Pflügers Archiv - European Journal of Physiology (2021) 473:785–792

https://doi.org/10.1007/s00424-020-02428-8

ORIGINAL ARTICLE

tDCS effects on brain network properties during physiological aging

Fabrizio Vecchio 1 & Francesca Miraglia 1 & Claudia Rodella 2 & Francesca Alù 1 & Carlo Miniussi 2,3 & Paolo Maria Rossini 1 &
Maria Concetta Pellicciari 2

Received: 11 March 2020 / Revised: 5 June 2020 / Accepted: 26 June 2020 / Published online: 4 July 2020
# Springer-Verlag GmbH Germany, part of Springer Nature 2020

Abstract
Brain neural networks undergo relevant changes during physiological aging, which affect cognitive and behavioral functions.
Currently, non-invasive brain stimulation techniques, such as transcranial direct current stimulation (tDCS), are proposed as tools
able to modulate cognitive functions in brain aging, acting on networks properties and connectivity. Segregation and integration
measures are used and evaluated by means of local clustering (segregation) and path length (integration). Moreover, to assess the
balancing between them, the Small World (SW) parameter is employed, evaluating functional coupling in normal brain aging and
in pathological conditions including neurodegeneration. The aim of this study was to systematically investigate the tDCS-
induced effects on brain network proprieties in physiological aging. In order to reach this aim, cortical activity was acquired
from healthy young and elderly subjects by means of EEG recorded before, during, and after anodal, cathodal, and sham tDCS
sessions. Specifically, the aim to exploring tDCS polarity-dependent changes in the age-dependent network dynamics was based
on a network graph theory application on two groups divided in young and elderly subjects. Eighteen healthy young (9 females;
mean age = 24.7, SD = 3.2) and fifteen elderly subjects (9 females; mean = 70.1, SD = 5.1) were enrolled. Each participant
received anodal, cathodal, or sham tDCS over the left prefrontal cortex (PFC) in three separate experimental sessions performed
1 week apart. SW was computed to evaluate brain network organization. The present study demonstrates that tDCS delivered in
PFC can change brain network dynamics, and tDCS-EEG coregistration data can be analyzed using graph theory to understand
the induced effects of different tDCS polarities in physiological and pathological brain aging.

Keywords Graph theory-EEG . Aging . Functional connectivity . Rehabilitation . tDCS

Introduction non-invasive brain stimulation techniques have attracted wide

interest for their ability to modify functional neural activation
In the last years, several efforts have been performed to amelio- patterns and the related cognitive functions in the aging brain
rate cognitive functions and behavioral outcomes in physiologi- [26]. Among the neuromodulation approaches, transcranial direct
cal aging with “neuroenhancement” approaches [6]. Recently, current stimulation (tDCS) represents an additional candidate
tool for maintaining and strengthening cognitive functioning of
the older adults, by promoting neuronal changes in brain activity
This article is part of the special issue on Aging Brain in Pflügers
and connectivity [6].
Archiv—European Journal of Physiology
Regardless of the polarity, tDCS affects neuronal excitabil-
* Fabrizio Vecchio
ity by delivering weak currents to cerebral tissues by means
[email protected]; [email protected] of, at least, two electrodes. In particular, anodal tDCS depo-
larizes the resting membrane potential of the neuron enhanc-
1 ing spontaneous neuronal firing rate and the cortical excitabil-
Brain Connectivity Laboratory, Department of Neuroscience &
Neurorehabilitation, IRCCS San Raffaele Pisana, Via Val Cannuta, ity increase, in the stimulated area, while cathodal tDCS
247, 00166 Rome, Italy causes a shift of the resting membrane potential towards hy-
2
Cognitive Neuroscience Section, IRCCS Istituto Centro San perpolarization, thus decreasing the excitability of the cortex
Giovanni di Dio, Fatebenefratelli, Brescia, Italy and reducing the neuronal firing [8].
3
Center for Mind/Brain Sciences – CIMeC, University of Trento, Albeit recent studies have adopted this technique to mod-
Rovereto, TN, Italy ulate neuronal activity underlying cognitive functions both in

