Neuroscience Letters 589 (2015) 115–120

Contents lists available at ScienceDirect

Neuroscience Letters
journal homepage: www.elsevier.com/locate/neulet

Research article

Effects of transcranial direct current stimulation on naming

and cortical excitability in stroke patients with aphasia
Dongyu Wu a,∗,1 , Jie Wang a,1 , Ying Yuan b
a
Department of Rehabilitation, China–Japan Friendship Hospital, No. 2, Yinghuadongjie, Chaoyang District, PO Box 100029, Beijing, China
b
Department of Rehabilitation, Xuanwu Hospital of Capital Medical University, China

h i g h l i g h t s

• Anodal tDCS over the left posterior perisylvian region can improve picture naming in new onset aphasic patients.
• Anodal tDCS may up-regulate excitability of a language network.
• Nonlinear dynamics analysis can reflect cortical activation during language tasks.

a r t i c l e i n f o a b s t r a c t

Article history: This study aimed to investigate the effects of anodal transcranial direct current stimulation (A-tDCS) over
Received 6 November 2014 the left posterior perisylvian region (PPR) on picture naming and cortical excitability measured with
Received in revised form electroencephalography (EEG) nonlinear dynamics analysis (NDA) in aphasic patients. Twelve apha-
15 December 2014
sic patients received 20 sessions of speech-language therapy during each of three phases: sham tDCS
Accepted 16 January 2015
(Phase A1); A-tDCS to the left PPR (Phase B); and sham tDCS (Phase A2). Picture naming and auditory
Available online 17 January 2015
word-picture identification were measured before and after each phase. The EEG nonlinear index of
approximate entropy (ApEn) was calculated for all subjects and 12 normal controls. Picture naming and
Keywords:
Aphasia
auditory word-picture identification was significantly improved after phase B. The EEG ApEn analysis
Brain stimulation indicated that improved picture naming correlated with a higher activation level in wide areas of the
Electroencephalography left hemisphere and in isolated areas of the right hemisphere after phase B. These results revealed that
Nonlinear dynamics A-tDCS over the left PPR coupled with speech-language therapy can improve picture naming and audi-
Rehabilitation tory comprehension in aphasic patients. tDCS not only modulates activity in the brain region directly
underlying the stimulating electrode but also in a network of brain regions that are function-related.
© 2015 Published by Elsevier Ireland Ltd.

1. Introduction
Transcranial direct current stimulation (tDCS), a non-invasive
method, delivers a weak polarizing direct current (1–2 mA) to the
cortex via two electrodes placed on the scalp. Cathodal stimulation
(C-tDCS) reduces cortex excitability, whereas anodal stimulation
(A-tDCS) increases cortex excitability.
Several studies of healthy and aphasic subjects have suggested
that tDCS could be used to improve language performance through
exciting or inhibiting corresponding cortical areas. Picture nam-
∗ Corresponding author. Tel.: +86 10 84205968.
E-mail address: wudongyu73@hotmail.com (D. Wu).
1
Both authors contributed equally to this work.
ing performance improved following C-tDCS over the right Broca’s
homologue area in ten aphasic patients [1]; and A-tDCS over the left
frontal cortex in 10 patients with chronic stroke-induced aphasia
[2], left posterior cortex in 8 chronic fluent aphasic patients [3], and
the right temporo-parietal cortex in 12 chronic post-stroke aphasia
patients [4]. Therefore, tDCS has been proposed as a clinical tool for
rehabilitation of picture naming in brain-damaged patients and as
a supplementary treatment approach for anomia [5].
Language function such as word repetition involves several
cerebral areas, particularly the superior temporal gyrus respon-
sible for spoken word recognition, articulatory representations in
the ventral premotor cortex (vPM), and the posterior part of the
inferior frontal gyrus (pIFG). The word repetition and picture-
naming tasks are performed through the interactions of these areas.
Therefore, local A-tDCS over the left PPR possibly could activate

