NEURAL REGENERATION RESEARCH www.nrronline.

org

RESEARCH ARTICLE

Age-related changes in resting-state functional

connectivity in older adults
Laia Farràs-Permanyer1, *, Núria Mancho-Fora1, 3, Marc Montalà-Flaquer1, David Bartrés-Faz2, 4, Lídia Vaqué-Alcázar2,
Maribel Peró-Cebollero1, 3, Joan Guàrdia-Olmos1, 3
1 Quantitative Psychology Section, Faculty of Psychology, University of Barcelona, Barcelona, Spain
2 Department of Medicine, Faculty of Medicine and Healthy Sciences, University of Barcelona, Barcelona, Spain
3 Institute of Neuroscience, UB Institute of Complex Systems, University of Barcelona, Barcelona, Spain
4 Institut d’Investigacions Biomèdiques August Pi i Sunyer, Barcelona, Spain

Funding: This study was supported by the Grup de Recerca en Tècniques Estadístiques Avançades Aplicades a la Psicologia (GTEAAP) members
of the Generalitat de Catalunya’s 2014 SGR 326 Consolidated Research Group (GRC) and was made possible by the PSI2013-41400-P project of
Ministerio de Economia y Competitividad of the Spanish Government.

Abstract *Correspondence to:

Laia Farràs-Permanyer, PhD,
Age-related changes in the brain connectivity of healthy older adults have been widely studied in recent [email protected].
years, with some differences in the obtained results. Most of these studies showed decreases in general func-
tional connectivity, but they also found increases in some particular regions and areas. Frequently, these orcid:
studies compared young individuals with older subjects, but few studies compared different age groups 0000-0002-2674-1101
only in older populations. The purpose of this study is to analyze whole-brain functional connectivity in (Laia Farràs-Permanyer)
healthy older adult groups and its network characteristics through functional segregation. A total of 114
individuals, 48 to 89 years old, were scanned using resting-state functional magnetic resonance imaging in doi: 10.4103/1673-5374.255976
a resting state paradigm and were divided into six different age groups (< 60, 60–64, 65–69, 70–74, 75–79,
≥ 80 years old). A partial correlation analysis, a pooled correlation analysis and a study of 3-cycle regions Received: January 23, 2018
with prominent connectivity were conducted. Our results showed progressive diminution in the functional Accepted: March 14, 2019
connectivity among different age groups and this was particularly pronounced between 75 and 79 years old.
The oldest group (≥ 80 years old) showed a slight increase in functional connectivity compared to the other
groups. This occurred possibly because of compensatory mechanism in brain functioning. This study pro-
vides information on the brain functional characteristics of every age group, with more specific information
on the functional progressive decline, and supplies methodological tools to study functional connectivity
characteristics. Approval for the study was obtained from the ethics committee of the Comisión de Bioética
de la Universidad de Barcelona (approval No. PSI2012-38257) on June 5, 2012, and from the ethics com-
mittee of the Barcelona’s Hospital Clínic (approval No. 2009-5306 and 2011-6604) on October 22, 2009 and
April 7, 2011 respectively.

Key Words: brain connectivity; resting state; default mode network; aging; healthy; functional connectivity;
resting state network; age groups

Chinese Library Classification No. R445

Introduction aging (fMRI) in resting-state paradigms. Spontaneous blood

Societal aging is a worldwide phenomenon with implica- oxygen level-dependent (BOLD) signals have been used to
tions in many aspects of life, among which we can identify characterize regional functional connectivity and investigate
social implications and individual consequences. Regarding changes in a variety of neurological and psychiatric disor-
individual implications, various studies have confirmed that ders (Chen et al., 2011). This is a non-invasive method that
aging is linked to a decline in cognitive functioning (Dam- provides an indirect proxy for the “ongoing” brain’s hemody-
oiseaux et al., 2008; Onoda et al., 2012), even if we consider namics (Gorges et al., 2014) and has become a powerful tool
the population of healthy older individuals whose activities to investigate functional coordination between brain areas
of daily living are within the normal range. Studies on cog- (Ystad et al., 2011). In particular, studies involving subjects
nitive decline confirm the shrinkage of global grey matter, in rest (in the absence of external stimuli) can reflect an in-
the breakage of anatomical connections (Raz and Rodrigue, trinsic property of brain functional organization that serves
2006), alterations in the anatomical connectivity between to stabilize brain ensembles, consolidate the past, and pre-
regions (O’Sullivan et al., 2001), and a generalized decrease pare us for future events (Raichle and Snyder, 2007).
in functional connectivity (Damoiseaux et al., 2008; Onoda The most consistently identified resting state network
et al., 2012; Huang et al., 2015). (RSN) is the default mode network (DMN), which is a set
Age-related changes in brain connectivity (both anatom- of different brain regions with decreased activity during
ical and functional) have recently been studied in several several tasks but increased activity during resting (Raichle
articles, especially using functional magnetic resonance im- et al., 2001). The main regions of the DMN are the medial
1544
 Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

