ARTICLE DE RECHERCHE - Recherche en IA spécialisée au MIT

ARTICLE DE RECHERCHE

Reduced Interference and Serial Dependency
Effects for Naming in Older but Not Younger Adults
after 1 Hz rTMS of Right Pars Triangularis

un accès ouvert

journal

Kaundinya S. Gopinath4

, and Bruce Crosson1,2,5

Jonathan H. Drucker1,2,3

, Charles M. Epstein2

, Keith M. McGregor1,2

, Kyle Hortman1,

1VA Rehabilitation Research & Development Center for Visual and Neurocognitive Rehabilitation, Decatur, GA, Etats-Unis
2Department of Neurology, Emory University, Atlanta, GA, Etats-Unis
3Aptima, Inc., Woburn, MA, Etats-Unis
4Department of Radiology, Emory University, Atlanta, GA, Etats-Unis
5Département de psychologie, Georgia State University, Atlanta, GA, Etats-Unis

Mots clés: aging, langue, inhibition, neuromodulation, stimulation magnétique transcrânienne répétitive,
Broca’s area

ABSTRAIT

1 Hz repetitive transcranial magnetic stimulation (SMTr) was used to decrease excitability of
right pars triangularis (R PTr) to determine whether increased R PTr activity during picture
naming in older adults hampers word finding. We hypothesized that decreasing R PTr
excitability would reduce interference with word finding, facilitating faster picture naming.
15 older and 16 younger adults received two rTMS sessions. In one, speech onset latencies for
picture naming were measured after both sham and active R PTr stimulation. In the other
session, sham and active stimulation of a control region, right pars opercularis (R POp), étaient
administered before picture naming. Order of active vs. sham stimulation within session was
counterbalanced. Younger adults showed no significant effects of stimulation. In older adults, un
trend indicated that participants named pictures more quickly after active than sham R PTr
stimulation. Cependant, older adults also showed longer responses during R PTr than R POp
sham stimulation. When order of active vs. sham stimulation was modeled, older adults
receiving active stimulation first had significantly faster responding after active than sham
R PTr stimulation and significantly faster responding after R PTr than R POp stimulation,
consistent with experimental hypotheses. Cependant, older adults receiving sham stimulation
first showed no significant differences between conditions. Findings are best understood,
based on previous studies, when the interaction between the excitatory effects of picture
naming and the inhibitory effects of 1 Hz rTMS on R PTr is considered. Implications regarding
right frontal activity in older adults and for design of future experiments are discussed.

INTRODUCTION

While the left hemisphere is normally dominant for language, older adults show right (R.) fron-
tal activity during language production that is not present in younger adults. It has been sug-
gested that activity during language production in aging plays a compensatory role, assisting a
declining left hemisphere with language tasks such as word finding (par exemple., Cabeza, 2001, 2002;
Cabeza et al., 2004; Dolcos et al., 2002). Increased R frontal activity has been shown for older
relative to younger adults both for picture naming (Berlingeri et al., 2013; Fridriksson et al.,

Citation: Drucker, J.. H., Epstein, C. M.,
McGregor, K. M., Hortman, K.,
Gopinath, K. S., & Crosson, B. (2022).
Reduced interference and serial
dependency effects for naming in older
but not younger adults after 1 HzrTMS
of right pars triangularis. Neurobiology
of Language, 3(2), 256–271. https://est ce que je
.org/10.1162/nol_a_00063

EST CE QUE JE:
https://doi.org/10.1162/nol_a_00063

Reçu: 15 Décembre 2020
Accepté: 25 Novembre 2021

Intérêts concurrents: Les auteurs ont
declared that competing interests exist.

Auteur correspondant:
Jonathan H. Drucker
jondrucker86@gmail.com

Éditeur de manipulation:
Steven Small

droits d'auteur: © 2021
Massachusetts Institute of Technology
Publié sous Creative Commons
Attribution 4.0 International
(CC PAR 4.0) Licence

La presse du MIT

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

Pars triangularis (PTr):
A structure in the anterior portion of
inferior frontal gyrus.

Repetitive transcranial magnetic
stimulation (SMTr):
In rTMS, repeated pulses of magnetic
energy penetrate the skull, inducing a
current that stimulates underlying
cortex.

Transcranial direct current
stimulation (tDCS):
In tDCS, direct current stimulation is
applied to the brain through
electrodes placed on the scalp.

2006; Hoyau et al., 2017; Wierenga et al., 2008) and for category-member generation
(Meinzer et al., 2009, 2012; Persson et al., 2004).

Usually this increased R frontal activity for older adults is located in R pars triangularis (PTr),
the anterior portion of the R hemisphere homologue for Broca’s area, though not always exclu-
sively so. Wierenga et al. (2008) found that for poorer performing older adults, picture naming
was negatively correlated with blood oxygen level-dependent (AUDACIEUX) activity in R PTr. More
importantly, Meinzer et al. (2009, 2012) found for all older adults that accuracy of category-
member generation was negatively correlated with R PTr BOLD activity during this task. Dans
autres mots, the greater the R PTr activity, the poorer word-finding performance was. These
findings raise the possibility that R PTr activity in older adults interferes with word finding.
En effet, Nocera et al. (2017) found that after a three-month aerobic exercise regimen, lower
post-exercise intervention activity in R inferior frontal cortex during category-member genera-
tion was associated with increased category-member generation accuracy and increased effi-
ciency in oxygen utilization from pre- to post-exercise intervention. The association of lower R
frontal activity post intervention with an increased accuracy in category member generation
across the intervention is highly suggestive that R frontal activity impedes word finding in
older adults.

A more direct way to determine if R PTr activity is interfering with word finding, cependant,
is to use non-invasive brain stimulation (NIBS) and observe the effects on a word-finding task.
Repetitive transcranial magnetic stimulation (SMTr) and transcranial direct current stimulation
(tDCS) are common forms of NIBS. While rTMS has advantages, tDCS has been used more
commonly to study frontal language functions in older adults and clinical populations
(Fridriksson et al., 2011). In tDCS, a constant, low amplitude current (1–2 mA) is passed
between an anode and a cathode. Generally, the anode is thought to increase cortical excit-
ability beneath it while the cathode is thought to decrease cortical excitability.

Holland et al. (2011) found that anodal tDCS of left (L) inferior frontal cortex during picture
naming, with the cathode over R frontopolar cortex, decreased verbal reaction times com-
pared to sham tDCS for older adults. There were no younger controls in this study. Meinzer
et autres. (2013) found that anodal tDCS of the L inferior frontal gyrus (IFG), with the cathode over
R supraorbital cortex, improved accuracy of category-member generation in older adults com-
pared to sham tDCS. For bilateral stimulation of the IFG (anode over L IFG, cathode over
R IFG), Lifshitz Ben-Basat and Mashal (2017) found faster reaction times during active than sham
tDCS in older participants, but only in the sham-first group and not in the active-first group
(active and sham conditions were separated by at least two or three days). There was no young
control group.