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

786 Pflugers Arch - Eur J Physiol (2021) 473:785–792

young and older adults [5, 25], the tDCS-induced effects at a network graph theory application on two groups of young
network level in physiological aging brain remains to be de- and elderly participants.
fined. A better understanding of the neural changes induced
by tDCS at cortical network level should precede the assess-
ment of its effects at behavioral and cognitive levels [2]. Material and methods
Starting from the well-established role of the prefrontal cortex
(PFC) in executive functions, working memory abilities, and Participants
speed of information processing, PFC could be considered an
ideal cortical target for evaluating the capability of tDCS to induce Eighteen healthy young (9 females; mean age = 24.7 years,
functional connectivity changes in brain aging. SD = 3.2) and fifteen elderly adult participants (9 females;
Considering that oscillatory activities in neural systems mean age = 70.1 years, SD = 5.1) participated in the present
play a key role in orchestrating brain functions and that study. All recruited subjects were right-handed, as evaluated
tDCS may be used to interact with brain oscillations and af- using the Edinburgh handedness inventory test (Oldfield,
fects in polarity-dependent manner specific cortical connec- 1971). Moreover, all participants reported no previous history
tivity patterns [14, 29], we used a network approach to eval- of neurological or psychiatric disorders and had no metal im-
uate age-dependent brain connectivity changes induced by plants. Individual written informed consent was obtained, and
different tDCS polarities. the study was approved by a local ethical committee.
At general level, recent methods based on network science Experimental procedures conformed to the Declaration of
were developed to evaluate brain connectivity changes. The Helsinki and received prior approval by the Ethical
brain tends to be modeled as a complex combination of net- Committee of IRCCS Istituto Centro San Giovanni di Dio,
works by the network science; in mathematical words, a net- Fatebenefratelli. The tDCS protocols were performed in ac-
work is defined by a set of nodes and links between pairs of cordance with safety guidelines procedures [1].
nodes. Physiologically, the brain regions could be represented
by the nodes, while the connections by the links [9] and the Data and code availability statement
connections that can be established between neuronal assem-
blies are the result of segregation and integration processes, as The data adopted by the authors comply with the requirements
mathematically revealed by local clustering (segregation) and of the institute and comply with institutional ethics approval.
path length (integration). Usually, a high degree of local clus- The data that support the findings of this study are available
tering (segregation) and long-distance connections from the corresponding author on request.
(integration) characterizes the network topology of the cere-
bral connections. A model of network organization, the Experimental design
“Small World” (SW) concept, was introduced allowing for
an optimal balance between global integration and local spe- Each participant received anodal, cathodal, or sham tDCS
cialization [32]. The architecture at the base of the brain func- over the left PFC in three separate experimental sessions per-
tional connectivity could be modeled by means of this ap- formed 1 week apart. The order of tDCS sessions was ran-
proach [4], in order to correlate it with behavior (i.e., task- domized and counterbalanced among participants. During
performance). This helps to evaluate if patterns of functional each session, the EEG signals were recorded before (5 min)
connectivity between brain regions reflect the organization of and after (5 min) tDCS, whereas an additional EEG recording
more-or-less strongly connected networks based on the phase was performed for the entire tDCS session (13 min). During
coherence of oscillatory firing synchronizations between the experiment, the participants were seated on a dedicated,
adjacent/remote neuronal assemblies, as shown by their EEG comfortable armchair in a Faraday-cage, sound-proofed room.
activity in a milliseconds time frame [7, 28]. Moreover, they were instructed to keep their eyes open, avoid
Within this theoretical framework, the present study aims blinking, and to look at a stationary fixed point in the center of
to systematically investigate the tDCS-induced effects on a computer screen. The participants were blind to the tDCS
brain network properties during physiological aging. At this conditions. Figure 1 shows the experimental protocol.
aim, EEG activity was acquired from healthy young and older
adult participants before, during, and after anodal, cathodal, EEG recordings and preprocessing
and sham tDCS applied over PFC. In order to explore online
polarity-dependent changes in functional connectivity, a net- Electroencephalographic recordings were performed with a
work approach was used to assess how the tDCS affects the standard montage (Easycap, GmbH, Brain Products) from
brain network, by the evaluation of the SW parameter. 31 electrodes positioned in accordance with the 10–20
Specifically, our aim to explore tDCS polarity-dependent International System (Fp1, Fp2, AF7, AF8, F7, F3, Fz, F4,
changes in the network dynamics during aging was based on F8, FC5, FC1, FC2, FC6, T7, C3, Cz, C4, T8, CP5, CP1, CP2,