posterior frontal cortex, the posterior cingulate cortex, the ger functional connections are mediated by less efficient and
inferior parietal lobe and the hippocampus. Changes in multi-step anatomical paths (Betzel et al., 2014). Longitudi-
DMN activity have been detected in studies related to differ- nal studies showed similar results, demonstrating changes
ent types of cognitive decline pathologies (e.g., mild cogni- in functional connectivity related with age, both increased
tive impairment or Alzheimer’s disease) and other disorders and decreased connectivity depending on brain areas and
such as major depression (Li et al., 2002, 2012; Rombouts et networks, including putamen-occipital and frontal-occipital
al., 2005; Zhou et al., 2008; Binnewijzend et al., 2012). For connectivity (Fjell et al., 2016, 2017). For a recent review of
example, Vemuri et al. (2012) found increased functional aging effects in structural and functional brain connectivity,
connectivity in DMN and other RSNs in Alzheimer’s disease. see Damoiseaux (2017). In summary, the author concludes
In addition, recent studies have found that DMN is also af- that changes, especially functional ones, seem to particularly
fected in cognitively preserved older adults, as well as other affect the DMN, determining that older adults would have
RSNs (Onoda et al., 2012; Huang et al., 2015; Damoiseaux, less connectivity between networks. In addition, the increase
2017). in age-related connectivity seems to indicate cognitive im-
Several studies that compared the connectivity between pairment, either present or in the near future. The results
healthy young adults and healthy older adults found several from whole brain data, and not focused on specific regions,
changes in RSNs. For example, previous studies found in- suggest that those areas involved in DMN play a central role
creased inter-network connections and decreased intra-net- in global connectivity. There is also a lower functional seg-
work connections, probably related to age-related cognitive regation in advanced ages. Finally, the author concludes that
decline (Geerligs et al., 2015; Huang et al., 2015; Grady et the existence of changes in functional connectivity patterns
al., 2016). This increased number of connections in RSNs in relation to age can be reaffirmed.
that occurs in healthy elderly is described as a compensatory Although age is markedly related to changes in functional
mechanism to explain overactivation in connectivity (Seidler and anatomical connectivity, more studies are needed to ex-
et al., 2010), but is also common in neurodegenerative pa- pand upon the characteristics of brain functional connectiv-
thologies (Vemuri et al., 2012). Age seems to be related with ity in healthy older adults since some results from different
functional connectivity reductions, especially in integrated investigations are still inconsistent. Most show decreased
networks and affecting significantly anterior and posterior functional connectivity, but in other cases, increased connec-
components of the DMN (Andrews-Hanna et al., 2007; Ng tivity was observed (especially between-networks functional
et al., 2016). The hippocampus is one of the most affected connectivity) (Damoiseaux, 2008, 2017; Onoda et al., 2012;
regions by this disruption, particularly between the poste- Geerligs et al., 2015; Huang et al., 2015). Additionally, Hir-
rior hippocampus to other hippocampal and DMN regions siger et al. (2016) found no association between functional
(Damoiseaux et al., 2016). Another study explains that aging connectivity strength and age. These differences could be
disrupts not only the DMN but also other RSNs such as the caused by the study characteristics and the approach chosen
sensorimotor (SM) or the visual (V), and these relationships by the authors (such as cross-sectional versus longitudinal or
appeared to be independent from grey matter changes (Ono- whole-brain versus specific brain systems), therefore more
da et al., 2012). Other regions, such as the right frontoinsular research is needed to address these issues. Moreover, to the
cortex, show decreased functional connectivity within the best of our knowledge, a comparison of brain connectivity
DMN in healthy older adults compared to young adults (He amongst different age groups only in older adult populations
et al., 2013). Also, the Dorsal Attention Network exhibited has rarely been approached in the existing literature.
age-related decreases, especially for long-range functional The main hypothesis in our work states that age affects the
connectivity density (Tomasi and Volkow, 2012). Recently, degree of cognitive decline differently, so we intend to com-
Siman-Tov et al. (2017) compared the functional connec- pare connectivity patterns and characteristics in different age
tivity in different age groups (young, middle-aged, and old) groups from middle to advanced age. Therefore, we contem-
and found that connectivity decline was already evident in plated two different objectives. First, we intended to analyze
the middle-aged group, suggesting that aging-related neural the whole-brain functional connectivity of each age group
changes can start early in adulthood. Additionally, Ystad et using a partial correlation analysis and a pooled correlation
al. (2011) found that age-related decline in cognitive pro- analysis. Second, we studied the network characteristics of
cessing speed was correlated to the fiber integrity between the DMN through functional segregation.
the subcortical nuclei, suggesting that widely distributed cor-
tical connections from the participants provide an important Participants and Methods
hub for the distributed resting-state network connectivity. Participants
Related with brain segregation, one study found a decrease The data used in this study included resting-state sequences
in segregation of brain systems defined by their patterns of from 114 healthy individuals aged 48–89 (68.93 ± 7.97) years
resting-state correlations (Chan et al., 2014). In terms of (50% females) from three different studies conducted at the
functional cohesion, Betzel et al. (2014) found that RSNs Department of Medicine, Faculty of Medicine and Health
corresponding to cognitive and sensorimotor functions un- Sciences, University of Barcelona. All participants were inde-
dergo significant age-related refinement, in general becom- pendent in daily life activity.
ing less functionally cohesive with age. Then, with age, stron- The exclusion criteria included illiteracy or an inability
1545
Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

to understand the protocol or undergo neuropsychological tion (Folstein et al., 1975; Tombaugh and McIntyre, 1992).
tests mentioned in the next section, prior cerebrovascular Language ability and verbal IQ were assessed with the
accident, any relevant psychiatric illness, advanced cognitive Boston Naming Test (BNT) (Goodglass et al., 1983), the Vo-
deterioration, dementia, or other neurodegenerative diseases cabulary Subtest of the Wechsler Adult Intelligence Scale 3rd
(e.g., Parkinson’s disease), any chronic illness expected to Edition and the National Adult Reading Test (NART) (Nel-
shorten survival (grave diseases such as heart failure, chronic son, 1982). The WAIS-Voc and the NART can also reflect
liver disease, kidney failure, blood disease or cancer) and any information about premorbid IQ (Lezak et al., 2004), which
MRI-related incompatibility (the presence of metallic objects is considered essential in healthy and cognitively impaired
within the body, pacemaker or claustrophobia). Inclusion older adults (Solé-Padullés et al., 2009; Parra et al., 2013).
criteria were related to the neuropsychological assessment, Verbal memory was assessed with the Rey Auditory Ver-
so they are described in Instruments section. bal Learning Test (RAVLT) (Rey, 1964) or the Grober and
Written informed consent (Additional files 1–3) was ob- Buschke Free and Cued Selective Reminding Test (Grober
tained from each participant prior to taking part in the study and Buschke, 1987) depending on the protocol. These tests
in accordance with the Declaration of Helsinki and approved evaluate memory-related functions such as short-term au-
by the institutional ethics committees. Details of ethics com- ditory-verbal memory, the retention of information, and
mittee agreement and affiliation are included in the section differences between free recall and delayed recall.
MR Image Acquisition.Although the original sample includ- Level of education was also registered in every participant
ed 122 participants, in the present study, individuals with with the following categories: primary school (less than 8
severe movement artifacts or incomplete rs-fMRI time series years of education), secondary school (between 8 and 15
were excluded from the analysis. More specifically, three in- years of education) and university level (more than 15 years
dividuals were discarded because their absolute root mean of education). Every age group of the proposed classification
square movement was above half a voxel (Power et al., 2012) had participants from every level of education.
and five participants had incomplete fMRI recordings. There-
fore, the remaining sample of 114 participants was analyzed. MR imaging
The participants were split into six age groups (< 60, MR image acquisition
60–64, 65–69, 70–74, 75–79, ≥ 80 years) to compare func- All participants were scanned with a Siemens Magnetom
tional connectivity states in a healthy aging process (with the Trio Tim syngo 3-T system at the Centre de Diagnòstic per
following group sizes: n1 = 12; n2 = 21; n3 = 29; n4 = 22; n5 la Imatge of Hospital Clínic (Barcelona, Spain). A high-res-
= 21 and n6 = 9). There were only two participants younger olution T2 and T1-weighted structural image was obtained
than 55 years old in the first group, and only four partici- for each subject with a magnetization-prepared rapid acqui-
pants older than 85 years old. These ranges were selected to sition gradient-echo (MPRAGE) 3-dimensional protocol
detect slight differences in relatively short aging intervals, (repetition time [TR] = 2300 ms, echo time [TE] = 2.98 ms,
and we considered that including these participants would 240 slices, slice thickness = 1 mm and field of view [FOV]
not produce a noticeable impact in the results. = 256 mm). A resting-state fMRI dataset was acquired with
the subjects instructed to lay down with their eyes closed.
Instruments The length of this acquisition was different depending on the
With regard to the neuropsychological assessment, normal sample:
cognitive functioning according to age standardized norms • Protocol 1: n = 32 participants, TR = 2000 ms, TE = 16
was determined through a neuropsychological evaluation, ms, slice thickness = 3 mm, interslice gap = 25%, FOV
including the administration of the Mini-Mental State Ex- = 220 mm, total: 5 minutes. The Ethics Committee from
amination (MMSE), the Boston Naming Test (BNT), the Na- Comisión de Bioética de la Universidad de Barcelona ap-
tional Adult Reading Test (NART), and the Vocabulary Scale proved this study on June 5, 2012 with approval number:
in the Wechsler Adult Intelligence Scale (WAIS-Voc). Partic- PSI2012-38257 (Additional file 4).
ipants from two out of the three protocols were also evaluat- • Protocol 2: n = 59 participants, TR = 2000 ms, TE = 16
ed with the Rey Auditory Verbal Learning Test (RAVLT) and ms, slice thickness = 3 mm, interslice gap = 25%, FOV =
the rest of the individuals were evaluated with the Grober 220 mm, total: 10 minutes. The Ethics Committee from
and Buschke Test (BUSCHKE). These tests were used as Barcelona’s Hospital Clínic approved this study on Oc-
exclusion/inclusion criteria to confirm that participants did tober 22, 2009 with approval number: 2009-5306 (Addi-
not have any cognitive decline in terms of behavioral perfor- tional file 5).
mance. Therefore, the performances were similar among all • Protocol 3: n = 23 participants, TR = 2000 ms, TE = 19 ms,
participants, because all of them must fit the classification slice thickness = 3 mm, interslice gap = 25%, FOV = 220
criteria about absence of cognitive impairment. Consequent- mm, total: 5 minutes. The Ethics Committee from Barce-
ly, the results in these tests have not been used as explanato- lona’s Hospital Clínic approved this study on April 7, 2011
ry variables in any further statistical analysis. with approval number: 2011-6604 (Additional file 6).
The MMSE is a 30-item screening tool designed to evaluate It is easy to see that protocols 1 and 3 recorded 150 dy-
orientations to time and place, immediate and delayed recall, namic points while protocol 2 recorded a total of 300 dy-
attention and calculation, and language and visual construc- namics. This difference between protocols complicates the