From the standpoint of our question about R PTr, there are potential limitations to tDCS
études. Two studies (Holland et al. 2011; Meinzer et al. 2013) investigated only L inferior
frontal anodal stimulation, and one study (Lifshitz Ben-Basat & Mashal, 2017) used anodal
stimulation in L and cathodal in R IFG. Ainsi, none of the three studies performed isolated
cathodal stimulation of the R IFG to decrease its excitability and determine effects on word
finding. Plus loin, two of the studies (Holland et al., 2011; Lifshitz Ben-Basat & Mashal, 2017)
did not include a young control group to which the results of the findings for older participants
could be referenced. Enfin, given the tDCS electrodes and their configuration, isolation of PTr
stimulation from stimulation of other IFG components would have been extraordinarily diffi-
cult, even if attempts to isolate R IFG stimulation had been made (Dmochowski et al., 2011).
Ainsi, the question of whether increased R PTr activity interferes with word finding remains an
open question.

Neurobiology of Language

257

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

Pars opercularis:
A structure occupying the posterior
portion of the inferior frontal gyrus.

1 HzrTMS:
When rTMS pulses are delivered at a
1/second rate, excitability of cortex is
decreased, making it more difficult to
activate.

The current study was designed to address this issue. We elected to use 1 Hz repetitive
transcranial magnetic stimulation (SMTr), which decreases cortical excitability, to deal with
some of the limitations to tDCS studies. rTMS induces rapid current changes that are limited
to a relatively small and well-defined brain region; c'est à dire., it has greater spatial specificity than
tDCS. When paired with neuronavigation guided by structural magnetic resonance images,
rTMS affords greater precision in stimulation of target cortical regions. Our study also included
a young control group to which we could reference findings from older adults. Plus loin, based
on previous work in aphasic patients and neurotypical adults (Naeser et al., 2005, 2011; Ren
et coll., 2014), we included a control region, R pars opercularis (POp), for R PTr and performed
active and sham rTMS on both R PTr and R POp. Since R POp is just posterior to R PTr in the
IFG, even high-definition tDCS (c'est à dire., using multiple channels on the scalp to selectively target
specific brain regions) could not provide enough spatial resolution to separate stimulation of
PTr from that of POp. We hypothesized that active 1 Hz rTMS of R PTr would reduce reaction
times for picture naming in older adults relative to sham stimulation of R PTr and relative to
active stimulation of R POp, because R PTr activity interferes with the word-finding functions
of L PTr, and that reducing the excitability of R PTr will reduce its activity, thereby reducing its
interference. Autrement dit, according to this hypothesis, this effect would be achieved
because the decreased excitability of R PTr would reduce its interference with picture naming.
Because young adults do not show R PTr activity during word finding (Meinzer et al., 2009,
2012; Wierenga et al., 2008), we did not expect to see changes in picture-naming reaction
times after 1 Hz rTMS in young adults.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

MATERIALS AND METHODS

Participants

Participants were 15 older adults (8 female) who were 65 à 79 years of age (mean = 70.67 années,
SD = 4.91) with at least a high school education (12–19 years, mean = 15.71 années, SD = 2.05) et
16 younger adults (8 female) who were 20 à 32 years of age (mean = 25.25, SD = 3.21) with at
least a high school education (16–21 years, mean = 16.79 années, SD = 1.58). (Due to a recording
error, years of education were not available for one older and two younger adults.) Participants
were recruited through flyers posted on the Emory University campus, in local libraries, et en
retirement communities, or by advertisements in local news papers. All participants had
English as a first language and were right-handed. The Montreal Cognitive Assessment (MoCA;
Nasreddine et al., 2005) was used as a cognitive screening tool. A cut-off score of 24 or higher
was used for inclusion, consistent with norms for the region in which the study took
place (Luis et al., 2009). Plus loin, participants had to be free from risks for magnetic resonance
imaging (IRM) scanning (par exemple., cardiac or other pacemakers, ferromagnetic implants not
anchored to bone, significant claustrophobia) and chronic conditions that could affect cognitive
les fonctions (par exemple., traumatic brain injury, epilepsy or family history of epilepsy, Parkinson’s disease,
Alzheimer’s disease, stroke, heart failure, kidney failure, malaria). Subjects could not be on anti-
seizure medications or other medications that might reduce responsiveness to rTMS, or on med-
ications that might reduce seizure thresholds (par exemple., bupropion, varenicline, chlorpromazine,
theophylline). Because the IFG lies below the temporalis muscle, which could be stimulated
by TMS pulses, persons with temporomandibular joint disorder were excluded from participa-
tion. Because exercise has been shown to affect language functions and brain activity in older
adultes (Zlatar et al., 2013; Nocera et al., 2015, 2017, 2020), and we wanted to measure the
effects of age without the confound of exercise effects, persons who regularly performed mod-
erate to high levels of exercise for at least 45 min on at least three days per week were excluded.
Participants gave written informed consent in accordance with procedures established by the

Neurobiology of Language

258

1 Hz rTMS of right pars triangularis prior to naming in older adults

Emory University/Atlanta VA Medical Center Institutional Review Board, consistent with the
Declaration of Helsinki. They were paid $50 per session for each of three sessions.

Procedures

Screening and cognitive testing

Respondents to advertisements and flyers underwent a brief telephone screening to ensure that
they met criteria for the study. Prior to MRI scanning and after informed consent procedures
during the first session, subjects participated in a brief cognitive assessment. As noted above,
the MoCA (Nasreddine et al., 2005) was administered as a cognitive screening tool. Potential
participants achieving a score of 24 points or higher proceeded to MRI scanning and subse-
quently were scheduled for their two rTMS sessions. To further characterize our sample, le
California Verbal Learning Test (Delis et al., 2000) and the Boston Naming Test (Kaplan et al.,
2001) were administered prior to scanning.

Magnetic resonance imaging

After consenting and cognitive testing, the younger and older adults participated in an MRI
scanning session. A T1-weighted (3 dimensional MP-RAGE) structural MRI was acquired to
assist in image-guided rTMS (TR = 2,300 ms TE = 2.89 ms, flip angle = 8 degrees, spatial res-
olution = 1 × 1 × 1 mm, matrix = 256 × 256, 176 sagittal slices). Sites for rTMS were selected
as follows. Anatomic landmarks were derived from the location of R PTr activity differences
between old and young participants during a previous picture-naming study (Wierenga et al.,
2008), providing optimal separation of stimulation between the two areas (PTr, POp) (Chiffre 1).
Spécifiquement, for R PTr a line was drawn from the intersection of the anterior horizontal ramus
(AHR) and the anterior ascending ramus (AAR) of the Sylvian fissure to the inferior frontal sul-
cus (IFS) at roughly a 45 degree angle to the AHR. The stimulation site was about half way up
this line from the vertex. For R POp, the stimulation site was in the superior portion of the area

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Chiffre 1. Stimulation sites for rTMS in R IFG. “1” represents a typical R PTr stimulation site. “2”
represents an example of a R POp site.

Neurobiology of Language

259

1 Hz rTMS of right pars triangularis prior to naming in older adults

(posterior to the AAR and near the IFS), in most instances providing at a least 1.5 cm separation
between stimulation sites.

rTMS and picture-naming sessions

Experimental procedures were modeled after those of Naeser et al. (2011), a study of low fre-
quency (1 Hz) rTMS of PTr, POp, and control regions in aphasic persons and cognitively neu-
rotypical adults. For Naeser et al., two or fewer stimulations occurred within a single session,
with at least 30 min between stimulations. In the current study, younger and older adults partic-
ipated in two rTMS sessions at least a day apart: one session for active and sham PTr stimulation
and one session for active and sham POp stimulation. The order of PTr and POp sessions was
counterbalanced across participants within each group. Across participants, the order of active
and sham stimulation within sessions also was counterbalanced, but each participant had the
same order for active vs. sham rTMS for both sessions. At least 30 min elapsed between active
and sham stimulation for each area.