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Pflugers Arch - Eur J Physiol (2021) 473:785–792 787

Fig. 1 Experimental protocol

CP6, P7, P3, Pz, P4, P8, PO7, PO8, O1, O2), acquired with a Italy) through two conductive-rubber electrodes. The active
0.1–1000-Hz band-pass filter and digitized at a 5000-Hz sam- (anodal or cathodal) electrode (area:16 cm2) was placed over
pling rate. The midfrontal (Fpz) electrode was used as the the left PFC (in correspondence with F3 position on the EEG
ground one. The online reference for all electrodes was the cap), whereas the other electrode (50 cm2) was fixed extra-
right mastoid electrode (M2), while the left mastoid electrode cephalically on the right arm. For the active condition, the
(M1) was used only offline in order to re-reference the record- current was delivered with an intensity of 1.5 mA for
ings to the average of the two mastoids. Two different chan- 13 min, with a ramping period of 8 s both at the beginning
nels with vertical and horizontal montages monitored the eye and at the end of the stimulation. For sham tDCS, the current
movement. The impedances between skin and electrode were was delivered only for 10 s at the beginning and at the end of
lowered below 5 KΩ. the stimulation time. The terms “anodal” and “cathodal” refer
Data were analyzed by means of scripts from EEGLAB to the polarity of the electrode placed over the left PFC.
(Swartz Center for Computational Neurosciences, La Jolla,
CA), a toolbox of Matlab (Math Works, Natick, MA) soft-
ware. The EEG signals were band-pass filtered from 0.2 to Functional connectivity analysis
47 Hz through a finite impulse response (FIR) filter and down-
sampled to 512 Hz sampling rate. The imported data were The exact low-resolution electromagnetic tomography
fragmented in 2-s duration epochs, and muscular, cardiac, (eLORETA) software was used to carry out EEG functional
ocular, and other types of artifacts were inspected. The proce- connectivity analysis [15, 16, 27, 28]. The eLORETA algo-
dure was the following: (1) an EEG expert reviewed the data, rithm is a linear inverse solution for EEG data without local-
discarding manually the epochs with evident artifactual activ- ization error which is able to provide sources under ideal
ity or with aberrant waveforms, and (2) an independent com- (noise-free) conditions [18]. eLORETA software, in accor-
ponent analysis (ICA) was lastly performed on the 2-s epochs dance with the EEG potential distribution of the scalp, was
to complete the detection and rejection of artifacts using the used to compute a discrete, three-dimensionally (3D) distrib-
Infomax ICA algorithm, which is implemented in the uted linear, minimum-norm, weighted, inverse solution. The
EEGLAB. ICA is a blind source decomposition algorithm weights used in eLORETA enable to localize exactly the
which enables to separate statistically independent sources sources, a property that is necessary to test point sources,
from data acquired from multiple channels. This method is yielding images of the current density with exact localization,
considered as particularly efficient in separating blink and even if with a low spatial resolution (i.e., there is a high cor-
ocular movement artifacts from EEG data. The EEG expert relation between neighboring neuronal sources).
visually inspected the components and, if artifact contamina- Brain connectivity was evaluated by means of eLORETA
tion was found, they were manually rejected. software in 84 regions to obtain a topographic view of the
entire brain, placing the center in the available 42 Brodmann
tDCS areas (BAs: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 17, 18, 19, 20,
21, 22, 23, 24, 25, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38,
The direct current stimulation was delivered by a battery- 39, 40, 41, 42, 43, 44, 45, 46, 47) in the left and right cerebral
driven electrical stimulator (Brain Stim, EMS, Bologna, hemispheres.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