1546
 Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

statistical processing of the data but there are two viable in which motion statistics such as DVARS (defined as the
ways to remediate this setback. The first one consists in trun- spatial standard deviation of successive difference images),
cating the temporal registry of protocol 2 and only using the the Framewise Displacement and Jenkinson’s Framewise
first 150 points so that all the input data has the same dy- Displacement (Power et al., 2012) were calculated using FSL.
namics. The other solution to the aforementioned problem This step allows for the computation of the movement of the
is to introduce a statistical weight to all the dynamic points subject during the MRI acquisition and permits discarding
and then globally average all the data (Hunter and Schmidt, of the data if there has been substantial displacement. The
2004). By doing this weighting, despite the protocol 2 has entire preprocessing pipeline was implemented with the FSL
more dynamic points, they will have the same impact as software.
the ones from protocols 1 and 3 on the overall count. In
the present study, we implemented the latter solution and Regions of interest
the data was treated so all the dynamic points had the same The regions of interest (ROI) were defined by the Automatic
statistical weight in the overall sum (see Statistical Analysis) Anatomical Labeling atlas (Tzourio-Mazoyer et al., 2002).
so no impact was expected to the functional connectivity as This atlas contains 90 cortical and subcortical areas, 45 on
this studies involve only a resting state analysis and no other each hemisphere, that are alternatively interspersed (i.e.,
time dependent task was performed. Also a difference in the starting with the left precentral, the right precentral, the left
TE on protocol 3 is reported but it is so slight that no further frontal sup, the right frontal_sup, etc.). To acquire the full
effect was seen on the sample data between protocols. signal of a given ROI, it is necessary to compute an average
The structural T2 images of every participant were revised over the entire time-series of all of the voxels of a given brain
to identify any possible abnormality before including it into area following the Automatic Anatomical Labeling atlas
the statistical analysis. No structural abnormalities or alter- (available by request).
ations were found in any participant.
Statistical analysis
Image preprocessing One of the most recurrent analyses in functional connectiv-
ity is the correlation coefficient matrices computed from the
The structural image data were analyzed using a FSL (FMRIB
fluctuations of the blood oxygen level-dependent (BOLD)
Software Library v5.0, Oxford, UK) preprocessing pipeline
signal. Two methods were used to obtain the functional con-
adapted under authorization from Diez et al. (2015), with its
nectivity: the partial correlation and the pooled correlation.
parameters adjusted to fit our experimental data. T1 images
were reoriented to match the same axes as the templates and
Neuropsychological assessment
a resampled Anterior Commissure-Posterior Commissure
One-way ANOVA was conducted to determine the existence
aligned image with 6 degrees of freedom (df) was created.
of differences in test scores of neuropsychological evalua-
Then, all non-brain tissue was removed to obtain an ana-
tions among age groups using R version 3.5.2 (R Core Team,
tomic brain mask that would be used to parcel and segment
Vienna, Austria).
the T1 data images. The final step involved registering our
structural data images to the normalized space using the Partial correlation
Montreal Neurological Institute reference brain (Ashburner The partial correlation is a matrix of N × N dimensions,
and Friston, 1999). where N = 90 is the number of ROIs that we intend to study
For the purpose of obtaining the functional connectivity and each element of the matrix represents the correlation
matrices, the fMRI images were preprocessed using FSL as coefficient between two given ROIs. Moreover, the truly im-
follows. First, a slice time correction based on the TR of the portant feature of this particular method is the fact that the
image acquisition was carried out to obtain thirty contiguous partial correlation does not consider the contribution by the
slices in the AC-PC plane. The input images were reoriented common neighbors to the correlation coefficient between
to match the template axes and motion correction was com- any two ROIs.
puted to co-register all of the volumes with the central one The partial correlation can be computed from a standard
so that all of the voxels of the different volumes belonged to correlation matrix defined as positive and invertible and it is
the same brain point. Then, all non-brain tissue was removed given by:
and, to get a better signal to noise ratio, the volumes were (1)
smoothed with a 6-mm Full Width at Half Maximum isotro-
pic Gaussian kernel. Also, intensity correction and band pass where p = C-1 is the inverse of the correlation matrix. Ad-
filtering between 0.01 and 0.08 Hz were applied to the data. ditionally, the partial correlation for all of the participants
The resulting functional data images were registered and was computed using the partialcorr function implemented
normalized to the standard MNI space. Finally, the white in MATLAB (R2016b, The MathWorks Inc., Natick, MA,
matter and the cerebrospinal fluid effects were removed so USA).
that no other interference was added to the fMRI signal.
To determine whether the subjects were able to stay still Pooled correlation matrix
during the session, an extra step was added at the pipeline Since that the purpose of this work is to compare the con-