One Hz rTMS was delivered via a MagVenture MagPro X100 stimulator and a Magventure
Cool-B65 A/P liquid-cooled figure of eight coil. Participants received 600 pulses over 10 min
(1 Hz) à 90% of resting motor threshold. R PTr and R POp were localized as described above
using a Rogue Research Brainsight 2 neuronavigation image guidance system. Two electrodes
for sham stimulation were placed prior to rTMS sessions approximately 1 cm below the
hairline and 2 cm apart on the right side of the forehead. For active stimulation, the coil
was oriented with the junction vertical (handle pointing downward) to limit stimulation pri-
marily to the target structure, with the stimulating side of the coil held directly over the scalp.
Position of the coil relative to the target was continuously monitored using the Brainsight neu-
ronavigation system so that real-time adjustments could be made by the operator to keep the
coil over the target site in R PTr or R POp. For sham stimulation, the inert side of the coil was
held over the scalp near the target location. The rTMS protocol (1 Hz stimulation for 10 min)
was run so that the audible clicks of the magnetic pulses were present during both active and
sham stimulation, but stimulation was delivered to the brain only during active rTMS. Electri-
cal stimulation on the forehead was used during active and sham stimulation to enhance the
similarity of sensations between the conditions. At the start of both active and sham sessions,
electrical pulses were delivered through the surface electrodes in synchrony with the audible
clicks of the coil and were adjusted to low levels so that the stimulation was not painful to
the participants.

Naming task

Target names for high imageability items were selected from the University of Western
Australia Psycholinguistic Database (https://websites.psychology.uwa.edu.au/school
/MRCDatabase/mrc2.html). Four lists of 30 items were composed of 15 medium frequency
(4–20 occurrences/million) et 15 low frequency (less than 4 occurrences per million) words,
and each list contained similar numbers of living and nonliving items. Lists also were counter-
balanced such that there were no significant differences between lists for frequency of words
in the English language, number of letters, number of syllables (1–4), familiarity, imageability,
or concreteness. Four sets of 30 colored pictures corresponding to the target words from the
four lists were selected from freely licensed internet databases. Immediately after each of the
four rTMS conditions (PTr active, PTr sham, POp active, POp sham), participants sat in front of
a Dell Latitude E6420 laptop computer on which one of the four picture sets was presented. UN
different picture set was presented for each of the conditions. Picture sets were counterba-
lanced across the conditions such that each picture set appeared an equal number of times

Neurobiology of Language

260

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

in each condition. To avoid participants getting into a rhythm that could bias response laten-
cies, presentation was paced by the experimenter, and a 500 ms pre-stimulus fixation pre-
ceded every trial. A new picture was presented after response to the previous picture was
completed. Vocal responses were recorded for off-line scoring. The first word uttered in
response to a picture was scored for correctness; non-word utterances were ignored for scoring
purposes. Plausible alternative responses to the target picture names were allowed. Only cor-
rect responses were used in the reaction time analysis; responses more than 3 SDs greater than
an individual participant’s mean reaction time were excluded from analysis. Once a correct
response was identified, the speech onset latency (SOL; c'est à dire., reaction time) for the response
was calculated using an automated MATLAB script written specifically for the current study,
removing any bias from derivation of verbal reaction times.

SOLs were analyzed using a linear mixed effects model that controlled for systematic var-
iance related to the specific items, thereby reducing error variance. The effects of interest were
âge (young vs. vieux, a between-subjects effect), site of stimulation (PTr vs. POp, a within-
subjects effect), kind of stimulation (active vs. sham, a within-subjects effect), and order of
stimulation (active first vs. sham first, a between-subjects effect). Since order effects were
not a part of experimental hypotheses, main experimental hypotheses were first assessed with-
out respect to order. As noted in the Introduction, our a priori hypothesis was that active stim-
ulation of R PTr would reduce reaction times for picture naming in older adults relative to
sham stimulation by decreasing its excitability, with no corresponding effect in younger par-
ticipants. A further a priori hypothesis was that older participants would name pictures more
quickly after active PTr than active POp stimulation, with no corresponding effect in younger
participants. Subsequently, post hoc analyses assessed if effects for order of stimulation within
sessions (active first vs. sham first) interacted with the effects of age (younger vs. older) and/or
type of rTMS (active vs sham).

RÉSULTATS

A Priori Hypotheses

A priori hypotheses were tested using the structure of a 2 cible (PTr vs. Pop) × 2 stimulation
(active vs. sham) Analysis of Variance (ANOVA), removing variance in SOLs related to the spe-
cific target words from error variance using a linear mixed model. The first a priori hypothesis,
derived from previous functional MRI studies (Berlingeri et al., 2013; Meinzer et al., 2009,
2012; Wierenga et al., 2008), was that for R PTr, active low-frequency rTMS would lead to
faster naming than sham stimulation in older but not younger adults, with no such effects
for our control region (R POp). Chiffre 2 shows mean SOLs for active vs. sham stimulation
in each of the target brain areas (R PTr vs. R POp) for older vs. younger adults. For PTr in older
participants, the mean difference for active vs. sham of −135 ms was marginally significant
(p = 0.0664). The corresponding mean difference of 79 ms for R PTr in younger adults was
not significant (p = 0.2648). For older participants, the effect for active vs. sham stimulation
of R POp (69 ms) was not significant (p = 0.1725), and the effect for R POp in younger
adultes (29 ms) was not significant (p = 0.5518).

Our other hypothesis was that active stimulation of R PTr would yield faster reaction times
than stimulation of the control region (R POp) in older participants. This hypothesis was not
confirmed (R POp minus R PTr = 59 ms, p = 0.3690), with similar results in younger partici-
pants (R POp minus R PTr = 54 ms, p = 0.3394). Enfin, inspection of Figure 2 indicated a
large difference between the sham conditions for PTr vs. POp in the older adults. While this
comparison was not planned, it might have explanatory value for our other findings, and this

Neurobiology of Language

261

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

Speech onset latencies (SOLs) for younger and older participants after 1 Hz active or
Chiffre 2.
sham rTMS of pars opercularis (POp) and of pars triangularis (PTr). The lighter gray bars represent
sham stimulation, and the darker gray bars represent active rTMS. For older participants (droite),
there was a marginal difference between active vs. sham stimulation for PTr (sham SOL minus active
SOL = 135 ms, p = 0.0664). Cependant, the sham condition for PTr was slower than that for POp
(PTr minus POp = 145 ms, p = 0.0143). There were no significance differences for the younger
group (gauche).

analysis indicated the difference was substantial (R PTr sham minus R POp sham = 145 ms,
p = 0.0143).

A perplexing aspect of our data was the difference in sham conditions for R PTr vs. R POp
stimulation. By definition, the sham conditions had no active stimulation of the cortex and thus
should not produce a measurable behavioral effect. The fact that sham stimulation of R PTr pro-
duced significantly longer SOLs than sham stimulation of R POp indicates that something was
affecting the cortex in one sham condition that was not affecting it in the other. Two facets of our
data led us to ask whether within-session order effects for active vs. sham stimulation in our study
might account for the difference in SOLs between the R PTr and the R POp sham conditions.