788 Pflugers Arch - Eur J Physiol (2021) 473:785–792

To estimate the electrical brain activity that is used to in- was carried out using the Brain Connectivity Toolbox (BCT,
vestigate cerebral-functional connectivity, the regions of inter- brain-connectivity-toolbox.net), a software instrument that
est (ROIs) are needed. The electric neuronal activities of all was adapted with the Matlab scripts developed in our
voxels belonging to each cortical ROI are averaged in order to laboratory.
get a single signal for each of them, as computed with On the brain networks, the SW parameter was evaluated,
eLORETA. For each brain hemisphere, the intracortical since it measures the balance between global integration and
lagged linear coherence among the eLORETA current density local connectedness of a network, providing a picture of brain
time series of the 84 ROIs, extracted via the method of “all network organization. The ratio of the normalized clustering
nearest voxels” [19], was evaluated between all possible pairs coefficient and the normalized path length constitutes the
of the 84 ROIs for all the seven independent EEG frequency small-worldness measure [22]. Before performing the small-
bands of delta, theta, alpha 1, alpha 2, beta 1, beta 2, and world measurements, a data normalization (i.e., relativization)
gamma (2–4 Hz; 4–8 Hz; 8–10.5 Hz; 10.5–13 Hz; 13– was applied. The clustering coefficient and the characteristic
20 Hz; 20–30 Hz; 30–45 Hz respectively) for each condition path length values were divided by the mean ones obtained by
and participant. the average measure of each parameter in the EEG frequency
Considering the definition of the complex valued coher- bands of each participant, in order to obtain individual mea-
ence between the two time series x and y in the frequency sures. As we computed from weighted networks, it was com-
band ω—that is based on the cross-spectrum given by the plicated to evaluate disgraphs with the same number of con-
variance and the covariance of the signals—the lagged linear nections (all connections were available) and nodes; thus, we
coherence computed in the frequency band ω is computed decided to employ relative values within each frequency band
according to the equation that follows [19]: [3, 21, 30].

½ImCovðx; yÞ2 Statistical evaluation

LagR2xyw ¼
VarðxÞ*VarðyÞ−½ReCovðx; yÞ2
Once extracted a graph pattern with eLORETA from the brain
where CoV and Var are the covariance and variance respec- network, the eLORETA statistical evaluation was performed.
tively of the time series. The Kolmogorov-Smirnov test was used to verify that data
This parameter provides a measure of true physiological normality, and the Gaussianity hypothesis could not be
cerebral connectivity not influenced by low spatial resolution rejected. A statistical ANOVA design for each group of sub-
and volume conduction [19]. The lagged linear connectivity jects (Young and Older adults) was performed for the Small
values computed for each frequency band between all pairs of World between the three factors: stimulation (anodal, cathod-
ROIs constitute the weights of the networks built in the graph al, sham), time (pre, during, post), and frequency band (delta,
analysis. theta, alpha 1, alpha 2, beta 1, beta 2, and gamma) in order to
validate the working hypothesis.
Graph analysis Finally, in order to address a direct match between the two
group of participants, a statistical analysis was evaluated in-
As already defined, a real-world complex system is character- cluding group of participants (Young and Older adults) as
ized by a series of nodes and links between pairs of nodes between factor and stimulation (anodal, cathodal) in the band
(vertices and edges). Cerebral regions usually are represented resulted statistically different in the previous analysis.
by nodes, while functional, anatomical, or effective connec- Furthermore, correlation analyses were carried out consider-
tions [9], depending on the dataset, are links. Normally, ana- ing all participants as a whole group separately for the fre-
tomical connections correspond to white matter fiber tracts quency bands presenting statistical differences between the
between couples of gray matter brain areas (cortical regions two groups of participants in each kind of stimulation.
or subcortical relays). Functional connections reflect the mag-
nitudes of correlations in activity which occur temporarily and Theory
which may occur even between pairs of regions that are not
structurally connected directly. Brain neural networks undergo relevant changes during phys-
A group of elements (vertices) that may be linked by means iological aging, which affect cognitive and behavioral func-
of connections of variable weights (edges) constitutes, from a tions. Currently, non-invasive brain stimulation techniques,
mathematical point of view, a weighted graph. such as transcranial direct current stimulation (tDCS), are
In this study, the undirected and weighted networks were adopted to modulate cognitive functions in brain aging, acting
built (the estimated cortical sources in the BAs constitute the on networks properties and connectivity. This could be used
network vertices) and the edges were weighted by the lagged to address rehabilitation treatment not only in physiological
linear value within each couple of vertices. The graph analysis aging but also in pathological brain aging.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Pflugers Arch - Eur J Physiol (2021) 473:785–792 789