1547
Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

nectivity structures between ROIs between age groups, the

matrices of partial correlations obtained in each participant (4)
presented some differences between them, especially due
to the fact that in some cases the duration of the sessions of where Pij represents the P-value of the ROIs i and j, α is
registration were slightly different. This implied that some of the significance level (0.05) and P represents the number of
the matrices of partial correlations showed slight changes in comparisons to be done.
the final degrees of freedom. To avoid the minimum bias in
the estimation of an average correlation matrix by age group, Density of connections within the RSN regions
instead of using a direct estimation, a correlation matrix For each group, structures of synchronous activation were
weighted by the number of registered brain volumes was detected in the pooled correlation analysis. This subset of
obtained for each age group. The best option to estimate a ROIs comprising DMN regions, visual areas, and sensorim-
unique correlation matrix by age group was to use the Hunt- otor areas was further analyzed through networks analysis,
er and Schmidt (2004) pooled correlation. The pooled cor- using pooled correlation coefficient as network links. Coef-
relation averages the simple Pearson correlation coefficients ficient values greater than 0.2 were drawn to illustrate graph
of the subjects within any of the aforementioned age groups density, while a threshold of r ≥ 0.5 was used to highlight
over all of their time-points, including the effect of number links with higher intensity of functional connectivity.
of observations. Obviously, the pooled correlation matrix
has an N × N dimension, where N = 90 represents the same Network characteristics of the DMN: functional segregation
ROIs as the ones evaluated in the partial correlation. The The functional segregation assumes that brain regions have
general expression is: the ability to develop specialized processing tasks by them-
selves and then integrate all of the information into more
(2) complex processing stages (Friston, 2011). This is possible
because these brain regions are interconnected, forming
where n represents the time-points, r is the correlation groups and clusters in the already known functional net-
coefficient between ROIs i, j and g is the number of subjects works. A simple measure of segregation is the clustering co-
in this particular case. i and j are the row and column indi- efficient, which can be calculated by computing the fraction
cators of the correlation matrix. For instance, r34 refers to the of triangles around a given node of the network. These trian-
correlation between ROIs 3 and 4. k is the participant’s indi- gles, or 3-cycles, represent the nearest functional neighbors
cator. It takes values form 1 to g, where g is the total number of a ROI that are functional neighbors of each other (Rubinov
of individuals in the sample. and Sporns, 2010).
To compute the pooled correlation matrix from the sub- Since we have no anatomical information from tractogra-
jects within a given age group, their simple correlation matri- phy on the subjects under study, we cannot know whether
ces must be homogeneous. This can be tested by computing their brain areas are strongly connected, but the pooled
the Q-test (Cheung and Chan, 2005), which assesses the uni- correlation matrices can be used to infer the intensity of the
formity of the variances among the correlation coefficients ROIs’ signal correlation. Moreover, if we imagine the cor-
within an age group. The Q statistic follows a χ2 distribution relation coefficients between two brain regions as the defini-
with N(N–1)/2 degrees of freedom and is expressed as: tion of the edges of the 3-cycles, we can compute the areas of
these triangles between neighboring ROIs. The areas of the
cycles can be calculated using Heron’s expression, which is:
(3)
(5)
where N is the total number of ROIs under study and is
the mean of the correlation coefficients of each i, j pair of where a, b and c are the side lengths of the triangle and s is
ROIs across the individuals of a given age group given by the semiperimeter, which is:
the letter g, i and j are the row and column indicators of the
correlation matrix. For instance, r34 refers to the correlation (6)
between ROIs 3 and 4. Moreover, the variance matrix was
upper-triangularized, and the main diagonal was not taken A threshold was applied to the 3-cycles to obtain only
into account. k is the participant’s indicator. It takes values those areas whose sides were above 0.6 to get the most repre-
form 1 to g, where g is the total number of individuals in the sentative figure.
sample. S stands for the variance among the correlation coef-
ficients within an age group. Results
The hypothesis of homogeneity is rejected when any of the Neuropsychological results
probability values obtained by testing individual correlation All of the individuals in our sample had scores higher than
coefficients of the matrix are smaller than a significance level 24 in the Mini-Mental State Examination. In addition, no
corrected for multiple comparisons, i.e., Bonferroni (Cheung statistically significant differences were observed between
and Chan, 2005). This condition is given by: the groups as determined by one-way ANOVA in either the
1548
 Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

Table 1 Statistical description of neuropsychological measures between age groups and significance of Q statistic

Age group (years) n BNT NART WAIS-Voc Q P-value

< 60 12 55.50±5.22 25.17±3.81 40.22±12.70 186.076 ≈1

60–64 21 54.81±2.94 24.86±3.45 45.05±11.29 164.121 ≈1
65–69 29 55.59±3.14 26.68±3.02 45.86±10.71 186.428 ≈1
70–74 22 54.50±3.90 25.33±3.62 41.86±9.06 193.353 ≈1
75–79 21 54.24±3.79 24.76±5.02 40.70±7.12 151.876 ≈1
≥ 80 9 48.33±3.12 22.22±5.54 41.00±9.81 169.995 ≈1

Data are expressed as the mean ± SD. BNT: Boston Naming Test; NART: National Adult Reading Test; WAIS-Voc: vocabulary subtest of WAIS.

National Adult Reading Test (F(5, 106) = 1.893, P = 0.102) or specifically, this decrease is accentuated in individuals aged
in the WAIS-Voc (F(5, 101) = 1.092, P = 0.369). Statistically between 65 and 79 years. Interestingly, compared to this
lower scores in the BNT were detected in the oldest group age group, the group of individuals ≥ 80 years old showed
compared to all of the others (F(5, 107) = 3.556, P = 0.005, slightly higher correlation values between brain areas.
Tukey’s HSD adjusted P < 0.001) (Table 1). Figure 2 shows the decreasing pattern described above in
relation to the number of statistically significant correlations
Functional connectivity in every age group. This pattern is maintained in different
Partial correlation threshold values, from P = 0.05 to P = 0.00001.
After deleting the autocorrelations and the anticorrelations,
no pattern was detected through the partial correlation Density of connections within the RSN regions
analysis in any of the groups (it can be seen by request). To further analyze the density of the functional connectivity
Furthermore, the range (r) and median (Md) values for each
age group were calculated: < 60 years old, r = 0.0026, 0.6792, Table 2 ROIs extracted from the AAL atlas and the RSN they belong
to in all age groups
Md = 0.1522; from 60 to 64 years old, r = 0.0100, 0.5840, Md
= 0.1388; from 65 to 69 years old, r = 0.0198, 0.6201, Md = #ROI Region name #ROI Region name
0.1428; from 70 to 74 years old, r = 0.0118, 0.5760, Md =
DMN SM
0.1368; from 75 to 79 years old = 0.0113, 0.5725, Md = 0.1530;
59 Parietal_Sup_L 1 Precentral_L
80 years old or older, r = 0.0003, 0.6767, Md = 0.1463. There-
60 Parietal_Sup_R 2 Precentral_R
by, the correlation between the majority of the regions was
61 Parietal_Inf_L 7 Frontal_Mid_L
proven weak and non-significant, except for some sparse
62 Parietal_Inf_R 8 Frontal_Mid_R
spikes along the matrix’s main diagonal. 85 Temporal_Mid_L 19 Supp_Motor_Area_L
86 Temporal_Mid_R 20 Supp_Motor_Area_R
Pooled correlation matrices DMNa 57 Postcentral_L
As for the homogeneity hypothesis of the correlation co- 29 Insula_L 58 Postcentral_R
efficients, all P-values associated with the Q statistic were 30 Insula_R 63 SupraMarginal_L
approximately 1 (Table 1), which implies that the values of 31 Cingulum_Ant_L 64 SupraMarginal_R
each variance matrix were equivalent. However, given the 32 Cingulum_Ant_R 69 Paracentral_Lobule_L
number of sum terms in the calculation of the Q statistic 87 Temporal_Pole_Mid_L 70 Paracentral_Lobule_R
(df = 4005) may be considered as a more descriptive than 88 Temporal_Pole_Mid_R Visual
inferential approach to estimation of correlation homoge- DMNv 43 Calcarine_L
neity. Altogether, the Q statistic took small values for each 35 Cingulum_Post_L 44 Calcarine_R
age group, which indicates that the variances were small and 36 Cingulum_Post_R 45 Cuneus_L
leads us to conclude that the correlation coefficients were 37 Hippocampus_L 46 Cuneus_R
homogeneous. 38 Hippocampus_R 47 Lingual_L
39 ParaHippocampal_L 48 Lingual_R
Additionally, Figure 1 shows the intensity of the pooled
40 ParaHippocampal_R 49 Occipital_Sup_L
positive correlations between Automatic Anatomical La-
55 Fusiform_L 50 Occipital_Sup_R
beling brain areas for each age group, where two main
56 Fusiform_R 51 Occipital_Mid_L
structures of synchronous activation can be detected in all 65 Angular_L 52 Occipital_Mid_R
groups. Namely, these structures comprise areas from 1 to 30 66 Angular_R 52 Occipital_Inf_L
on the one hand and from 43 to 70 on the other hand, which 67 Precuneus_L 54 Occipital_Inf_R
include several DMN regions. Table 2 shows a detailed clas- 68 Precuneus_R
sification of the ROIs examined and the RSN they belong to.
The analysis of the pooled correlation across the six age AAL: Automatic Anatomical Labeling; RSN: Resting State Network;
DMN: Default Mode Network; DMNa: anterior Default Mode Network;
groups reveals a decreasing pattern in the functional con- DMNv: ventral Default Mode Network; SM: Sensorimotor; #ROI:
nectivity of the brain areas associated with aging. More number of the regions of interest in the atlas; L: left; R: right.