D'abord, while our within-session time span between active and sham rTMS stimulation (30 min)
was similar to other rTMS studies of language (par exemple., Naeser et al., 2011) at the time our study
was planned, reports of order effects in older participants, even with days between sessions,
began to emerge for tDCS (par exemple., Lifshitz Ben-Basat & Mashal, 2017) and rTMS (par exemple., James
et coll., 2017). En effet, in the motor system, Siebner et al. (2004) had previously shown that
the effects of tDCS can last at least 35–45 min. Ainsi, when active stimulation is done first
with sham 30 min later, the effects of active stimulation may still be present when sham stim-
ulation is given that would not be present when sham stimulation was done first.

The second consideration indicating that the order effects were worth examining was that
based on the findings of Wierenga et al. (2008), our picture-naming tasks involved repetitive
excitation (30 events) of R PTr over a few minutes. Lang et al. (2004) showed that precondi-
tioning weak excitatory stimulation with stimulation that reduces cortical excitability tends to
enhance the effects of a weak excitatory stimulus 20–30 min after the latter, and Siebner et al.
(2004) showed that preconditioning stimulation that is normally inhibitory evokes a weak
excitatory stimulation that, à son tour, lessens effects of the inhibitory stimulation 25–35 min after
preconditioning. Ainsi, these studies raise the possibility that our picture-naming tasks were
interacting with our 1 Hz (inhibitory) r TMS. As we discuss in detail later (in the Discussion
section), these sequences in Lang’s and Seibner’s studies have analogues in our study. For pur-
poses of studying order effects, the important point here is that the interaction of 1 Hz rTMS of

Neurobiology of Language

262

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

R PTr with the excitatory effects on R PTr of picture naming might produce order effects in
older adults affecting results of the stimulation condition (active or sham) that was adminis-
tered second within sessions. Autrement dit, there were no experimental events that could
precondition the stimulation condition (active or sham) that was administered first, mais le
picture naming or active rTMS administered during the first stimulation condition could pre-
condition the events that followed it to affect picture-naming SOLs in the second condition.
Ainsi, taking these possibilities into consideration, within session order was added to the
model to determine if it could further explain our findings.

Adding Order of Stimulation to the Analysis Model

Ainsi, we repeated the above linear mixed model ANOVA adding order of stimulation within
session (active first vs. sham first) as a third independent variable. Since significant effects for
the above a priori comparisons were found for the older but not the younger group, nous
focused these analyses on the older group, but also performed a parallel analysis in the youn-
ger group to determine whether order effects could have contributed to any lack of findings in
the younger subjects. For the older group, there was a significant 3-way interaction (p =
0.0426) of Cortical Target (PTr vs. POp) × Stimulation (Active vs. Sham) × Order within Ses-
sions (active-first vs. sham-first). Ainsi, we next determined whether there were Target × Stim-
ulation interactions at the different levels of order within sessions. For the active-first group, le
Target × Stimulation interaction was significant (p = 0.0018), but for the sham-first group, ce
interaction was not significant (p = 0.7962). Given this finding, we next looked at pairwise
comparisons representing stimulation effects at the different targets (R PTr vs. R POp), aussi
as the target effects at the different types of stimulation, for the active-first older group.

For the older participants, the active-first effects (Figure 3A) are an amplified version of
the results when order of stimulation was not modeled, specifically for the R PTr conditions:

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

(UN) Speech onset latency (SOL) reactions times for the active-first (gauche) and sham-first (droite) conditions for the older adults. Light
Chiffre 3.
gray bars represent sham stimulation, and dark gray bars represent the active stimulation. For the active-first group, participants named pictures
significantly faster for active than sham rTMS in PTr but not in POp. Older participants also named pictures faster after active PTr stimulation
than after active POp stimulation. Cependant, they also named pictures faster after sham POp stimulation than after sham PTr stimulation. Le
unexpected difference between sham conditions is explained in the Discussion section. (B) SOLs for the active-first (gauche) and sham-first (droite)
conditions for younger adults. Younger participants named pictures marginally faster after active PTr stimulation than after the active POp
condition, but no other comparisons approached significance.

Neurobiology of Language

263

1 Hz rTMS of right pars triangularis prior to naming in older adults

Tableau 1. Number of participants with different error rates per list for younger and older participants

Groupe
Younger

Older

Error rates per 30-word list

0 Errors / List
10

0.25 Errors / List
4

0.5 Errors / List
1

≥0.5 Errors / List
1

The R PTr active minus sham difference (−291 ms) was significant (p = 0.005), consistent
with our a priori hypothesis. Plus loin, the R PTr active minus the R POp active difference of
−183 ms was significant for active-first participants (p = 0.038), which also is consistent
with our a priori hypothesis. Le 226 ms difference between the sham conditions (R PTr
sham vs. R POp sham) also was significant (p = 0.008). Le 26 ms difference for R POp active
minus sham was not significant (p = 0.708). There were no significant interactions for the
younger participants.

Accuracy

Average accuracy of picture-naming performance was 99.54% correct for younger participants
(average error percentage = 0.47%, SD = 0.74%) et 98.10% correct for older participants
(average error percentage = 1.90%, SD = 3.26%). Twelve of the 15 older subjects had error
rates within the distribution limits of the younger participants (0–3 errors for all four of the
word lists, 120 items). The other three older participants had 4, 6, et 15 errors for all four
of the word lists, accounting for 73.53% of the total errors for the older group. Tableau 1 shows
the error rates for the younger and older participants per 30-word list. Given the high accuracy
rates for both groups (all but 1 younger and 3 older participants averaged ≤0.5 errors per list),
it was not surprising that further analysis of errors with respect to stimulation conditions and
sites of stimulation did not yield significant differences characterizing the performance of
either group.

DISCUSSION

The main hypothesis for this study was that decreasing excitability of R PTr in older adults
would improve the efficiency of picture naming because this R PTr activity interferes with
word finding. If suppression of R PTr activity via 1 Hz rTMS in the older group resulted in faster
picture naming, it would provide clear evidence that R PTr activity had been interfering with
word retrieval, as opposed to compensating for diminished function. Our study was designed
to enable us to test this hypothesis against two control conditions: (1) sham (as opposed to
active) stimulation, et (2) a control target in R POp, a brain region that previous literature
(Naeser et al., 2011; Wierenga et al., 2008) indicated would not respond to 1 HzrTMS. Results
support our hypotheses with a caveat that warrants detailed discussion and suggests future
avenues of research.

Namely, despite precedent in the literature for a 30-min washout period between condi-
tion, our results suggested that 30 min between the active and sham conditions was insuffi-
cient, and that the inhibitory effects of rTMS (active-first condition) or excitatory effects of
picture naming (sham condition) interacted with other experimental events that followed in
the session, confounding the effects of stimulation (active vs. sham rTMS) and/or target
(R PTr vs. R POp) (Lang et al., 2004; Siebner et al., 2004). To explore this possible interaction,
we stratified our sample by the order in which they received active vs. sham stimulation within

Neurobiology of Language

264

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

session and repeated our analysis. This was statistically and practically feasible because the
groups were well-counterbalanced. Ce faisant, we discovered that the hypothesized effects
in one older group (active-first) were obscured by the lack of any effects in the other group
(sham-first).