Results Sham stimulation vs real stimulations

Graph theory parameter analyses In young, Duncan-planned post hoc testing showed lower
values in theta (p < 0.000022) band, and higher values in al-
Both the ANOVAs (Fig. 2) for the evaluation of the SW in pha 2 band (p < 0.008954) during anodal tDCS with respect to
Young and Older adults showed statistically significant inter- sham stimulation. Duncan-planned post hoc testing showed
actions (young: F(24, 408) = 2.2962, p = 0.00057; older also lower values in delta (p < 0.000713) and theta
adults: F(24,360) = 2.8435; p < 0.00002) between stimulation (p < 0.000039) bands and higher values in alpha 1
(anodal, cathodal, sham), time (pre, during, post), and fre- (p < 0.005393) and alpha 2 bands (p < 0.001567) during cath-
quency band (delta, theta, alpha 1, alpha 2, beta 1, beta 2, odal tDCS with respect to sham stimulation.
and gamma) factors. While considering older adults, Duncan-planned post hoc
The post hoc tests showed no significant differences in pre testing showed lower values in delta (p < 0.001617) and theta
and post condition, probably also due to the long time distance (p < 0.028757) bands and higher values in alpha 2 band
passed after stimulation [31], while several differences were (p < 0.003434) during anodal tDCS with respect to sham stim-
found during the stimulation, as reported in the following ulation. Duncan-planned post hoc testing showed also lower
paragraphs. values in delta (p < 0.000308) band and higher values in alpha

Fig. 2 ANOVAs for the

evaluation of the SW in young
and older adults

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

790 Pflugers Arch - Eur J Physiol (2021) 473:785–792

1 (p < 0.016822) and alpha 2 bands (p < 0.023037) during Discussion

cathodal tDCS with respect to sham stimulation.
The typical organization of a healthy brain is characterized by
an optimal balance of functional segregation and integration;
Anodal stimulation vs cathodal stimulation SW is the term used to define this kind of scenario. The intri-
cate inhibitory and excitatory cerebral networks consist of
In young, Duncan-planned post hoc testing showed higher areas which are functionally specialized, and that reciprocally
values in alpha 1 band (p < 0.011404) during cathodal with and continuously cooperate to acquire, share, and integrate
respect to anodal tDCS. While considering older adults, any kind of information; all of this reflects the SW features
Duncan-planned post hoc testing showed no differences be- and happens in a constant state of dynamic fluctuations. Our
tween anodal and cathodal stimulation. study demonstrates that tDCS delivered over PFC can change
In order to address a direct match among age, a statistical ongoing network dynamics, in terms of SW properties.
analysis was evaluated including group of participants (young Additionally, our findings support the idea that the graph the-
and older adults, as between factor) and stimulation (anodal, ory can be properly used for analyzing tDCS-EEG
cathodal) in the alpha 1 band. The result showed a statistically coregistration data with the purpose to understand the effects
significant interactions (F(1, 32) = 4.0169, p = 0.049) and induced by different tDCS polarities including in exploring
Duncan-planned post hoc testing showing higher values in age-related brain changes.
cathodal stimulation for young with respect to older subjects Specifically, our results highlight a statistically significant dif-
(p < 0.002797). ference in the SW organization during real tDCS respect to sham
Finally, considering all participants as a whole group, stimulation both in young and older adult participants while spe-
correlation analysis on alpha 1 band showed a negative cific frequency differences during cathodal with respect to anodal
correlation with age (p = 0.0368, r = − 0.3706), namely, stimulation were found only in young participants.
higher the age lower the alpha 1 values during cathodal Particularly, a SW increase in alpha and a reduction in low
stimulation. (delta and theta) frequencies for both groups of subjects were
Summarizing, elderly participants present different effects found for real tDCS respect to sham stimulation. This effect
regarding real tDCS with respect to sham stimulation but no should be explained keeping in mind the physiological role
specific changes induced by different tDCS polarities. that the alpha rhythm plays. The waking rest EEG activity is
Otherwise, young participants present different effects with characterized by alpha frequencies, usually defined as the
respect to tDCS polarity. “idling rhythms” of the adult brain [17]. Alpha has been
proved by several studies to be a brain rhythm involved in
several cerebral functions, ranging from sensory-motor pro-
Functional coupling evaluation cessing to memory formation [23]. Alpha rhythm constitutes
the base of information transmission in healthy subject brains,
Figure 3 reports the functional coupling distribution, as because it works as an oscillatory component of cerebral ac-
revealed by the lagged linear coherence, in the above tivity [11]. Furthermore, studies dealing with event related
significant EEG frequency bands (delta, theta, alpha 1, EEG have found that alpha rhythm is positively correlated
alpha 2) in the two groups of participants during tDCS with the information processing speed and with a good cog-
period in each stimulation. nitive performance [11].