1549
Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

Figure 1 Intensity of pooled positive correlations between Automatic Anatomical Labeling brain areas for each age group.
The color bar of the legend indicates the intensity of the correlations found between the brain regions, with 1 (yellow) representing a full positive
correlation and 0 (deep blue) the lack of correlation between the regions analyzed.

the qgraph (version 1.5) package for R version 3.5.2 (R Core

Team, Vienna, Austria) (Epskamp et al., 2012).
Figure 3 shows a network representation of the functional
connectivity in each age group, where nodes are RSN brain
regions and links are Pearson correlations between ROIs.
Only correlations higher than 0.2 were drawn and correla-
<60 tions equal or higher than 0.5 were drawn in thicker lines.
60–64 The first group of individuals, up to 60 years old, showed
65–69
70–74
the highest density of functional connectivity compared to
75–79 the other groups, especially in the visual and sensorimotor
80+ regions. Links between the visual region and the DMNv and
between the sensorimotor region and the DMN showed high
intensity in most cases. We also found high intensity in the
regions of the DMN between them.
Participants between 60 and 64 years old and between 65
and 69 years old showed similar connectivity patterns related
p-value to the density and the intensity of the connections. A slight
Figure 2 Statistically significant pooled correlations between group
density decrease was observed in all regions, particularly in
and threshold. the sensorimotor region, the DMNv and the DMNa. Less in-
There were 12 participants in the < 60-year-old group, 21 participants tensity was also found in the connections of these structures,
in the 60–64-year-old group, 29 participants in the 65–69-year-old except in the visual system, which has a very similar pattern
group, 22 participants in the 70–74-year-old group, 21 participants
in the 75–79-year-old group, and 9 participants in the ≥ 80-year-old
of density and intensity in those groups.
group. Individuals between 70 and 74 years old exhibited less in-
tensity in their resting-state networks relative to the previous
networks across age groups, we studied the structures that groups but not prominently in the density of the functional
arose in the whole-brain analysis, including the DMN areas connections.
and other sensorimotor (SM) and visual (V) regions in- The group of participants between 75 and 79 years old
volved in the resting state (Table 2). The distinction between showed the most significant changes in the density of the
the ventral DMN (DMNv) and the anterior DMN (DMNa) links and in the intensity, as Figure 3 shows. A remarkable
was determined by previous work with healthy older adults reduction was observed in both cases and this group seemed
by Huang et al. (2015). The non-ventral or the non-anteri- to be the most affected relative to the loss of connectivity
or DMN regions were also considered and included in the intensity and density compared to all of the other groups,
DMN group classification. Graph plots were built through including the oldest group.

1550
 Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

In the oldest group including individuals aged ≥ 80 years, be further discussed. It is important to notice that the older
we found more functional connectivity compared to the oth- group presents fewer cycles of small areas but more cycles of
er groups. The DMNv showed a density in its connections medium and large areas.
similar to that of the 60–64-year-old age group and an in- Furthermore, for the purpose of obtaining a more precise
crease in the sensorimotor areas and the DMN regions com- description of the functional connectivity in each age group,
pared to the previous groups. However, no regions reached we examined the statistical estimators of the triangles be-
the connectivity intensity and density shown by the younger tween the ROIs (Table 3). Although the main descriptors
group, which included participants aged < 60 years. remain invariant across the groups, two parameters present
variability: the number of triangles and the skewness.
Network characteristics of the DMN: functional The first value confirms what we have already stated: there
segregation is a non-negligible density variation of functional connec-
The 3-cycle regions whose edges, i.e., correlation coefficients, tivity in the DMN regions. On the other hand, it is easy to
were over the 0.6 threshold were plotted (Figure 4) to obtain observe that groups three and six show an increase and a
their frequency distribution and to highlight any difference decrease, respectively, in their distribution asymmetry.
among the groups. The Kruskal-Wallis sum rank test high-
light differences in distribution among the age groups (χ2 = Discussion
38.97, df = 5, P < 0.001). At first glance, all groups showed This study has two objectives. First, we intended to analyze
the same tendency in their area distributions, i.e., more tri- the whole-brain functional connectivity of different older
angles of small areas and fewer cycles with larger surfaces. age groups through a pooled correlation analysis. Second,
The number of triangles was not homogeneous across all we studied the network characteristics of the DMN through
groups (χ2 = 147.39, df = 5, P < 0.001), with the group of functional segregation.
individuals between 75 and 79 years old exhibiting a re- Regarding the first objective, the pooled correlation analy-
markable decrease in the overall counts, with evidence to sis showed evidence of a decrease in aging-related functional
connectivity. Specifically, the groups including participants
between 65 and 79 years showed this decrease more marked-
ly as the age group advanced. However, the group of 80-year-
olds and older showed increased functional connectivity
compared to the younger age groups. Partial correlation

Figure 3 Density of functional connectivity in age groups (Pearson

correlation rxy > 0.2, line width threshold: rxy ≥ 0.5). Figure 4 Histogram of the 3-cycle Default Mode Network areas by
G1: Group 1 (< 60 years old, n = 12); G2: group 2 (60–64 years old, n age group.
= 21); G3: group 3 (65–69 years old, n = 29); G4: group 4 (70–74 years The interpretation of brain regions as polygonal vertices makes possible
old, n = 22); G5: group 5 (75–79 years old, n = 21); G6: group 6 (≥ 80 to define the area comprehended between theses DMN brain regions,
years old, n = 9). DMN: Default Mode Network; DMNa: anterior De- which were calculated with Matlab 2016b using Heron’s expression. A
fault Mode Network; DMNv: ventral Default Mode Network; SM: sen- histogram of the area values was performed to highlight the differences
sorimotor; V: visual. Each color shows a different resting state network both in counts and distribution between age groups. The unit for Y axis
included in the study. is count.