The remainder of the discussion deals with the following: (1) We interpret the results of the
R PTr vs. the R POp stimulation conditions in the active-first group within the context of our
experimental hypothesis. (2) We explain how experimental manipulations in the condition
administered first (active or sham) interacted with subsequent experimental conditions to affect
the results of the condition administered second in older adults. (3) The implications of our
findings for future experimental design as well as the broader implications for rehabilitation
are discussed. Enfin, a brief summary is offered.

R PTr vs. R POp in Active-First Group

We examined the possibility that order of stimulation within session (active first vs. sham first)
interacted with type of stimulation (active vs. sham) and target of stimulation (R PTr vs. R POp),
emphasizing older adults. This order of stimulation within sessions had been counterbalanced
across participants, with half in each group receiving active rTMS first on each of the two visits,
and half receiving sham rTMS first on each visit. When the temporal order of active vs. sham
rTMS was included in our statistical model, we found convincing support for our experimental
hypotheses in the active-first group (Chiffre 3). This evidence had been partially obscured in
our primary analysis by the lack of an effect in the sham-first group. Plus loin, there was a sig-
nificant difference in the sham conditions for R PTr vs. R POp, with the sham condition for
R PTr having significantly longer SOLs (c'est à dire., slower responding) than for R POp in the
active-first group, calling into question the validity of the comparisons of the active and sham
conditions. Ainsi, we rely on the comparison of R PTr and R POp active stimulation in the
active-first group to address our hypothesis regarding R PTr activity during picture naming.

After 1 Hz rTMS of R PTr in the older active-first group, SOLs for picture naming were faster
than after 1 Hz rTMS of the control area, R POp. This finding supports the idea that decreasing
the excitability of (c'est à dire., inhibiting) R PTr is removing its interference with picture naming. It also
contradicts the concept that R PTr activity is a compensatory mechanism that aids picture
naming because inhibiting such a compensatory mechanism would make picture naming less
efficient (c'est à dire., slower), the opposite effect from our findings. One might ask: Why does R PTr
activity increase during picture naming if it is not compensatory? The work of McGregor et al.
(2013, 2018) shows an analogous situation in the motor system to that in the language system.
Spécifiquement, R motor cortex (M1) shows decreased activity in younger adults during R hand
movements that changes to increased activity in sedentary older (middle aged) adultes
(McGregor et al., 2013), similar to Wierenga et al.’s (2008) picture-naming findings in
R PTr. (Recall that adults in the present study also were sedentary.) The increased R M1 activity
in older adults is accompanied by poorer performance in speeded and skilled motor activities
of the R hand (McGregor et al., 2013). Plus loin, decreased R hand motor skills in sedentary
older adults is associated with decreased interhemispheric inhibition of R M1 after single-pulse
TMS stimulation of L M1 (McGregor et al., 2013), indicating that reduced interhemispheric
inhibition in older adults might account for the poorer motor performance. Since the subjects
in McGregor’s and Wierenga’s experiments were right-handed, we can assume left-hemisphere
dominance for motor and language functions in these experiments. These parallels suggest
that the relationship between interhemispheric inhibition and word-finding efficiency should
be investigated in older adults.

Neurobiology of Language

265

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

Serial Dependency Effects and Active vs. Sham Comparisons in Older Adults

The reader will remember that for the active-first group, sham stimulation for the R PTr session
showed significantly slower SOLs than sham stimulation for the R POp session. Theoretically,
this should not happen because sham stimulation should have no effects on performance.
Plus loin, the reader will recall that there were no differences in SOLs after the active vs. le
sham condition for the sham-first group of older adults. It was asserted earlier that these two
findings, which were contrary to our predictions, could be explained by interactions between
the final picture-naming measure in the stimulation paradigm that was administered second
and the experimental events that preceded it. En effet, the veracity of the above interpretation
of R PTr vs. R POp stimulation effects depends to some degree on our ability to explain these
unexpected findings.

Why SOLs in the sham condition for R PTr are increased

To understand why the SOLs for sham stimulation in the R PTr session of the active-first group
were longer than those for sham stimulation in the R POp (control) session, we must under-
stand the order of events that produce the serial dependencies for the picture-naming measure
in the sham condition. Picture naming for sham stimulation in the active-first group was pre-
ceded first by active 1 HzrTMS, which has the effect of inhibiting R PTr. This stimulation was
followed by repetitive picture naming which tends to excite R PTr in older adults (Wierenga
et coll., 2008). The reader may recall from the study of Lang et al. (2004) that inhibitory stim-
ulation followed by weak excitatory stimulation (20 s of 5 HzrTMS) significantly enhances the
effects of excitatory stimulation, producing enhanced excitatory effects above that of receiving
only sham stimulation at 20–30 min post excitatory stimulation, similar to the timeframe of the
current experiment. Ainsi, the increased excitatory effects of picture naming that are precondi-
tioned by 1 Hz rTMS should increase the interference effects of R PTr activity evoked by picture
naming after sham stimulation, lequel, à son tour, would cause increased SOLs for the measurement
taken after sham stimulation. The effects of this sequence of events are illustrated in Figure 4.
Plus loin, it must be emphasized that this explanation requires that R PTr interferes with picture
naming, providing further evidence for our hypothesis that R PTr in older adults interferes with
picture naming.

Why there is no difference between active and sham stimulation in the sham-first group

To understand the serial dependency for the active condition in the sham-first group, we must
again review the order of events. The first active event for this group is the picture-naming
measurement after sham stimulation, which has an excitatory impact on R PTr. This is followed
later by 1 Hz (inhibitory) SMTr. A study speaking to the order of events for the sham-first group
is that of Siebner et al. (2004). They performed 10 min of anodal tDCS stimulation (lequel
increases cortical excitability), followed 10 min later by 15 min of 1 HzrTMS. Siebner
et autres. found that greater excitability in response to anodal tDCS leads to greater suppression
of cortical excitability after 1 HzrTMS, and that weaker response to anodal tDCS results in
weaker suppression of excitability after 1 HzrTMS. Ainsi, if the excitatory stimulation
afforded by 30 picture-naming trials is weak in terms of its excitatory after-effects (because
of a small number of trials, or the spacing between trials (~2–4 s), etc.), then its precondition-
ing could weaken the effects of 1 HzrTMS, thereby negating the rTMS effects. Un tel
séquence, alors, could lead to a lack of difference between the sham and active 1 HzrTMS
for the sham-first older group. (It should be remembered that the effects on the sham trial of the
active-first group discussed in the preceding paragraph require only weak excitatory stimula-
tion; Lang et al., 2004).

Neurobiology of Language

266

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

Chiffre 4. Reversed polarity effects during sham PTr condition for participants receiving active rTMS first. (UN) Active 1 Hz rTMS is applied to R
PTr. (B) First picture-naming task. A moderate decrease in R PTR excitability leads to faster (c'est à dire., shorter) speech onset latency. Cependant, the act
of word retrieval, potentiated by the preceding rTMS, begins to strongly increase excitability of R PTr. (C) Sham rTMS. Even though no rTMS is
delivered, the excitability effects just described are accompanied by an increase in R PTr excitability. (D) Second picture-naming task. A strong
increase in R PTr excitability leads to slower (c'est à dire., longer) speech onset latency. (E) It is uncertain how long the increased R PTr excitability
might last after the second picture-naming task, so this phase is depicted by a dashed line.