Fig. 3 Functional coupling. An

arbitrary threshold was used to
illustrate these patterns

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Pflugers Arch - Eur J Physiol (2021) 473:785–792 791

Regarding the low-frequency bands, it is affirmed that, Funding information This work was partially supported by the Italian
Ministry of Health for Institutional Research (Ricerca corrente) and for
in a waking state, these EEG rhythms are poorly repre-
the project GR-2011-02349998.
sented, as if alpha and delta rhythms are in a state of
“reciprocal inhibition” [21]. Furthermore, it is well Data availability The data will be available upon request.
known that spontaneous oscillations in the delta frequen-
cies in almost all recorded neurons are generated in case Compliance with ethical standards The data adopted by the
of functional or anatomical disconnection of damaged authors comply with the requirements of the institute and comply with
cortical regions [12]. The found SW decrease in the institutional ethics approval.
low-frequency bands could reflect a sort of structured
behavior consisted of a functional inhibition and an in- Conflict of interest The authors declare that they have no conflict of
interest.
crease in activity. The opposite is true for the alpha band.
Finally, older adults present different effects with respect to
sham tDCS but no specific effects for the polarity of stimula-
References
tion. Otherwise, young participants present different effects
with respect to the tDCS polarities: cathodal stimulation pro- 1. Antal A, Alekseichuk I, Bikson M, Brockmöller J, Brunoni AR,
vides higher effect in alpha 1 than anodal one, namely, higher Chen R, Cohen LG, Dowthwaite G, Ellrich J, Flöel A, Fregni F,
SW values in cathodal are observed. This finding is strongly George MS, Hamilton R, Haueisen J, Herrmann CS, Hummel FC,
in line with the non-linear effects of cathodal tDCS recently Lefaucheur JP, Liebetanz D, Loo CK, McCaig CD, Miniussi C,
Miranda PC, Moliadze V, Nitsche MA, Nowak R, Padberg F,
reported [24]. The high level of alpha 1 SW was also found Pascual-Leone A, Poppendieck W, Priori A, Rossi S, Rossini PM,
negatively correlated with the age of participants, namely, the Rothwell J, Rueger MA, Ruffini G, Schellhorn K, Siebner HR,
higher the age, the lower the SW values in alpha 1 during Ugawa Y, Wexler A, Ziemann U, Hallett M, Paulus W (2017)
cathodal stimulation. The specificity of alpha effect in young Low intensity transcranial electric stimulation: safety, ethical, legal
regulatory and application guidelines. Clin Neurophysiol 128:
participants during cathodal stimulation could be seen as an 1774–1809. https://doi.org/10.1016/j.clinph.2017.06.001
increase in activity and a functional excitation no more present 2. Antonenko D, Schubert F, Bohm F, Ittermann B, Aydin S, Hayek
in aging brain; a mechanism of brain reactivity is still present D, Grittner U, Flöel A (2017) tDCS-induced modulation of GABA
in young participants. levels and resting-state functional connectivity in older adults. J
Neurosci 37:4065–4073. https://doi.org/10.1523/JNEUROSCI.
Albeit recent studies failed to show tDCS polarity- 0079-17.2017