1551
Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

Table 3 Statistical estimators of the triangles between ROIs in each age group

Age group (years) N triangles Mean SD Median Min Max Skewness

< 60 393 0.19 0.03 0.18 0.16 0.29 1.21

60–64 376 0.19 0.03 0.18 0.16 0.30 1.18
65–69 489 0.19 0.03 0.19 0.16 0.31 1.99
70–74 493 0.19 0.03 0.19 0.16 0.31 1.02
75–79 205 0.19 0.03 0.18 0.16 0.28 1.24
≥ 80 466 0.20 0.03 0.19 0.16 0.33 0.65

N Triangles: number of triangles. Skewness was calculated by Fisher’s coefficient g1. When skewness is different from 0 it indicates that the
distribution of the variable is not symmetric.

results did not show any pattern in any of the groups. the DMN, the visual system and other networks. They also
With regard to the second objective, we can discuss dif- found increased connectivity between specific regions of
ferent results. The study on the density in the connections different networks that involved some DMN regions. These
inside of the DMN showed a progressive decrease in density results potentially match ours, with marked functional
relative to aging, except for participants of aged 80 years or connectivity decreases that appear early, but with specific
more, who showed denser functional connectivity in their increases in particular regions. We did not notice the de-
RSNs. The study of functional segregation through trian- scribed decrease in the visual network, but we found it in the
gles showed a similar number of triangles in the different other described regions. Therefore, the results of the present
groups, except for participants between 75 and 79 years old, study did not match with that reported by Hirsiger et al.
who manifested an abrupt reduction in the number of tri- (2016), evincing an association between functional connec-
angles. Apart from the frequency of the triangles, their dis- tivity and age. Finally, in terms of functional segregation, our
tributions resembled one another except for the participants results are hardly consistent with those reported by Chan et
aged 80 years or more, which showed less asymmetry than al. (2014), showing an age-related decrease in segregation in
the others. different brain systems.
These findings confirm the results that were observed The present research provides new details on the effects of
previously in other studies. Increased between-networks aging in the estimation of brain functional connectivity. Age
connectivity in older adults had already been identified (Bet- has often been used as a continuum variable in correlation or
zel et al., 2014; Geerligs et al., 2015; Huang et al., 2015; Fjell regression in previous documents, so the use of age groups is
et al., 2016) but not as noticeably in a particular age group new in this area. Studying age-related alterations in connec-
including subjects older than 79 years old. This could be tivity with age-group classifications also contributed to the
explained as a compensatory mechanism or a compensatory clarification of which age shows more noticeable functional
reaction, found in healthy older adults, as well as in mild decreases and when a compensatory mechanism of this con-
cognitive impairment and in the early stages of Alzheimer’s nectivity dysfunction could appear. Therefore, our results
disease. showed a progressive diminution in functional connectivity
Before the age of 79 years, our results showed that every relative to the number of connections and their intensities
age group showed progressive but marked connectivity up to 74 years old. Between 75 and 79 years of age, the most
decline, particularly in intra-networks connectivity. These noticeable decrease appeared, with important reductions in
results are consistent with previous studies that described a intra- and inter-network connectivity and a lower intensity
disruption in the RSNs related with aging (Andrews-Han- of these connections. Finally, from 80 years old onward,
na et al., 2007; Onoda et al., 2012; He et al., 2013; Ng et al., more connectivity than in previous groups was observed
2016) and are partially consistent with other previous studies in our sample. These differences in functional connectivity
(Tomasi and Volkow, 2012; Huang et al., 2015; Siman-Tov et comparing the number of correlations are statistically signif-
al., 2017). Specifically, Huang et al. (2015) found a diminu- icant across different P-values. Therefore, noticeable changes
tion in functional connectivity in the ventral DMN and in related with the number of connections seem to occur while
the sensorimotor and visual systems. Our results showed a age draw on.
marked decrease in the DMNv but not in the sensorimotor The presence of more functional connectivity in partici-
or visual systems. Especially in the visual system, we found pants over 80 years old is probably related to compensatory
a very similar functional connectivity pattern across the dif- mechanisms that have been thoroughly described, but also
ferent age groups and a subtle decrease in the sensorimotor could be explained because of a high degree of resilience
network, particularly in individuals between 60 and 69 years of these participants. Since they have shown a correct per-
old. We also found a slight increase in functional connectiv- formance in neuropsychological assessment and they have
ity in the sensorimotor system and the DMN after 80 years surpassed their life expectancy, another valid reason could
old, which is consistent with the results of Geerlings et al. be just survivability. More specifically, participants older
(2015). In addition, Siman-Tov et al. (2017) found early sig- than 80 years old who are cognitively preserved could reflect
nificant reductions in the functional connectivity between better functional connectivity as our results evidenced. More
1552
 Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

studies regarding this population could help to clarify the first group (below 60 years old) and the last group (above 80
observed differences in functional connectivity. years old) include fewer individuals than the other groups.
The study of functional connectivity and, especially, results Finally, we can identify two limitations related to the sta-
visualization has become a challenge. Studying the whole tistical procedure. On one hand, the threshold applied to
brain implies working with a large amount of data and anal- the 3-cycle segregation analysis was selected to highlight the
ysis techniques should be carefully selected. Therefore, we regions with a Pearson’s correlation value above 0.6 and to
decided to use a 90-ROI atlas to simplify this information avoid selecting those with links that had weaker intensities.
to study the most important brain regions related to rest- However, further research should be conducted on the opti-
ing-state networks. mization of the intensity threshold. On the other hand, it is
Correlation matrices are commonly used in functional important to remember that the Q statistic used in the vari-
connectivity research and density maps are being used pro- ance homogeneity test has limited statistical power. None-
gressively because of their clarity in the identification of con- theless, it is more important to remark that it was not used
nectivity inter- and intra-networks and their intensities (Far- for the purpose of hypothesis testing but rather to quantify
ràs-Permanyer et al., 2015). Therefore, using 3-cycle regions, the variances of correlation coefficients in each group.
we complemented the information from previous results. It should also be noted that this study has some strong
These analyses allowed for more details on the functional points to consider. The use of different analysis strategies re-
connectivity characteristics of our participants, thus comple- inforces all of the results and conclusions obtained, therefore
menting all of the results with more accuracy, specifically in increasing the results’ consistency. In particular, 3-cycle ar-
the DMN network. eas allow for the confirmation of correlation matrices results
As it was mentioned before, the use of age groups is rela- and provide strong information about the characteristics of
tively new in this area. Studies usually distinguish between the functional connectivity in our sample.
young and older individuals, or compare young, middle Regarding the sample, it is important to emphasize that
aged and older adults, as in Siman-Tov et al. (2017), but we this study has a large sample of healthy older individuals.
did not find any article that studied one population, as older This large sample permitted the creation of different age
adults, dividing age groups. The criterion of group classifi- groups from many age ranks, which allowed for the contem-
cation, separating every group by 5 years, was selected for plation of the analysis. Usually, in older populations, partic-
different reasons. One reason was the intention to define ipants present with symptoms of different diseases (physical
groups with a similar number of participants, trying to form and mental) that impede their participation in fMRI re-
them as homogeneous as possible. Another reason was relat- search. Therefore, an exhaustive evaluation has been con-
ed to the aim of finding differences between groups: we had ducted in this study to include only healthy older individuals
to establish a criterion that permitted to spot these differenc- of different ages.
es and it had to be not so narrow (for example, 2 or 3 years) In conclusion, we demonstrated that the characteristics
and not so wide (8 or 10 years). We did not expect to find of brain functional connectivity are related to aging. The
many differences between groups with differences of 2 or 3 age groups showed differences in the number of activated
years, and we expected to find greater differences between regions during resting state and differences in intra- and in-
participants differenced by 10 years. For these reasons, and ter-network connectivity related to the number and intensity
having no examples of previous research with this kind of of the connections. We showed an aging-related progressive
classification, we chose a 5-year distinction between groups. decrease in functional connectivity that was particularly
There are some limitations that merit consideration in this apparent in individuals between 75 and 79 years old, with
study. First, resting-state fMRI includes the possible con- slight increased connectivity after 80 years. These results
founding of automatic bodily activity (for example, cardiac were consistent with previous studies that found age-related
pulsations and respiratory rhythm) (Birn et al., 2006). It is connectivity diminution and the existence of compensatory
important to keep in mind that this is an indirect measure of mechanisms in older adults (Andrews-Hanna et al., 2007;
brain activity, which could involve inaccuracies, and it also Onoda et al., 2012; Geerligs et al., 2015; Huang et al., 2015;
provides much valuable information on brain functioning. Ng et al., 2016; Damoiseaux, 2017; Siman-Tov et al., 2017).
Second, the participants in our study came from three All of the different estimators showed the same pattern. Fi-
different protocols that used almost the same neuropsycho- nally, the present research could expand the understanding
logical tests for assessment, except in the case of the memory of brain functioning in various older age populations, and
examination. Two different tests were used for this evalu- future studies in this area may provide more details about
ation depending on the protocol, namely the RAVLT and the associated characteristics.
the BUSCHKE. Even though memory results could differ,
these tests are admissible as exclusion criteria for individuals Author contributions: Concept of study: JGO; design of study: JGO;
below the thresholds. For this reason, these tests were not literature search: LFP, NMF, MMF; definition of intellectual content:
LFP, MMF, MPC, JGO; data analysis: NMF; data acquisition: DBF, LVA;
taken into account in the later analyses. statistical analysis: LFP, NMF, MMF; manuscript preparation, editing, re-
Another limitation is the sample size, which is not homo- view: LFP, NMF, MMF, MPC; clinical studies: DBF; guarantor: LFP, JGO;
geneous across the groups under study. More specifically, the approval of final manuscript for publication: all authors.