http://dx.doi.org/10.1016/j.neulet.2015.01.045
0304-3940/© 2015 Published by Elsevier Ireland Ltd.
116 D. Wu et al. / Neuroscience Letters 589 (2015) 115–120
these anatomic regions and improve language performance in post-
stroke aphasic patients.
In a study of arterial spin labeling, A-tDCS not only modulated
activity in the brain region directly underlying the stimulating elec-
trode, but also in a network of brain regions that were functionally
related to the stimulated area [6]. Several issues have yet to be
clearly addressed, such as the excitability status of the underly-
ing cortical tissues before and after tDCS; how the cortical status
of other regions is impacted or influenced if one area of the net-
work was stimulated by A-tDCS; and whether cortical excitability
is associated with language performance.
To the best of our knowledge, cortical neuroelectrical activation
after A-tDCS in aphasic patients has not been studied. Moreover,
tDCS may directly change the electrical activation of cortical neu-
rons, and the most easily performed clinical technique for observing
cortical electrical activation under different language tasks is elec-
troencephalography (EEG).
EEG is thought to be complex and of limited predictability
because its ultra-high dimensional nature makes it a stochastic sys-
tem [7,8]. Nonlinear dynamics analysis (NDA) can characterize the
dynamics of the neural networks underlying the EEG and provide
a powerful tool for studying the dynamic changes and abstracting
correlative information from cortical networks [9]. With NDA, the
degree of suppression for unconscious patients is quantifiable. The
changes in brain function in unconscious subjects can be captured
by NDA, and may predict the prognosis of unconscious patients by
assessing the interrelation among sensory and other residual cor-
tical functional islands [10]. Therefore, NDA could be utilized to
provide some information about cortical excitability.
The purpose of this study was to determine whether aphasic
patients with a post-stroke onset of 3–6 months could benefit from
tDCS plus speech-language therapy and to investigate cortical acti-
vation changes using NDA. Our hypotheses included the following:
(1) A-tDCS over the left PPR may improve picture naming in post-
stroke patients with aphasia and potentially enhance the cortex to
a “near normal” activation pattern, which could be associated with
better recovery. (2) NDA could reflect cortical activation under lan-
guage tasks. To investigate, A-tDCS and sham tDCS were applied
over the left PPR and the EEG nonlinear index of ApEn was calcu-
lated for all of the patients.
2. Methods
2.1. Design and procedures
Thirty minutes of daily speech-language therapy was provided
over four weeks using an A1-B-A2 design. Sham tDCS was adminis-
tered during Phases A1 and A2, and A-tDCS was given during Phase
B. Each subject received a standardized aphasia assessment battery
before and after each treatment phase. An EEG was recorded after
each treatment phase. The trial structure is illustrated in Fig. 1.
2.2. Participants
The study was performed in the Department of Rehabilita-
tion, Xuanwu Hospital of Capital Medical University, Beijing, China.
All participants were right-handed and native speakers of Man-
darin Chinese. The hospital ethics committee approved the study.
Informed written consent was obtained from each subject.
2.3. Patient group
Inclusion criteria included: (1) 3–6 months after the onset of a
stroke, (2) no previous brain injury, (3) no significant improvement
in picture naming found after a 4-week language-based treatment
period (Phase A1), and (4) no significant difference found in left
Wernicke’s point (see Section 5) via EEG ApEn index during word
repetition task compared with the resting condition before entering
Phase B. Exclusion criteria included: (1) severely impaired audi-
tory verbal comprehension (auditory word-picture identification
<6/60), (2) inability to repeat three-syllable words, (3) history of
seizures in the last 12 months, (4) history of implanted metal
objects, (5) history of seizures or other neurological condition, and
(6) history of psychiatric disease or dementia.
Twelve (10 men) of the original fifteen aphasic patients quali-
fied for the study. Two patients withdrew from the trial for personal
reasons, and one patient did not meet the inclusion criteria. Patient
demographics and stroke characteristics are reported in Table 1.
Twelve healthy age-, sex-, and education-matched controls (10
men) also participated in this study.
2.4. Aphasia assessment
Aphasia type and severity were evaluated using the Boston
Diagnostic Aphasia Examination-Chinese version. All subjects