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Implications of Serial Dependency Effects

The current report has introduced the idea that there may be serial dependency between
repeated behaviors and NIBS techniques that may introduce unexpected effects on subsequent
behavior. In the following paragraphs, we will discuss two important issues regarding these
serial dependencies. The first issue is how to avoid the kind of confounds that interaction
between NIBS and repetitive behavior introduce into experimental designs and their interpre-
tation. Hopefully, this discussion can be of some help avoiding such interpretative complica-
tions in the future. Deuxième, NIBS interventions are receiving a great deal of attention in the
rehabilitation literature (Crosson et al., 2019). Ainsi, interactions between NIBS and behav-
ioral treatments could invoke unexpected consequences (positive or negative) during rehabil-
itation studies or treatments. These two topics are covered below.

Avoiding serial dependencies between NIBS and repetitive behavior

An important implication of the preceding discussion is that repetitive picture naming (ou autre
repetitive behaviors) should not be viewed solely as inert measurements. They may have a
weak excitatory effect that interacts with preceding or subsequent brain stimulation or another
episode of picture naming. This hypothesis is itself worth putting to further test because a
knowledge of the excitatory effects of repetitive picture naming could affect the design of
future experiments in which NIBS techniques are used to test neurocognitive theories. Le
key consideration in such experiments is how to mitigate the interactive effects of NIBS and
picture naming. One obvious remedy for the order effects described above is to put more time
between active inhibitory, active excitatory, and/or sham conditions with repetitive behavioral
measurements of their effects. Le 30 min between conditions of the current study obviously
was not enough, but how much more time is enough? In the tDCS neuromodulation literature,
it is common to put a week between sessions, consistent with recent recommendations by
Woods et al. (2016) for avoiding long-term neuroplasticity effects of tDCS. Par exemple, dans

Neurobiology of Language

267

1 Hz rTMS of right pars triangularis prior to naming in older adults

their study of the effects of tDCS on word retrieval, Meinzer et al. (2013) had at least a week
between active and sham tDCS sessions. Plus loin, in the one study with only two to three
days between sessions (Lifshitz Ben-Basat & Mashal, 2017) there was also an order effect.
Based upon these considerations, the recommendation of a week between different active
and/or sham sessions to avoid accumulation of long-term effects seems appropriate for any
NIBS technique.

Implications for rehabilitation

We have discussed ways in which repeated behavioral activation of cortex and NIBS of the
same cortex might interact to potentiate or negate the effects of NIBS. Potentially, there are
numerous ways in which such serial dependencies might be studied and eventually applied
in rehabilitation. Ici, we mention one example: It has become common for studies to use
anodal stimulation of language-eloquent dominant-hemisphere cortices combined with lan-
guage training for aphasia in rehabilitation studies (see Crosson et al., 2019, for a review, ou
Fridriksson et al., 2018, for a more recent example). Lang et al. (2004) showed that anodal
stimulation of M1 followed by a weak excitatory stimulation of the same cortex resulted in
a reversal of the excitatory effects of anodal tDCS so that they became inhibitory effects
that began somewhere between 10–20 min after the initial tDCS. It has not been uncommon
to begin language training during tDCS and to continue it after tDCS has been ended. Mais,
if language training has a weak excitatory effect on the language-eloquent cortex that is
stimulated, alors, could continuing language training much past the end of tDCS be
counter-productive because it eventually leads to inhibitory effects on the stimulated
language-eloquent cortex that is needed for successful language training? Plus loin, pourrait
differences in whether or how long the behavioral component of therapy continued after
tDCS account for some of the differences in outcomes between combined tDCS and
behavioral treatment that were noted by Crosson et al. (2019). This analysis suggests that it
would be worth investigating the best time after tDCS to stop language training to achieve
optimal results.

Conclusions

The current study was designed to test the proposition that R PTr activity in older adults inter-
feres with picture naming. Spécifiquement, 1 HzrTMS, shown to reduce cortical excitability, était
administered to R PTr immediately before SOLs to picture naming were measured. It was
hypothesized that this low-frequency rTMS would suppress the interference of R PTr on picture
naming, leading to faster SOLs. When within-session order effects were modeled, findings
indicated that there were serial dependencies between picture naming and 1 Hz rTMS that
invalidated comparisons with sham stimulation for r PTr. Néanmoins, in the active-first group
of older adults, comparison of active R PTr stimulation with active stimulation of a control
area, R POp, showed faster SOLs for R PTr, indicating that R PTr activity in older adults inter-
feres with word finding. Plus loin, when taken in the context of the work of Lang et al. (2004),
the lengthening of SOLs after sham stimulation in the active-first group also indicates R PTr
activity interferes with word finding in older adults. No evidence from this study supports
the idea that R PTr activity provides compensatory support of word finding in older adults.
Serial dependencies between behavioral activation and NIBS, such as those found in the cur-
rent study, can be avoided in future studies of the cognitive effects of NIBS by allowing the
effects of NIBS to dissipate over one week before performing other active or sham NIBS con-
ditions. Enfin, the implications of interactions between repetitive behavioral manipulations
and NIBS for rehabilitation should be studied.

Neurobiology of Language

268

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

REMERCIEMENTS

Work on this study was supported by the following grants: Grant # I21 RX00099401 (Bruce
Crosson), Senior Research Career Scientist Award # B6364-L (Bruce Crosson), and Career
Development Award – Level 2 # IK2 RX000956 (Keith M. McGregor), all from the United States
Department of Veterans Affairs Rehabilitation Research and Development Service. The views
expressed in this work do not necessarily reflect those of the United States Government or
the Department of Veterans Affairs. All authors contributed significantly to the production of
this work.

INFORMATIONS SUR LE FINANCEMENT

Bruce A. Crosson, U.S. Department of Veterans Affairs (https://dx.doi.org/10.13039
/100000738), Award ID: I21 RX00099401. Bruce A. Crosson, U.S. Department of Veterans
Affaires (https://dx.doi.org/10.13039/100000738), Award ID: B6364-L. Keith M. McGregor,
U.S. Department of Veterans Affairs (https://dx.doi.org/10.13039/100000738), Award ID: IK2
RX000956.

CONTRIBUTIONS DES AUTEURS

Jonathan H. Drucker: Conceptualisation: Supporting; Conservation des données: Lead; Analyse formelle:
Lead; Enquête: Lead; Méthodologie: Equal; Gestion de projet: Supporting; Logiciel:
Lead; Visualisation: Equal; Rédaction – ébauche originale: Equal; Rédaction – révision & édition: Equal.
Charles M. Epstein: Conceptualisation: Supporting; Méthodologie: Equal; Ressources: Support-
ing; Surveillance: Supporting; Rédaction – ébauche originale: Supporting; Rédaction – révision & édition:
Equal. Keith M. McGregor: Conceptualisation: Equal; Conservation des données: Equal; Funding acquisi-
tion: Equal; Enquête: Supporting; Méthodologie: Supporting; Surveillance: Supporting;
Rédaction – révision & édition: Equal. Kyle Hortman: Conservation des données: Supporting. Kaundinya S.
Gopinath: Méthodologie: Supporting. Bruce Crosson: Conceptualisation: Lead; Formal analy-
sis: Supporting; Acquisition de financement: Lead; Méthodologie: Equal; Gestion de projet: Lead;
Ressources: Lead; Surveillance: Lead; Visualisation: Supporting; Rédaction – ébauche originale: Equal;
Rédaction – révision & édition: Lead.