dependent changes in resting-state EEG highlighting the lack 3. Babiloni C, Vecchio F, Lizio R, Ferri R, Rodriguez G, Marzano N,
of specific after effects on cortical activity as evaluated by Frisoni GB, Rossini PM (2011) Resting state cortical rhythms in
spectral power analysis [10], our findings about the tDCS- mild cognitive impairment and Alzheimer’s disease: electroenceph-
alographic evidence. J Alzheimers Dis 26(Suppl 3):201–214.
induced effects at network level are in line with previous re- https://doi.org/10.3233/jad-2011-0051
sults on inter-hemispheric changes [20] and resting-state brain 4. Bassett DS, Bullmore E (2006) Small-world brain networks.
connectivity modulations after tDCS [13], highlighting also Neu roscie ntis t 1 2:5 12 –52 3. ht tps: //doi.org/10.1177/
that the polarity-specific effects of tDCS on cortical connec- 1073858406293182
5. Cespón J, Rodella C, Rossini PM, Miniussi C, Pellicciari MC
tivity at rest could be partly age-dependent. (2017) Anodal transcranial direct current stimulation promotes
As general remarks, further studies could be carried out by frontal compensatory mechanisms in healthy elderly subjects.
non-linear techniques and also to look at other graph parame- Front Aging Neurosci 9:420. https://doi.org/10.3389/fnagi.2017.
ters such as centrality, modularity, hub, sub graphs, and so on 00420
6. Cespón J, Miniussi C, Pellicciari MC (2018) Interventional
to see for example if there are any isolated clusters to be more programmes to improve cognition during healthy and pathological
modulated by the stimulations. ageing: cortical modulations and evidence for brain plasticity.
In conclusion, our results underline that tDCS affects the Ageing Res Rev 43:81–98. https://doi.org/10.1016/j.arr.2018.03.
brain organization of specific functional networks at specific 001
7. de Haan W, Pijnenburg YA, Strijers RL, van der Made Y, van der
frequency bands in a different manner related to the subject Flier WM, Scheltens P, Stam CJ (2009) Functional neural network
age. This might be used to address rehabilitation treatment not analysis in frontotemporal dementia and Alzheimer’s disease using
only in physiological aging but also in pathological brain ag- EEG and graph theory. BMC Neurosci 10:101. https://doi.org/10.
ing in a way that needs further studies. Furthermore, since 1186/1471-2202-10-101
8. Fertonani A, Miniussi C (2017) Transcranial electrical stimulation:
resting-state patterns of brain connectivity are the result of what we know and do not know about mechanisms. Neuroscientist
robust and specific intrinsic neural activity, our results high- 23:109–123. https://doi.org/10.1177/1073858416631966
light that tDCS is able to affect the ongoing spontaneous neu- 9. Friston KJ (2004) Functional and effective connectivity in neuro-
ronal fluctuations; this seems particularly interesting in view imaging: a synthesis. Hum Brain Mapp 2:56–78. https://doi.org/10.
1002/hbm.460020107
of age-related changes and the modulation of the dynamics of 10. Gordon PC, Zrenner C, Desideri D, Belardinelli P, Zrenner B,
the brain at rest to study specific effects in motor or mental Brunoni AR, Ziemann U (2018) Modulation of cortical responses
activities. by transcranial direct current stimulation of dorsolateral prefrontal