underwent an evaluation involving picture naming and auditory
word identification according to the Psycholinguistic Assessment
in Chinese Aphasia (PACA). Four 60-test-item lists were randomly
selected from a total of 90 items in the PACA. The lists were
controlled for word frequency, familiarity, visual complexity, gram-
matical class (nouns) and length in syllables (two syllables, i.e.,
two Chinese characters). The four 60-test-item lists were randomly
used in four aphasia assessments. The reaction time for each item
was limited within 20 s. No cue was provided if the patient did not
respond correctly.
2.5. Speech-language therapy
Speech-language therapy was conducted with a computer sys-
tem. A total of 500 treatment items were used. The subject received
a picture-naming treatment, which consisted of a series of steps
encouraging the subjects to activate and link information about
meaning and sound characteristics of a target picture/word. In the
treatment, a picture was presented, the subject was asked to name
the picture, if failed the semantic cues were provided, such as the
object’s preposition, function and features. The initial phoneme of
the word was provided if the patient could not retrieve the first
sound of the word.
2.6. Transcranial direct current stimulation (tDCS)
To administer tDCS, direct current was applied through a
saline-soaked pair of surface sponge electrodes (4.5 cm × 5.5 cm)
and delivered by a specially developed, battery-driven, constant-
current stimulator (IS200, Chengdu, China). A constant current with
an intensity of 1.2 mA was applied for 20 min once daily, 5 days per
week.
In this study, A-tDCS was applied over the left posterior peri-
sylvian region (PPR). tDCS and speech-language therapy started
simultaneously. According to Friederici’s method, the left Wer-
nicke’s area was defined in the posterior temporal region as a
crossing point between T3-P3 and C3-T5 [11]. This crossing point
was known as the “left Wernicke’s point (LWP)”. The cathodal elec-
trode was placed on the unaffected shoulder to prevent any adverse
effects. For the sham stimulation, the stimulator was turned off
after 30 s. For both active and sham tDCS treatments, current inten-
sity was gradually increased and decreased [12].
2.7. EEG recording
The subject was awake during the entire recording process and
lay comfortably in a quiet ward. A 16-channel EEG (Wireless Digital
 D. Wu et al. / Neuroscience Letters 589 (2015) 115–120
Fig. 1. Flow chart of the trial structure.
EEG, ZN16E, Chengdu, China) recorded signals using the inter-
national 10–20 system. A linked earlobe electrode was used for
reference. Signals were digitized with a sample rate of 500 Hz, a
bandwidth of 0.3–100 Hz, and 12-bit analog-to-digital conversion
resolution. Data were stored on a hard drive for further analysis.
In order to minimize the ocular artifact and interference from the
electrodes not involved with word repetition, data from FP1, FP2,
O1 and O2 were excluded. The left Broca’s area was defined in the
fronto-temporal region as a crossing point between T3-Fz and F7-
Cz [11]. The crossing point was called “left Broca’s point (LBP)”. In
this study, T5 was placed on the LWP, T6 on the RWP (homologous
region of LWP in the right hemisphere), F7 on the LBP and F8 on the
RBP (homologous region of LBP on the right hemisphere).
The EEG was recorded under two conditions: resting with eyes
closed for approximately 5 min, followed by three-syllable-word
repetition with eyes closed. The duration of each three-syllable-
word repetition was 3 s. The repetition task was a subtest of the
PACA that consisted of 20 content nouns. In this study, word
117
repetition was chosen rather than picture naming primarily
because word repetition could be performed under the closed-eyes
condition. Details of the reason are described elsewhere [13].
2.8. Nonlinear dynamics analysis (NDA)
Approximate entropy (ApEn) is a measure of system complexity
proposed by Pincus and Singer [14]. For details on the algorithm, see
Wu et al. [10]. ApEn calibrates the extent of serial interrelationships,
quantifying a continuum that ranges from totally ordered (zero) to
completely random (infinite). It assigns a non-negative number to a
time series, with larger values corresponding to more complexity or
irregularity in the data. With increasing irregularity, knowing past
values will not enable reliable prediction of future values, and ApEn
will increase. Thus, increasing irregularity (increased ApEn) will
cause higher complexity in the time series, such as raised nonlinear
cell-dynamics or interaction of cortical networks.
118 D. Wu et al. / Neuroscience Letters 589 (2015) 115–120