COMPETING INTERESTS

Charles M. Epstein received royalties from Neuronetics, Inc., which manufactures transcranial
magnetic stimulators. No Neuronetics equipment was used in this study.

RÉFÉRENCES

Berlingeri, M., Danelli, L., Bottini, G., Sberna, M., & Paulesu, E.
(2013). Reassessing the HAROLD model: Is the hemispheric
asymmetry reduction in older adults a special case of
compensatory-related utilisation of neural circuits? Experimental
Brain Ressearch, 224(3), 393–410. https://doi.org/10.1007
/s00221-012-3319-x, PubMed: 23178904

Cabeza, R.. (2001). Cognitive neuroscience of aging: Cotisations
of functional neuroimaging. Scandinavian Journal of Psychology,
42(3), 277–286. https://doi.org/10.1111/1467-9450.00237,
PubMed: 11501741

Cabeza, R.. (2002). Hemispheric asymmetry reduction in older adults:
The HAROLD model. Psychological Aging, 17(1), 85–100. https://
doi.org/10.1037/0882-7974.17.1.85, PubMed: 11931290

Cabeza, R., Daselaar, S. M., Dolcos, F., Prince, S. E., Budde, M., &
Nyberg, L. (2004). Task-independent and task-specific age effects
on brain activity during working memory, visual attention and
episodic retrieval. Cortex cérébral, 14(4), 364–375. https://est ce que je
.org/10.1093/cercor/bhg133, PubMed: 15028641

Crosson, B., Rodriguez, UN. D., Copland, D., Fridriksson, J.,
Krishnamurthy, L., Meinzer, M., Raymer, UN. M., Krishnamurthy,
V., & Leff, UN. (2019). Neuroplasticity and aphasia treatments:
New approaches for an old problem. Journal of Neurology,
Neurochirurgie, and Psychiatry, 90, 1147–1155. https://est ce que je.org/10
.1136/jnnp-2018-319649, PubMed: 31055282

Delis, D. C., Kramer, J.. H., Kaplan, E., & Obler, B. UN. (2000). California
Verbal Learning Test – Second Edition. Psychological Corporation.

Neurobiology of Language

269

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

Dmochowski, J.. P., Datta, UN., Bikson, M., Su, Y., & Parra, L. C.
(2011). Optimized multi-electrode stimulation increases focality
and intensity at target. Journal of Neural Engineering, 8(4), Article
046011. https://doi.org/10.1088/1741-2560/8/4/046011,
PubMed: 21659696

Dolcos, F., Rice, H. J., & Cabeza, R.. (2002). Hemispheric asymme-
try and aging: Right hemisphere decline or asymmetry reduction.
Neuroscience and Biobehavioral Reviews, 26(7), 819–825.
https://doi.org/10.1016/s0149-7634(02)00068-4

Fridriksson, J., Morrow, K. L., Moser, D., & Baylis, G. C. (2006).
Age-related variability in cortical activity during language
traitement. Joural of Speech, Language, and Hearing Research,
49(4), 690–697. https://doi.org/10.1044/1092-4388(2006/050),
PubMed: 16908869

Fridriksson, J., Richardson, J.. D., Boulanger, J.. M., & Rorden, C. (2011).
Transcranial direct current stimulation improves naming reaction
time in fluent aphasia: A double-blind, sham-controlled study.
Stroke, 42(3), 819–821. https://doi.org/10.1161/STROKEAHA
.110.600288, PubMed: 21233468

Fridriksson, J., Rorden, C., Elm, J., Sen, S., George, M.. S., & Bonilha,
L. (2018). Transcranial direct current stimulation vs. sham stimu-
lation to treat aphasia after stroke: A randomized clinical trial.
JAMA Neurology, 75(12), 1470–1476. https://doi.org/10.1001
/jamaneurol.2018.2287, PubMed: 30128538

Holland, R., Leff, UN. P., Joseph, O., Galea, J.. M., Desikan, M.,
Prix, C. J., Rothwell, J.. C., & Crinion, J.. (2011). Speech facilita-
tion by left inferior frontal cortex stimulation. Biologie actuelle,
21(16), 1403–1407. https://doi.org/10.1016/j.cub.2011.07.021,
PubMed: 21820308

Hoyau, E., Boudiaf, N., Cousin, E., Pichat, C., Fournet, N., Krainik,
UN., Jaillard, UN., & Baciu, M.. (2017). Aging modulates the hemi-
spheric specialization during word production. Frontiers in Aging
Neurosciences, 9, Article 125. https://doi.org/10.3389/fnagi.2017
.00125, PubMed: 28536520

James, G. UN., Thostenson, J.. D., Brun, G., Carter, G., Hayes, H.,
Tripathi, S. P., Dobry, D. J., Govindan, R.. B., Dornhoffer, J.. L.,
Williams, D. K., Kilts, C. D., & Mennemeier, M.. S. (2017). Neural
activity during attentional conflict predicts reduction in tinnitus
perception following rTMS. Brain Stimulation, 10(5), 934–943.
https://doi.org/10.1016/j.brs.2017.05.009, PubMed: 28629874
Kaplan, E., Goodglass, H., & Weintraub, S. (2001). Boston Naming

Test – Second Edition. Lippincott, Williams & Wilkins.

Lang, N., Siebner, H. R., Ernst, D., Nitsche, M.. UN., Paulus, W.,
Lemon, R.. N., & Rothwell, J.. C. (2004). Preconditioning with
transcranial direct current stimulation sensitizes the motor cortex
to rapid-rate transcranial magnetic stimulation and controls the
direction of after-effects. Biological Psychiatry, 56(9), 634–639.
https://doi.org/10.1016/j.biopsych.2004.07.017, PubMed:
15522246

Lifshitz Ben-Basat, UN., & Mashal, N. (2017). Improving naming abil-
ities among healthy young-old adults using transcranial direct
current stimulation. Journal of Psycholinguistic Research, 47,
113–124. https://doi.org/10.1007/s10936-017-9516-9, PubMed:
28856553

Luis, C. UN., Keegan, UN. P., & Mullan, M.. (2009). Cross validation of
the Montreal Cognitive Assessment in community dwelling older
adults residing in the Southeastern US. Revue internationale de
Geriatric Psychiatry, 24(2), 197–201. https://doi.org/10.1002/gps
.2101, PubMed: 18850670

McGregor, K. M., Crosson, B., Mammino, K., Garcia, P.. S., &
Nocera, J.. (2018). Influences of 12-week physical activity inter-
ventions on TMS measures of cortical network inhibition and
upper extremity motor performance in older adults – A feasibility

étude. Frontiers in Aging Neuroscience, 9, Article 422. https://est ce que je
.org/10.3389/fnagi.2017.00422, PubMed: 29354049

McGregor, K. M., Nocera, J.. R., Sudhyadhom, UN., Patten, C.,
Manini, T., Kleim, J.. UN., Crosson, B., & Butler, UN. J.. (2013). Effects
of aerobic fitness on aging-related changes of interhemispheric
inhibition and motor performance. Frontiers in Aging Neurosci-
ence, 5, Article 66. https://doi.org/10.3389/fnagi.2013.00066,
PubMed: 24198784

