
Figu e 6 illus a es he spa ial dis ibu ion o he TEPs
o bo h popula ions, along wi h hei di e ences and s a-
is ical opog aphy, speci ically ocusing on he Pe sonal-
ized TW. In his case, SCZ pa ien s exhibi highe alues o
ampli ude o N45 in he igh - on al b ain a ea and lowe
ampli ude in he le -occipi al egion. Rega ding N100, s a-
is ically signi ican di e ences a e obse ed in on al and
cen o-pa ie al elec odes, wi h a g ea e ampli ude in he
on al elec odes o SCZ pa ien s and a smalle one in he
cen o-pa ie al b ain a ea. In he case o P140, he opog a-
phy o SCZ g oup exhibi s a highe ampli ude in he cen al
elec odes compa ed o he HC g oup, eaching s a is ically
signi icance (p < 0.01). Finally, he e a e ewe elec odes
wi h s a is ically signi ican di e ences be ween popula-
ions ega ding he P180 po en ial.
TEP ampli ude and la ency
As obse ed in Fig. 7, ampli udes alues a e g ea e when
he TEPs a e e alua ed in a mo e indi idualized TW. Fo he
GMA TW, he P140 wa e o SCZ pa ien s exhibi s nega-
i e alues, and no signi ican di e ences be ween popula-
ions a e iden i ied.
Fig. 7 TEP ampli udes o he h ee ypes o TW in he channel ob ained in he GMA TW me hodology (le o igh : Fixed, GMA and Pe sonalized
TWs) and bo h popula ion g oups (HC pu ple, SCZ o ange)
1 3
Eu opean A chi es o Psychia y and Clinical Neu oscience
ixed la encies o all subjec s (Fixed and GMA TW) ei he
missed TEPs in some subjec s o do no cen e he TEP
wi hin he window, esul ing in an inaccu a e e alua ion o
he po en ial. This is a ibu ed o he inhe en a iabili y in
he la encies o he TEPs, as i is illus a ed in Fig. 1. While
he la encies o N45, P60, N100, and P180 po en ials using
he GMA TW closely align wi h he median la ency o he
same TEPs when indi idually e alua ed, he e a e ins ances
whe e some TEPs a e missed. The P140 po en ial, being
less p onounced and dis inc , is mo e p one o smoo hing
du ing a e aging, leading o a poo ma ch be ween he
GMA la ency and he median la ency o he popula ion. In
con as , he Fixed TW is o en cen e ed dis an ly om he
median o he pe sonalized la encies, missing e en mo e
TEPs o some subjec s. Ne e heless, due o he la ge
wid h o he Fixed TW, his decen aliza ion o he TEP can
some imes be mi iga ed. The downside is ha he signal is
smoo hed, educing he compu ed ampli ude. Fu he mo e,
he Fixed TW o en encompasses wo TEPs ( ypically he
ea ly TEPs) in one single TW, which is in ended o be exclu-
si e o one TEP. O e all, his esul s in a subop imal selec-
ion o milliseconds o e alua ing he signal ( he TW used
o assess each TEP), leading o a loss o empo al esolu ion
o indi idual TEPs.
The spa ial dis ibu ion o he TEPs is also a ec ed by
he TW used, as shown in Fig. 3. The N45 po en ial, along
wi h P180, exhibi s b oade di e ences among TWs due o
i s sho du a ion. Despi e i s sho du a ion, he spa ial dis-
ibu ion o N45 shows some simila i y be ween GMA and
Pe sonalized TW, as he la ency o N45 in GMA is close o
he median popula ion. Addi ionally, he wid h is smalle in
he GMA TW compa ed o he Fixed TW. The same pa -
e n is obse ed wi h P180 and P60 (median la ency close
o he GMA TW la ency). Howe e , he scalp dis ibu ion o
he Pe sonalized and he GMA TWs is mo e dis inc in he
TW. Conce ning ampli udes, hei alues inc ease wi h he
pe sonaliza ion o he TW. Las ly, ea ly TEPs appea ea lie
in he SCZ popula ion han in he HC popula ion, whe eas
he opposi e occu s o la e TEPs.
Discussion
The ou comes o his s udy unde sco e he signi ican in lu-
ence o bo h he la ency and he wid h o he ime window
employed cha ac e izing he TMS-e oked po en ials, as
hese ac o s wield a subs an ial impac on he inal esul s.
Mo eo e , he use o an indi idualized TW o each subjec
enhances he cha ac e iza ion o he single-pulse TMS-EEG
signal and maximizes he disce nible di e ences be ween
popula ions. Al hough he mos p onounced di e ences a e
obse ed o an indi idualized TW, in all h ee TW exhibi
consis en ends in di e ences among popula ions, pa icu-
la ly in he same b ain egions. The e o e, a non-da a depen-
den TW may o e a comp ehensi e o e iew o ongoing
e en s, while da a dependen TWs (i.e. GMA and Pe sonal-
ized TWs) p o ide mo e p ecise opog aphical localiza ion.
The obse ed di e ences be ween SCZ and HC popula-
ions o e addi ional suppo o an imbalance in co ical
exci abili y/inhibi ion wi hin he SCZ g oup. Fu he mo e,
he indings sugges ha he schizoph enic b ain may exhibi
a baseline hype ac i e s a e, as indica ed by he ea lie onse
o ea ly TEPs in he SCZ popula ion.
TW compa ison o HC popula ion
The choice o di e en TW signi ican ly in luences he
cha ac e iza ion o he single-pulse TMS-EEG, leading o
di e ences in pa ame e s de ining he TEP, such as ampli-
ude and scalp dis ibu ion. As shown in Fig. 2, TWs wi h
Fig. 8 Violin plo s o he la encies o HC (pu ple) and SCZ (o ange) o each TEP in he Pe sonalized TW. The TEP wi h he as e isk ha e a s a is i-
cally signi ican p- alue (p < 0.05) when e alua ed be ween popula ions
1 3