2.9. Data selection 3.4. Normal subjects

Artifact-free epoch selection was performed offline by an expe-
rienced physician through visual inspection of the recordings. For
this analysis, a 50-Hz notch filter was used to minimize electrical
noise.
For the resting condition, 32,768 consecutive data points
(65.536 s) were chosen for further analysis. Subsequently, 20,480
data points consisting of 10 segments of the language task
epoch (2048 data points, 4.096 s epoch each) under the three-
syllable-word repetition condition were collected offline. ApEn was
calculated for all of the subjects.
2.10. Statistical analysis
SPSS (version 17) was used to perform all of the analyses.
To examine the effect of tDCS on treatment outcome, repeated-
measures analysis of variance was performed to analyze the
number of correctly named and comprehended items before and
after each treatment phase using time (baseline, T1–T3). Bonferroni
post hoc multiple comparisons were used to compared different
treatment phase. A paired t-test was used to analyze the follow-
ing: differences in the ApEn indices of normal subjects between
the resting condition and repetition task; and differences in the
ApEn indices of patients between the resting and repetition task
after each treatment phase. Statistical significance was defined as
p < 0.05, except for comparisons of ApEn indices with paired t-test
(p < 0.01).
3. Results
3.1. The clinical and baseline characteristics of the patients
Ultimately, 12 patients completed the trial. The entire patient
group tolerated tDCS well and reported no adverse effects. The
baseline characteristics of the patient group are listed in Table 1.
3.2. Results of PACA assessment
At T2, the number of correctly named and comprehended was
significantly higher than baseline and T1. However, there were no
significant differences between T1 and baseline, between T3 and
T2, which is suggestive of stable language function (Table 2).
3.3. Results of EEG ApEn analysis
EEG ApEn indices of all participants during the resting condition
and word repetition task are listed in Table 3.
When comparing the word repetition task with the resting con-
dition, the EEG ApEn indices were significantly higher in the LWP,
T3 (left middle temporal site), C3 (left central site), P3 (left parietal
site) and LBP.
3.5. Aphasic patients
EEG ApEn indices of the aphasic patients under the resting con-
dition and repetition task before and after phase B are listed in
Table 3. At T1, ApEn indices were significantly higher only in the P3
(left parietal site) during word repetition than during the resting
condition. However, ApEn indices were significantly higher in the
LWP, LBP, C3, P3, P4, T3 and RWP at T2 and in the LWP, C3 and P3
at T3.
4. Discussion
Our results confirmed both hypotheses. First, A-tDCS over the
left PPR coupled with speech-language therapy improved the
performance of picture naming and auditory word-picture iden-
tification in post-subacute aphasic patients. Second, before A-tDCS
(Phase A1), the EEG ApEn indices in the patient group were signifi-
cantly higher only in the P3 (left parietal site) and lower in the RBP
and T4 compared with the resting condition. However, after A-tDCS,
they were significantly higher in the LWP, LBP, C3, P3, P4, T3 and
RWP compared with the resting condition. The EEG ApEn analysis
indicated that improved picture naming correlated with a higher
activation level in wide areas of the left hemisphere (“near normal”
activation pattern) and in isolated areas of the right hemisphere.
The functional imaging studies demonstrated that auditory
word repetition can activate left lateralized regions in the frontal
operculum (Broca’s area), posterior superior temporal gyrus (Wer-
nicke’s area), posterior inferior temporal cortex, a mid-superior
temporal region [15], and the supplementary motor area [16]. In
this study, normal subjects demonstrated increased ApEn indices
in Wernicke’s area (LWP), Broca’s area (LBP), fronto-parietal (C3),
parietal (P3) and middle temporal (T3) regions during word rep-
etition tasks. Therefore, investigation of cortical activation using
a word repetition task with NDA produced results similar to the
above fMRI studies.
Many regions are activated during picture naming, including
the mid-to-posterior lateral temporal cortex, which is involved in
lexical-conceptual and phonological processing; and Broca’s area,
which is involved in speech production. Thus, evidence suggests
that a majority of the activated areas in word repetition overlapped
with those in picture naming, particularly the left posterior supe-
rior temporal and posterior inferior frontal areas. In this study, the

Table 1
Clinical characteristics of the patient group.