1553
Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

Conflicts of interest: The authors declare that the research was conduct- Chen G, Ward BD, Xie C, Li W, Wu Z, Jones JL, Franczak M, Antu-
ed in the absence of any commercial or financial relationships that could ono P, Li SJ (2011) Classification of Alzheimer disease, mild cog-
be construed as a potential conflict of interest. nitive impairment, and normal cognitive status with large-scale
Financial support: This study was supported by the Grup de Recerca en
Tècniques Estadístiques Avançades Aplicades a la Psicologia (GTEAAP) network analysis based on resting-state functional MR imaging.
members of the Generalitat de Catalunya’s 2014 SGR 326 Consolidated Radiology 259:213-221.
Research Group (GRC) and was made possible by the PSI2013-41400-P Cheung MW, Chan W (2005) Meta-analytic structural equation
project of Ministerio de Economia y Competitividad of the Spanish modeling: a two-stage approach. Psychol Methods 10:40-64.
Government. Damoiseaux JS (2017) Effects of aging on functional and structural
Institutional review board statement: Approval for the study was ob-
tained from the ethics committee of the Comisión de Bioética de la Uni- brain connectivity. Neuroimage 160:32-40.
versidad de Barcelona (approval No. PSI2012-38257) on June 5, 2012, Damoiseaux JS, Beckmann CF, Arigita EJ, Barkhof F, Scheltens P,
and from the ethics committee of the Barcelona’s Hospital Clínic (approval Stam CJ, Smith SM, Rombouts SA (2008) Reduced resting-state
No. 2009-5306 and 2011-6604) on October 22, 2009 and April 7, 2011 brain activity in the “default network” in normal aging. Cereb
respectively. All participants provided written informed consent for their Cortex 18:1856-1864.
inclusion in the present study.
Declaration of patient consent: The authors certify that they have Damoiseaux JS, Viviano RP, Yuan P, Raz N (2016) Differential effect
obtained all appropriate participant consent forms. In the forms, the of age on posterior and anterior hippocampal functional connec-
participants have given their consent for their images and other clinical tivity. Neuroimage 133:468-476.
information to be reported in the journal. The participants understand Diez I, Bonifazi P, Escudero I, Mateos B, Muñoz MA, Stramaglia S,
that their names and initials will not be published and due efforts will be Cortes JM (2015) A novel brain partition highlights the modular
made to conceal their identity, but anonymity cannot be guaranteed.
Reporting statement: This study followed the STrengthening the Report-
skeleton shared by structure and function. Sci Rep 5:10532.
ing of OBservational studies in Epidemiology (STROBE) statement. Epskamp S, Cramer AOJ, Waldorp LJ, Schmittmann VD, Borsboom
Biostatistics statement: The statistical methods of this study were re- D (2012) qgraph: Network visualizations of relationships in psy-
viewed by the professors in biostatistics and the author of this article, Dr. chometric data. J Stat Softw 48:1-18.
Joan Guàrdia Olmos. Farràs-Permanyer L, Guàrdia-Olmos J, Peró-Cebollero M (2015)
Copyright license agreement: The Copyright License Agreement has
Mild cognitive impairment and fMRI studies of brain functional
been signed by all authors before publication.
Data sharing statement: For data sharing, individual participant data, connectivity: the state of the art. Front Psychol 6:1095.
study protocol or informed consent will not be available. However, if there Fjell AM, Sneve MH, Grydeland H, Storsve AB, Walhovd KB (2017)
is any question about the details of statistical analysis plan or other infor- The disconnected brain and executive function decline in aging.
mation in the manuscript that need further explanation, researchers can Cereb Cortex 27:2303-2317.
contact the research group via e-mail to [email protected].
Fjell AM, Sneve MH, Storsve AB, Grydeland H, Yendiki A, Walho-
Plagiarism check: Checked twice by iThenticate.
Peer review: Externally peer reviewed. vd KB (2016) Brain events underlying episodic memory changes
Open access statement: This is an open access journal, and articles are in aging: a longitudinal investigation of structural and functional
distributed under the terms of the Creative Commons Attribution-Non- connectivity. Cereb Cortex 26:1272-1286.
Commercial-ShareAlike 4.0 License, which allows others to remix, tweak, Folstein MF, Folstein SE, McHugh PR (1975) “Mini-mental state”.
and build upon the work non-commercially, as long as appropriate credit
A practical method for grading the cognitive state of patients for
is given and the new creations are licensed under the identical terms.
Open peer reviewer: Gabriele Siciliano, Universita degli Studi di Pisa, the clinician. J Psychiatr Res 12:189-198.
Neurological Clinic, Clinical and Experimental Medicine, Italy. Friston KJ (2011) Functional and effective connectivity: a review.
Additional files: Brain Connect 1:13-36.
Additional files 1–3: Model consent forms 1–3. Geerligs L, Renken RJ, Saliasi E, Maurits NM, Lorist MM (2015) A
Additional files 4–6: Ethical approval documents 1–3.
brain-wide study of age-related changes in functional connectivi-
Additional file 7: Open peer review report 1.
ty. Cereb Cortex 25:1987-1999.
Goodglass H, Kaplan E, Weintraub S (1983) Boston naming test.
References 2nd ed. Philadelphia: Lea & Febiger.
Andrews-Hanna JR, Snyder AZ, Vincent JL, Lustig C, Head D, Gorges M, Müller HP, Ludolph AC, Rasche V, Kassubek J (2014)
Raichle ME, Buckner RL (2007) Disruption of large-scale brain Intrinsic functional connectivity networks in healthy elderly sub-
systems in advanced aging. Neuron 56:924-935. jects: a multiparametric approach with structural connectivity
Ashburner J, Friston KJ (1999) Nonlinear spatial normalization us-
analysis. Biomed Res Int 2014:947252.
ing basis functions. Hum Brain Mapp 7:254-266.
Grady C, Sarraf S, Saverino C, Campbell K (2016) Age differences in
Betzel RF, Byrge L, He Y, Goñi J, Zuo XN, Sporns O (2014) Changes
the functional interactions among the default, frontoparietal con-
in structural and functional connectivity among resting-state net-
trol, and dorsal attention networks. Neurobiol Aging 41:159-172.
works across the human lifespan. Neuroimage 102 Pt 2:345-357.
Binnewijzend MA, Schoonheim MM, Sanz-Arigita E, Wink AM, Grober E, Buschke H (1987) Genuine memory deficits in dementia.
van der Flier WM, Tolboom N, Adriaanse SM, Damoiseaux JS, Dev Neuropsychol 3:13-36.
Scheltens P, van Berckel BN, Barkhof F (2012) Resting-state fMRI He X, Qin W, Liu Y, Zhang X, Duan Y, Song J, Li K, Jiang T, Yu C
changes in Alzheimer’s disease and mild cognitive impairment. (2013) Age-related decrease in functional connectivity of the
Neurobiol Aging 33:2018-2028. right fronto-insular cortex with the central executive and de-
Birn RM, Diamond JB, Smith MA, Bandettini PA (2006) Separating fault-mode networks in adults from young to middle age. Neuro-
respiratory-variation-related fluctuations from neuronal-activi- sci Lett 544:74-79.
ty-related fluctuations in fMRI. Neuroimage 31:1536-1548. Hirsiger S, Koppelmans V, Mérillat S, Liem F, Erdeniz B, Seidler
Chan MY, Park DC, Savalia NK, Petersen SE, Wig GS (2014) De- RD, Jäncke L (2016) Structural and functional connectivity in
creased segregation of brain systems across the healthy adult healthy aging: Associations for cognition and motor behavior.
lifespan. Proc Natl Acad Sci U S A 111:E4997-5006. Hum Brain Mapp 37:855-867.

