Effects of aging on task- and stimulus-related cerebral attention networks

Accepted Manuscript Effects of aging on task- and stimulus-related cerebral attention networks Sophie Kurth, Steve Majerus, Christine Bastin, Fabienne Collette, Mathieu Jaspar, Mohamed Ali Bahri, Eric Salmon PII:

S0197-4580(16)30049-5

DOI:

10.1016/j.neurobiolaging.2016.04.015

Reference:

NBA 9593

To appear in:

Neurobiology of Aging

Received Date: 1 October 2015 Revised Date:

18 April 2016

Accepted Date: 20 April 2016

Please cite this article as: Kurth, S., Majerus, S., Bastin, C., Collette, F., Jaspar, M., Bahri, M.A., Salmon, E., Effects of aging on task- and stimulus-related cerebral attention networks, Neurobiology of Aging (2016), doi: 10.1016/j.neurobiolaging.2016.04.015. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Effects of aging on task- and stimulus-related cerebral attention networks

Sophie Kurtha, Steve Majerusb, Christine Bastina, Fabienne Collettea, Mathieu Jaspard,

RI PT

Mohamed Ali Bahria & Eric Salmona,c

GIGA - Cyclotron Research Center in vivo imaging, University of Liège, Belgium

b

Psychology and Neuroscience of Cognition Research Unit (PsyNCog): University of Liège, Belgium

SC

a

Memory Clinic, CHU Liège, Belgium

d

M AN U

c

Adaptation, Resilience & Change Research Unity, Cognitive Ergonomy & Work Intervention,

TE D

University of Liège, Belgium

Sophie Kurth, GIGA - Cyclotron Research Center, University of Liège, Allée du 6 août (Bat. B30), 4000 Liège, Belgium. Email: [email protected]

EP

Steve Majerus, Psychology and Neuroscience of Cognition Research Unit (PsyNCog): University of Liège, Boulevard du Rectorat (Bat. B33), 4000 Liège, Belgium. Email: [email protected] Christine Bastin, GIGA - Cyclotron Research Center, University of Liège, Allée du 6 août (Bat. B30), 4000 Liège, Belgium. Email: [email protected]

AC C

Mathieu Jaspar, Adaptation, Resilience & Change Research Unity, Cognitive Ergonomy & Work Intervention, University of Liège, Boulevard du Rectorat (Bat. B32), 4000 Liège, Belgium. Email:[email protected] Mohamed Ali Bahri, GIGA - Cyclotron Research Center, University of Liège, Allée du 6 août (Bat. B30), 4000 Liège, Belgium. Email: [email protected] Fabienne Collette, GIGA - Cyclotron Research Center, University of Liège, Allée du 6 août (Bat. B30), 4000 Liège, Belgium. Email: [email protected] Corresponding author: Eric Salmon, GIGA - Cyclotron Research Center, University of Liège, Allée du 6 août (Bat. B30), 4000 Liège, Belgium, Telephone: +32(0)4 3662365, Fax: +32(0)4 3662946, Email: [email protected]

1

ACCEPTED MANUSCRIPT

ABSTRACT

Interactions between a dorsal attention (DAN) and a ventral attention cerebral network (VAN) have been reported in young participants during attention or short term memory (STM) tasks.

RI PT

Since it remains an under-investigated question, age effects on DAN and VAN activity and their functional balance were explored during performance of a STM task. Older and young groups showed similar behavioral patterns of results. At the cerebral level, DAN activation increased as a function of increasing STM load in both groups, suggesting preserved activity

SC

in DAN during healthy aging. Age-related over-recruitment in regions of the DAN in the higher task load raised the question of compensation attempt versus less efficient use of

M AN U

neural resources in older adults. Lesser decrease of VAN activation with increasing load and decreased stimulus-driven activation in the VAN, especially in the higher load, in older participants suggested age-related reduced response in the VAN. However, functional connectivity measures showed that VAN was still functionally connected to the DAN in older

AC C

EP

TE D

participants.

Keywords: Aging; fMRI; Dorsal attention network (DAN); Ventral attention network (VAN); Short term memory (STM).