Subject no. Age (y) Sex Education (y) Stroke etiology Lesion site Post-stroke onset (m) Aphasia type Aphasia severity

1 41 Male 16 Hemorrhage MCA BG/Sc (F, T, P) 3 Broca’s 2

2 45 Male 16 Hemorrhage MCA BG/Sc (F, P) 3 Anomic 3
3 39 Male 19 Hemorrhage MCA Co (T, P) 6 Conductive 3
4 42 Male 16 Hemorrhage MCA BG/Sc (F, T, P) 3 Mixed 1
5 41 Male 16 Thrombosis MCA BG/Co (F, T, P) 4 Broca’s 1
6 45 Female 12 Thrombosis MCA Co (F, T, P) 3 Broca’s 2
7 47 Female 9 Thrombosis MCA BG/Co (F, T, P) 5 Broca’s 1
8 42 Male 12 Hemorrhage MCA BG/Sc (T, P) 6 Broca’s 3
9 46 male 15 Thrombosis MCA Co (F, T, P) 4 Mixed 3
10 36 Male 15 Thrombosis MCA BG/Co (F, T, P) 6 Broca’s 2
11 37 Male 16 Thrombosis MCA BG/Co (F, T, P) 6 Broca’s 2
12 57 Male 16 Thrombosis MCA BG/Co (F, T, P) 5 Broca’s 2
Mean ± SD 43.2 ± 5.6 14.8 ± 2.6 4.5 ± 1.3 2.1 ± 0.8

BG, basal ganglia; Co, cortical; F, frontal; MCA, middle cerebral artery; P, parietal; Sc, subcortical; T, temporal.
 D. Wu et al. / Neuroscience Letters 589 (2015) 115–120 119

Table 2
PACA assessment of the patient group.

Patient Picture naming Auditory word-picture identification

Baseline T1 T2 T3 Baseline T1 T2 T3

1 13 12 19 17 45 40 49 50
2 27 30 40 42 51 54 54 52
3 21 24 33 34 51 54 53 54
4 10 12 22 24 26 32 48 50
5 7 10 17 18 29 32 38 38
6 29 27 37 44 44 46 52 57
7 0 2 8 12 44 38 46 38
8 31 35 51 49 52 54 55 54
9 30 28 48 51 56 55 57 56
10 12 11 31 30 44 42 46 50
11 18 20 34 32 46 48 52 54
12 22 23 29 29 50 46 54 53
Mean ± SD 18.3 ± 9.6 19.5 ± 9.5 30.8 ± 12.2 31.8 ± 12.2 44.8 ± 8.6 45.1 ± 8.0 50.3 ± 5.0 50.5 ± 6.0

PACA, Psycholinguistic Assessment in Chinese Aphasia. T1, after Phase A1; T2, after Phase B; T3, after Phase A2. Values in cells are either baseline or the number of correctly
named and comprehended items.