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

792 Pflugers Arch - Eur J Physiol (2021) 473:785–792

cortex: a resting-state EEG and TMS-EEG study. Brain Stimul 11: 22. Rubinov M, Sporns O (2010) Complex network measures of brain
1024–1032. https://doi.org/10.1016/j.brs.2018.06.004 connectivity: uses and interpretations. Neuroimage 52:1059–1069.
11. Klimesch W (1999) EEG alpha and theta oscillations reflect cogni- https://doi.org/10.1016/j.neuroimage.2009.10.003
tive and memory performance: a review and analysis. Brain Res 23. Schürmann M, Başar E (2001) Functional aspects of alpha oscilla-
Brain Res Rev 29:169–195 tions in the EEG. Int J Psychophysiol 39:151–158
12. Kooi KA (1922) Fundamentals of electroencephalography. 24. Shilo G, Lavidor M (2019) Non-linear effects of cathodal transcra-
Medical Dept., Harper & Row, New York nial direct current stimulation (tDCS) of the primary motor cortex
13. Luft CD, Pereda E, Banissy MJ, Bhattacharya J (2014) Best of both on implicit motor learning. Exp Brain Res 237:919–925. https://doi.
worlds: promise of combining brain stimulation and brain org/10.1007/s00221-019-05477-3
connectome. Front Syst Neurosci 8:132. https://doi.org/10.3389/ 25. Summers JJ, Kang N, Cauraugh JH (2016) Does transcranial direct
fnsys.2014.00132 current stimulation enhance cognitive and motor functions in the
14. Mancini M, Brignani D, Conforto S, Mauri P, Miniussi C, ageing brain? A systematic review and meta- analysis. Ageing Res
Pellicciari MC (2016) Assessing cortical synchronization during Rev 25:42–54. https://doi.org/10.1016/j.arr.2015.11.004
transcranial direct current stimulation: a graph-theoretical analysis. 26. Tatti E, Rossi S, Innocenti I, Rossi A, Santarnecchi E (2016) Non-
Neuroimage 140:57–65. https://doi.org/10.1016/j.neuroimage. invasive brain stimulation of the aging brain: state of the art and
2016.06.003 future perspectives. Ageing Res Rev 29:66–89. https://doi.org/10.
1016/j.arr.2016.05.006
15. Miraglia F, Vecchio F, Bramanti P, Rossini PM (2016) EEG char-
27. Vecchio F, Miraglia F, Bramanti P, Rossini PM (2014) Human
acteristics in “eyes-open” versus “eyes-closed” conditions: small-
brain networks in physiological aging: a graph theoretical analysis
world network architecture in healthy aging and age-related brain
of cortical connectivity from EEG data. J Alzheimers Dis 41:1239–
degeneration. Clin Neurophysiol 127:1261–1268. https://doi.org/
1249. https://doi.org/10.3233/JAD-140090
10.1016/j.clinph.2015.07.040
28. Vecchio F, Miraglia F, Marra C, Quaranta D, Vita MG, Bramanti P,
16. Miraglia F, Vecchio F, Rossini PM (2017) Searching for signs of Rossini PM (2014) Human brain networks in cognitive decline: a
aging and dementia in EEG through network analysis. Behav Brain graph theoretical analysis of cortical connectivity from EEG data. J
Res 317:292–300. https://doi.org/10.1016/j.bbr.2016.09.057 Alzheimers Dis 41:113–127. https://doi.org/10.3233/jad-132087
17. Niedermeyer E, Lopes da Silva F (1993) Electroencephalography, 29. Vecchio F, Pellicciari MC, Miraglia F, Brignani D, Miniussi C,
basic principles, clinical applications and related fields. Oxford Rossini PM (2016) Effects of transcranial direct current stimulation
University Press, Oxford on the functional coupling of the sensorimotor cortical network.
18. Pascual-Marqui RD (2002) Standardized low-resolution brain elec- Neuroimage 140:50–56. https://doi.org/10.1016/j.neuroimage.
tromagnetic tomography (sLORETA): technical details. Methods 2016.01.051
Find Exp Clin Pharmacol 24(Suppl D):5–12 30. Vecchio F, Miraglia F, Piludu F, Granata G, Romanello R, Caulo
19. Pascual-Marqui RD (2007) Instantaneous and lagged measure- M, Onofrj V, Bramanti P, Colosimo C, Rossini PM (2017) “Small
ments of linear and nonlinear dependence between groups of mul- world” architecture in brain connectivity and hippocampal volume
tivariate time series: frequency decomposition. eprint arXiv: in Alzheimer’s disease: a study via graph theory from EEG data.
0711.1455 Brain Imaging Behav 11:473–485. https://doi.org/10.1007/s11682-
20. Pellicciari MC, Brignani D, Miniussi C (2013) Excitability modu- 016-9528-3
lation of the motor system induced by transcranial direct current 31. Vecchio F, Di Iorio R, Miraglia F, Granata G, Romanello R,
stimulation: a multimodal approach. Neuroimage 83:569–580. Bramanti P, Rossini PM (2018) Transcranial direct current stimu-
https://doi.org/10.1016/j.neuroimage.2013.06.076 lation generates a transient increase of small-world in brain connec-
21. Rossini PM, Del Percio C, Pasqualetti P, Cassetta E, Binetti G, Dal tivity: an EEG graph theoretical analysis. Exp Brain Res 236:1117–
Forno G, Ferreri F, Frisoni G, Chiovenda P, Miniussi C, Parisi L, 1127. https://doi.org/10.1007/s00221-018-5200-z
Tombini M, Vecchio F, Babiloni C (2006) Conversion from mild 32. Watts DJ, Strogatz SH (1998) Collective dynamics of ‘small-world’
cognitive impairment to Alzheimer’s disease is predicted by networks. Nature 393:440–442. https://doi.org/10.1038/30918
sources and coherence of brain electroencephalography rhythms.
Neuroscience 143:793–803. https://doi.org/10.1016/j. Publisher’s note Springer Nature remains neutral with regard to jurisdic-
neuroscience.2006.08.049 tional claims in published maps and institutional affiliations.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

[email protected]