1554
 Farràs-Permanyer L, Mancho-Fora N, Montalà-Flaquer M, Bartrés-Faz D, Vaqué-Alcázar L, Peró-Cebollero M, Guàrdia-Olmos J (2019)
Age-related changes in resting-state functional connectivity in older adults. Neural Regen Res 14(9):1544-1555. doi:10.4103/1673-5374.255976

Huang CC, Hsieh WJ, Lee PL, Peng LN, Liu LK, Lee WJ, Huang Rey A (1964) Clinical tests in psychology. Paris: Presses Universi-
JK, Chen LK, Lin CP (2015) Age-related changes in resting-state taires de France.
networks of a large sample size of healthy elderly. CNS Neurosci Rombouts SA, Barkhof F, Goekoop R, Stam CJ, Scheltens P (2005)
Ther 21:817-825. Altered resting state networks in mild cognitive impairment
Hunter JE, Schmidt FL (2004) Methods of meta-analysis - Correct- and mild Alzheimer’s disease: an fMRI study. Hum Brain Mapp
ing error and bias in research findings. 2nd ed. Thousand Oas, 26:231-239.
CA: SAGE. Rubinov M, Sporns O (2010) Complex network measures of brain
Lezak MD, Howieson DB, Loring DW (2004) Neuropsychological connectivity: uses and interpretations. Neuroimage 52:1059-
assessment. 4th ed. New York: Oxford University Press. 1069.
Li B, Liu L, Friston KJ, Shen H, Wang L, Zeng L, Hu D (2012) A Seidler RD, Bernard JA, Burutolu TB, Fling BW, Gordon MT, Gwin
treatment-resistant default mode subnetwork in major depres- JT, Kwak Y, Lipps DB (2010) Motor control and aging: Links to
sion. Biol Psychiatry 74:48-54. age-related brain structural, functional, and biochemical effects.
Li SJ, Li Z, Wu G, Zhang MJ, Franczak M, Antuono PG (2002) Alz- Neurosci Biobehav Rev 34:721-733.
heimer disease: evaluation of a functional MR imaging index as a Siman-Tov T, Bosak N, Sprecher E, Paz R, Eran A, Aharon-Peretz J,
marker. Radiology 225:253-259. Kahn I (2017) Early age-related functional connectivity decline
Nelson HE (1982) The National Adult Reading Test (NART): Test in high-order cognitive networks. Front Aging Neurosci 8:330.
Manual. Windsor: NFER-Nelson. Solé-Padullés C1, Bartrés-Faz D, Junqué C, Vendrell P, Rami L,
Ng KK, Lo JC, Lim JKW, Chee MW, Zhou J (2016) Reduced func- Clemente IC, Bosch B, Villar A, Bargalló N, Jurado MA, Barrios
tional segregation between the default mode network and the M, Molinuevo JL (2009) Brain structure and function related
executive control network in healthy older adults: A longitudinal to cognitive reserve variables in normal aging, mild cognitive
study. Neuroimage 133:321-330. impairment and Alzheimer’s disease. Neurobiol Aging 30:1114-
O’Sullivan M, Jones DK, Summers PE, Morris RG, Williams SCR, 1124.
Markus HS (2001) Evidence for cortical “disconnection” as a Tomasi D, Volkow ND (2012) Aging and functional brain net-
mechanism of age-related cognitive decline. Neurology 57:632- works. Mol Psychiatry 17:549-558.
638. Tombaugh TN, McIntyre NJ (1992) The mini-mental state exam-
Onoda K, Ishihara M, Yamaguchi S (2012) Decreased functional ination: a comprehensive review. J Am Geriatr Soc 40:922-935.
connectivity by aging is associated with cognitive decline. J Cogn Tzourio-Mazoyer N, Landeau B, Papathanassiou D, Crivello F,
Neurosci 24:2186-2198. Etard O, Delcroix N, Mazoyer B, Joliot M (2002) Automated
Parra MA, Pattan V, Wong D, Beaglehole A, Lonie J, Wan HI, Hon- anatomical labeling of activations in SPM using a macroscopic
ey G, Hall J, Whalley HC, Lawrie SM (2013) Medial temporal anatomical parcellation of the MNI MRI single-subject brain.
lobe function during emotional memory in early Alzheimer’s Neuroimage 15:273-289.
disease, mild cognitive impairment and healthy ageing: an fMRI Vemuri P, Jones DT, Jack CR Jr (2012) Resting state functional MRI
study. BMC Psychiatry 13:76. in Alzheimer’s disease. Alzheimers Res Ther 4:2.
Power JD, Barnes KA, Snyder AZ, Schlaggar BL, Petersen SE (2012) Ystad M, Hodneland E, Adolfsdottir S, Haász J, Lundervold AJ,
Spurious but systematic correlations in functional connectivity Eichele T, Lundervold A (2011) Cortico-striatal connectivity and
MRI networks arise from subject motion. Neuroimage 59:2142- cognition in normal aging: A combined DTI and resting state
2154. fMRI study. Neuroimage 55:24-31.
Raichle ME, Macleod AM, Snyder AZ, Powers WJ, Gusnard DA, Zhou Y, Dougherty JH, Hubner KF, Bai B, Cannon RL, Hutson RK
Shulman GL (2001) A default mode of brain function. Proc Natl (2008) Abnormal connectivity in the posterior cingulate and
Acad Sci U S A 98:676-682. hippocampus in early Alzheimer’s disease and mild cognitive im-
Raichle ME, Snyder AZ (2007) A default mode of brain function: A pairment. Alzheimers Dement 4:265-270.
brief history of an evolving idea. Neuroimage 37:1083-1090.
Raz N, Rodrigue KM (2006) Differential aging of the brain: Pat- P-Reviewer: Siciliano G; C-Editor: Zhao M; S-Editor: Li CH; L-Editor:
terns, cognitive correlates and modifiers. Neurosci Biobehav Rev Song LP; T-Editor: Jia Y
30:730-748.

1555