left PPR was stimulated, and the NDA results showed an increased
ApEn index in the LWP, LBP, C3, P3, P4, T3 and RWP under the word
repetition condition for the patient group. This indicated that the
activation in Wernicke’s area and Broca’s area under word repeti-
tion was similar to that during picture naming.
The accuracy of picture naming (57.9%) improved more signif-
icantly than that of auditory word picture identification (11.8%)
after Phase B. This may be primarily because the accuracy of pic-
ture naming was much lower than that of the auditory word
picture identification before Phase B, which is indicative of room
for improvement in picture naming.
After A-tDCS, the NDA resulted in an increased ApEn index in the
LWP, LBP, C3, P3, P4, T3 and RWP under the word repetition condi-
tion for the patient group. The increase in accuracy score for picture
naming in all 12 subjects suggested a close correlation between
the state of cortical activation and improvement of picture nam-
ing. Furthermore, cortical excitability that increased significantly
after A-tDCS in both the stimulated area (LWP) and many non-
stimulated language-related areas indicated that other task-related
brain areas also were activated probably through cortico–cortical
connection. Wise et al. [17] have suggested that the junction
between the posterior supratemporal region (LWP) and the inferior
frontal lobe (LBP) acts as a center for binding speech perception and
speech production, or lexical recall. This is relevant for tasks involv-
ing repetition and auditory comprehension. Matsumoto et al. [18]
used electrodes to directly stimulate anterior and posterior corti-
cal regions and record evoked potentials in humans. Language areas
were defined as those which, when stimulated, impaired sentence
reading in the individual patient. Anterior regions included Broca’s
area or adjacent regions and posterior regions included the supra-
marginal gyrus, the middle and posterior superior temporal gyrus,
and the adjacent middle temporal gyrus. As expected, stimulation
of posterior language areas resulted in evoked potentials in ante-
rior language areas, supporting the idea of processing progressing
from posterior to anterior. In the present study, the enhancement
of activation of the stimulated area (PPR) could facilitate the lan-
guage processing from posterior to anterior and also increase the
activation of the anterior region (LBP). The phonological aspects of
repetition are supported by the SLF-tp (the temporoparietal com-
ponent of superior longitudinal fascicle, a pathway connecting the
posterior temporal cortex (T5) with the angular gyrus (P3)), and
the articulatory aspects by the SLFIII (the third component of supe-