Meinzer, M., Flaisch, T., Wilser, L., Eulitz, C., Rockstroh, B.,
Conway. T., Gonzalez-Rothi, L., & Crosson, B. (2009). Neural
signatures of semantic and phonemic fluency in young and old
adultes. Journal des neurosciences cognitives, 21(10), 2007–2018.
https://doi.org/10.1162/jocn.2009.21219, PubMed: 19296728
Meinzer, M., Lindenberg, R., Antonenko, D., Flaisch, T., Flöel, UN.
(2013). Anodal transcranial direct current stimulation temporarily
reverses age-associated cognitive decline and functional brain
activity changes. Journal des neurosciences, 33(30), 12470–12478.
https://doi.org/10.1523/ JNEUROSCI.5743-12.2013, PubMed:
23884951

Meinzer, M., Seeds, L., Flaisch, T., Harnish, S., Cohen, M.. L.,
McGregor, K., Conway, T., Benjamin, M., & Crosson, B. (2012).
Impact of changed positive and negative task-related brain
activity on word-retrieval in aging. Neurobiology of Aging, 33(4),
656–669. https://doi.org/10.1016/j.neurobiolaging.2010.06.020,
PubMed: 20696496

Naeser, M.. UN., Martine, P.. JE., Nicholas, M., Boulanger, E. H., Seekins, H.,
Kobayashi, M., Theoret, H., Fregni, F., Maria-Tormos, J., Kurland,
J., Doron, K. W., & Pascual-Leone, UN. (2005). Improved picture
naming in chronic aphasia after TMS to part of right Broca’s area:
An open-protocol study. Brain and Language, 93(1), 95–105.
https://doi.org/10.1016/j.bandl.2004.08.004, PubMed:
15766771

Naeser, M.. UN., Martine, P.. JE., Nicholas, M., Boulanger, E. H., Seekins, H.,
Kobayashi, M., Theoret, H., Fregni, F., Maria-Tormos, J., Kurland,
J., Doron, K. W., & Pascual-Leone, UN. (2011). TMS suppression of
right pars triangularis, but not pars opercularis, improves naming
in aphasia. Brain and Language, 119(3), 206–213. https://doi.org
/10.1016/j.bandl.2011.07.005, PubMed: 21864891

Nasreddine, Z. S., Phillips, N. UN., Bédirian, V., Charbonneau, S.,
Whitehead, V., Collin, JE., Cummings, J.. L., & Chertkow, H.
(2005). The Montreal Cognitive Assessment, MoCA: A brief
screening tool for mild cognitive impairment. Journal of the
American Geriatric Society, 53(4), 695–699. https://est ce que je.org/10
.1111/j.1532-5415.2005.53221.x, PubMed: 15817019

Nocera, J., Crosson, B., Mammino, K., & McGregor, K. M.. (2017).
Changes in cortical activation patterns in language areas follow-
ing an aerobic exercise intervention in older adults. Neural Plas-
ticity, 2017, Article 6340302. https://doi.org/10.1155/2017
/6340302, PubMed: 28367334

Nocera, J.. R., Mammino, K., Kommula, Y., Wharton, W., Crosson, B.,
& McGregor, K. M.. (2020). Effects of combined aerobic exercise
and cognitive training on verbal fluency in older adults.
Gerontology and Geriatric Medicine, 6, Article 2333721419896884.
https://doi.org/10.1177/2333721419896884, PubMed: 31950073
Nocera, J.. R., McGregor, K. M., Hass, C. J., & Crosson, B. (2015).
Spin exercise improves semantic fluency in previously sedentary
older adults. Journal of Aging and Physical Activity, 23(1), 90–94.
https://doi.org/10.1123/japa.2013-0107, PubMed: 24425525
Persson, J., Sylvestre, C. Y., Nelson, J.. K., Welsh, K. M., En Ionie, J., &
Reuter-Lorenz, P.. UN. (2004). Selection requirements during verb
generation: Differential recruitment in older and younger adults.
NeuroImage, 23(4), 1382–1390. https://est ce que je.org/10.1016/j
.neuroimage.2004.08.004, PubMed: 15589102

Neurobiology of Language

270

D
o
w
n
o
un
d
e
d

F
r
o
m
h

t
t

:
/
/

d
je
r
e
c
t
.

je
t
.

e
d
toi
n
o

je
/

un
r
t
je
c
e
–
p
d

F
/

3
2
2
5
6
2
0
0
2
5
3
1
n
o
_
un
_
0
0
0
6
3
p
d

b
oui
g
toi
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

1 Hz rTMS of right pars triangularis prior to naming in older adults

Ren, C.-L., Zhang, G.-F., Xian N., Jin, C. H., Zhang, X. H., Hao, J.. F.,
Guan, H. B., Tang, H., Li, J.. UN., & Cai, D. L. (2014). Effect of low-
frequency rTMS on aphasia in stroke patients: A meta-analysis of
randomized controlled trials. PLOS ONE, 9(7), Article e102557.
https://doi.org/10.1371/journal.pone.0102557, PubMed:
25036386

Siebner, H. R., Lang, N., Rizzo, V., Nitsche, M.. UN., Paulus, W.,
Lemon, R.. M., & Rothwell, J.. C. (2004). Preconditioning of
low-frequency repetitive transcranial magnetic stimulation with
transcranial direct current stimulation: Evidence for homeostatic
plasticity in the human motor cortex. Journal des neurosciences,
24(13), 3379–3385. https://doi.org/10.1523/ JNEUROSCI.5316
-03.2004, PubMed: 15056717

Wierenga, C. E., Benjamin, M., Gopinath, K., Perlstein, W. M.,
Leonard, C. M., Rothi, L. J.. G., Conway, T., Cato, M.. UN., Briggs,
R., & Crosson, B. (2008). Age-related changes in word retrieval:

Role of bilateral frontal and subcortical networks. Neurobiology of
Aging, 29(3), 436–451. https://doi.org/10.1016/j.neurobiolaging
.2006.10.024, PubMed: 17147975

Woods, UN. J., Antal, UN., Bikson, M., Boggio, P.. S., Brunoni, UN. R., Celnik,
P., Cohen, L. G., Fregni, F., Herrmann, C. S., Kappenman, E. S.,
Knotkova, H., Liebetanz, D., Miniussi, C., Miranda, P.. C., Paulus,
W., Priori, UN., Reato, D., Stagg, C., Wenderoth, N., & Nitsche, M.. UN.
(2016). A technical guide to tDCS, and related non-invasive brain
stimulation tools. Neurophysiologie clinique, 127(2), 1031–1048.
https://doi.org/10.1016/j.clinph.2015.11.012, PubMed: 26652115
Zlatar, Z. Z., Towler, S., McGregor, K. M., Dzierzewski, J.. M.,
Bauer, UN., Phan, S., Cohen, M., Marsiske, M., Manini, T. M., &
Crosson, B. (2013). Functional language networks in sedentary
and physically active older adults. Journal of the International
Neuropsychological Society, 19(6), 625–634. https://est ce que je.org/10
.1017/S1355617713000246, PubMed: 23458438