rior longitudinal fascicle, a pathway which connects the posterior
inferio-frontal cortex (F7), including ventral premotor cortex (C3),
with the supramarginal gyrus) [19]. These cortico–cortical connec-
tions may facilitate the activation in non-stimulated regions.
The EEG ApEn indices of the right PPR (RWP) and parietal site
(P4) after A-tDCS increased significantly for the aphasic patients.
With support by functional imaging evidence [20], this may indi-
cate that listening to speech activates the superior temporal gyrus
(STG) bilaterally, including the dorsal STG (close to T5, T6) and
superior temporal sulcus (close to P3, P4), which is. No significant
increase was found in P4 and T6 during word repetition for normal
subjects because speech perception in the right hemisphere might
be too rapid to be captured by NDA.
There were no significant differences in the accuracy of pic-
ture naming, and LWP, C3 and P3 maintained their excitability
after Phase A2. This may indicate that the effect of tDCS on

Table 3
EEG ApEn indices of participants under different conditions.

Montage Normal subjects T1 T2 T3

Resting repetition p Resting repetition p Resting repetition p Resting repetition p

F3 0.78 ± 0.09 0.80 ± 0.07 0.188 0.68 ± 0.08 0.68 ± 0.08 0.832 0.68 ± 0.08 0.71 ± 0.07 0.047 0.69 ± 0.06 0.69 ± 0.07 0.683
F4 0.79 ± 0.06 0.77 ± 0.05 0.446 0.79 ± 0.05 0.75 ± 0.07 0.045 0.78 ± 0.06 0.78 ± 0.14 0.831 0.79 ± 0.06 0.78 ± 0.07 0.620
F7 (LBP) 0.78 ± 0.06 0.89 ± 0.06 0.001 0.70 ± 0.09 0.71 ± 0.10 0.538 0.70 ± 0.09 0.74 ± 0.08 0.008 0.72 ± 0.08 0.78 ± 0.13 0.062
F8 (RBP) 0.89 ± 0.10 0.89 ± 0.10 0.877 0.82 ± 0.07 0.75 ± 0.07 0.004 0.80 ± 0.07 0.79 ± 0.07 0.610 0.86 ± 0.13 0.81 ± 0.10 0.168
C3 0.78 ± 0.06 0.84 ± 0.05 0.001 0.68 ± 0.11 0.70 ± 0.10 0.051 0.70 ± 0.10 0.74 ± 0.12 0.008 0.70 ± 0.08 0.75 ± 0.09 0.003
C4 0.85 ± 0.06 0.84 ± 0.07 0.399 0.82 ± 0.06 0.81 ± 0.07 0.385 0.81 ± 0.06 0.82 ± 0.04 0.258 0.80 ± 0.04 0.82 ± 0.05 0.268
P3 0.78 ± 0.05 0.85 ± 0.06 0.002 0.70 ± 0.11 0.74 ± 0.12 <0.001 0.70 ± 0.08 0.75 ± 0.09 0.005 0.69 ± 0.09 0.75 ± 0.08 0.008
P4 0.80 ± 0.05 0.76 ± 0.07 0.066 0.79 ± 0.06 0.78 ± 0.07 0.893 0.77 ± 0.07 0.81 ± 0.07 0.009 0.77 ± 0.07 0.82 ± 0.07 0.163
T3 0.79 ± 0.09 0.90 ± 0.08 0.001 0.77 ± 0.17 0.78 ± 0.17 0.244 0.79 ± 0.16 0.83 ± 0.16 0.010 0.76 ± 0.12 0.84 ± 0.12 0.039
T4 0.90 ± 0.08 0.93 ± 0.08 0.293 0.91 ± 0.11 0.85 ± 0.12 0.006 0.90 ± 0.13 0.89 ± 0.13 0.809 0.90 ± 0.13 0.90 ± 0.17 0.955
T5 (LWP) 0.79 ± 0.07 0.89 ± 0.06 <0.001 0.72 ± 0.14 0.73 ± 0.14 0.191 0.70 ± 0.13 0.80 ± 0.15 <0.001 0.74 ± 0.11 0.80 ± 0.11 0.005
T6 (RWP) 0.82 ± 0.08 0.84 ± 0.06 0.350 0.80 ± 0.09 0.80 ± 0.09 0.829 0.79 ± 0.10 0.84 ± 0.06 0.010 0.81 ± 0.10 0.86 ± 0.08 0.194

Values in cells are mean ± standard deviation. T1, after Phase A1; T2, after Phase B; T3, after Phase A2. See Section 2 for calculations. Bold numbers indicate significant
difference (Significant p value: p < 0.01) between the resting condition and repetition task.
120 D. Wu et al. / Neuroscience Letters 589 (2015) 115–120
cortical excitability beneath the stimulated site and its nearby areas
could last for a minimum of one month.
Moreover, before A-tDCS, the EEG ApEn indices of the patients
were significantly lower (compared with normal subjects) in the
RBP and T4, which indicated that RBP and T4 were depressed during
word repetition. Due to a lack of studies regarding suppression of
the cortical areas under word repetition in post-subacute aphasic
patients, further research is indicated, to elucidate a more detailed
suppression mechanism.
This study has a number of limitations. First, because of the rel-
atively small sample size, the correlation between improvement of
language performance and lesion location and size was not calcu-
lated. Second, there was not long enough post-treatment follow-up
to observe the long-term effect of tDCS. Third, the patients whose
LWP had already activated measured with EEG ApEn index during
word repetition task were excluded in this study. It would be inter-
esting to determine what could happen if such patients continued
to receive A-tDCS over the left PPR.
5. Conclusions
A-tDCS over the left PPR coupled with speech-language ther-
apy in post-acute aphasic patients can improve picture naming
and auditory word-picture identification, and increase cortical
excitability in the language network. The therapeutic effect of tDCS
may be based on the ability of the anodal tDCS to directly increase
cortical excitability under the stimulation electrode and to facilitate
the interconnection with language-related cortical networks. NDA
may reflect cortical activation during language tasks and provide a
method for observing cortical activation during tDCS treatment in
aphasia.
Acknowledgements
This study was supported by two grants from the National
Natural Science Foundation of China(Grant Nos. 81171011 and
81272173), one grant from Science and Technology projects of Bei-
jing (Grant No. Z121107001012144) and one grant from National
Science &Technology Pillar Program (2013BAH14F03).
References
[1] E.K. Kang, Y.K. Kim, H.M. Sohn, L.G. Cohen, N.J. Paik, Improved picture naming
in aphasia patients treated with cathodal tDCS to inhibit the right Broca’s
homologue area, Restor. Neurol. Neurosci. 29 (2011) 141–152.
[2] J.M. Baker, C. Rorden, J. Fridriksson, Using transcranial direct-current
stimulation to treat stroke patients with aphasia, Stroke 41 (2010)
1229–1236.
[3] J. Fridriksson, J.D. Richardson, J.M. Baker, C. Rorden, Transcranial direct
current stimulation improves naming reaction time in fluent aphasia: a
double-blind sham-controlled study, Stroke 42 (2011) 819–821.
[4] A. Floel, M. Meinzer, R. Kirstein, S. Nijhof, M. Deppe, S. Knecht, C. Breitenstein,
Short-term anomia training and electrical brain stimulation, Stroke 42 (2011)
2065–2067.
[5] B. Crosson, K. McGregor, K.S. Gopinath, T.W. Conway, M. Benjamin, Y.L. Chang,
A.B. Moore, A.M. Raymer, R.W. Briggs, M.G. Sherod, C.E. Wierenga, K.D. White,
Functional MRI of language in aphasia: a review of the literature and the
methodological challenges, Neuropsychol. Rev. 17 (2007) 157–177.
[6] X. Zheng, D.C. Alsop, G. Schlaug, Effects of transcranial direct current
stimulation (tDCS) on human regional cerebral blood flow, Neuroimage 58
(2011) 26–33.
[7] B.H. Jansen, Quantitative analysis of electroencephalograms: is there chaos in
the future? Int. J. Biomed. Comput. 27 (1991) 95–123.
[8] W.S. Pritchard, D.W. Duke, Measuring chaos in the brain: a tutorial review of
EEG dimension estimation, Brain. Cogn. 27 (1995) 353–397.
[9] B. Jelles, R.L. Strijers, C. Hooijer, C. Jonker, C.J. Stam, E.J. Jonkman, Nonlinear
EEG analysis in early Alzheimer’s disease, Acta Neurol. Scand. 100 (1999)
360–368.
[10] D.Y. Wu, G. Cai, Y. Yuan, L. Liu, G.Q. Li, W.Q. Song, M.B. Wang, Application of
nonlinear dynamics analysis in assessing unconsciousness: a preliminary
study, Clin. Neurophysiol. 122 (2011) 490–498.
[11] A.D. Friederici, A. Hahne, D.Y. von Cramon, First-pass versus second-pass
parsing processes in a Wernicke’s and a Broca’s aphasic: electrophysiological
evidence for a double dissociation, Brain Lang. 62 (1998) 311–341.
[12] D. Wu, L. Qian, R.D. Zorowitz, L. Zhang, Y. Qu, Y. Yuan, Effects on decreasing
upper-limb poststroke muscle tone using transcranial direct current
stimulation: a randomized sham-controlled study, Arch. Phys. Med. Rehabil.
94 (2013) 1–8.
[13] J. Wang, D. Wu, Y. Chen, Y. Yuan, M. Zhang, Effects of transcranial direct
current stimulation on language improvement and cortical activation in
nonfluent variant primary progressive aphasia, Neurosci. Lett. 549 (2013)
29–33.
[14] S. Pincus, B.H. Singer, Randomness and degrees of irregularity, Proc. Natl.
Acad. Sci. U.S.A. 93 (1996) 2083–2088.
[15] C.J. Price, D. Winterburn, A.L. Giraud, C.J. Moore, U. Noppeney, Cortical
localisation of the visual and auditory word form areas: a reconsideration of
the evidence, Brain Lang. 86 (2003) 272–286.
[16] M.W. Burton, D.C. Noll, S.L. Small, The anatomy of auditory word processing:
individual variability, Brain Lang. 77 (2001) 119–131.
[17] R.J. Wise, S.K. Scott, S.C. Blank, C.J. Mummery, K. Murphy, E.A. Warburton,
Separate neural subsystems within ‘Wernicke’s area’, Brain 124 (2001) 83–95.
[18] R. Matsumoto, D.R. Nair, E. LaPresto, I. Najm, W. Bingaman, H. Shibasaki, H.O.
Luders, Functional connectivity in the human language system: a
cortico-cortical evoked potential study, Brain 127 (2004) 2316–2330.
[19] S.M. Gierhan, Connections for auditory language in the human brain, Brain
Lang. 127 (2013) 205–221.
[20] G. Hickok, D. Poeppel, The cortical organization of speech processing, Nat.
Rev. Neurosci. 8 (2007) 393–402.