2

ACCEPTED MANUSCRIPT

I. Introduction Attentional control is not a uniform cognitive function but results from the interaction of multiple cerebral networks and associated cognitive processes. Two attention networks have

RI PT

received particular interest as they appear to have antagonistic effects on attentional control. The dorsal attention network (DAN) is centered on the intra-parietal sulcus (IPS), and connects with the superior parietal lobule and the dorsal frontal cortex; it is considered to

SC

support voluntary orientation of attention in a top-down manner (Asplund et al., 2010; Corbetta and Shulman, 2002; Chica et al., 2011). A second network is the ventral attention

M AN U

network (VAN) centered on the temporo-parietal junction (TPJ) and connecting with the orbito-frontal cortex, inferior frontal gyrus, and anterior insula (Corbetta et al., 2008). Like the ventral parietal cortex more generally (Cabeza et al. 2012), it is considered to be involved in the attraction of attention in a bottom-up manner by salient and novel events in the environment (Corbetta et al., 2008; Todd et al., 2005). To date, these networks have been

TE D

mainly studied in young participant populations. The present study explored the integrity of DAN and VAN and their interactions during healthy aging.

EP

Attentional control is considered as the primary component of working memory capacity in classical working memory theories (Engle and Kane, 2004). More recent models of short term

AC C

memory (STM) consider that controlled, task-related attentional processes are not only linked to executive processes such as updating or manipulation of information in working memory, but are also involved in the maintenance of information in STM (Cowan et al. 2005, 2011, Majerus et al. 2010, 2012). In young participants, this claim is supported at the cerebral level by the observation that part of the neural substrate of STM corresponds to an interaction between the DAN and the VAN (Corbetta et al., 2008; Majerus et al., 2012; Todd and Marois, 2004, Todd et al., 2005). Indeed, the regions consistently activated during verbal and visuospatial STM tasks include the bilateral inferior parietal cortex, the dorsolateral prefrontal 3

ACCEPTED MANUSCRIPT

cortex, as well as premotor and cerebellar areas (Linden et al., 2003; Palesu et al., 1993); these regions partially overlap with the attentional networks (Corbetta et al., 2008; Majerus et al., 2012; Todd and Marois, 2004, Todd et al., 2005). This is particularly the case for the IPS

RI PT

region which is part of the DAN and whose activity has been shown to correlate with attentional and working memory task demands (Todd et al., 2004) and is sensitive to STM load (Ravizza et al., 2004).

SC

Studies by Todd et al. (2004, 2005) and Majerus et al. (2012) have shown that with increasing STM load, the DAN increases its activity while the VAN shows progressive deactivation. A

M AN U

distractor stimulus (DS) presented during a STM maintenance phase further induces stimulusdriven activity in the VAN, but only when participants maintain low load memory lists. This is also in line with some aspects of the perceptual load theory by Lavie et al. (2005). At earlystage perceptual processing, for DS stimuli presented in low perceptual load conditions, this theory predicts that DS will interfere with an ongoing task when occurring in low but not high

TE D

attentional load task contexts. These patterns of results suggest the existence of a trade-off between stimulus-related attention networks, supported by VAN, and task-related attention networks, supported by DAN during performance of a STM task. The increased activation of

EP

the DAN as a function of increasing STM load has been interpreted as reflecting increased demands of task-related attentional control, while the decreased activation of the VAN would

AC C

protect information in STM against interference from task-irrelevant stimuli (Majerus et al., 2012; Todd et al., 2005). Interaction between these two networks has also been shown during attentional processing (Vossel et al., 2014 for a review). Vossel et al. (2012) showed a modulation of connectivity from TPJ to IPS during reorientation of spatial attention. More recently, Leitao et al. (2015) observed causal influences from the IPS to the TPJ during a sustained target detection task using a concurrent fMRI-TMS study design.

4

ACCEPTED MANUSCRIPT

While the ability of older adults to simply maintain items in STM has been shown to be relatively preserved (Craik and Jennings, 1992 for a review), age-related cognitive declines have been reported for different attentional components. A number of studies have

RI PT

documented age-related impairment in tasks involving selective attention (Maylor and Lavie, 1998), divided attention and task switching (Verhaegen and Cerella, 2002) as well as for attentional capture (Zanto and Gazzaley, 2014), although some of these effects appear to be at

SC

least partially mediated by age-related decreases in processing speed (Salthouse, 1996), sensory decline (Lorenzo-Lopez et al., 2002) or task difficulty (Zanto et al., 2014). The effects

M AN U

of aging on the neural correlates of attentional components are still poorly studied and understood. A common observation is the over-recruitment of fronto-parietal regions overlapping with the DAN during performance of various tasks in older populations, such as working memory tasks (Cabeza et al., 2004), visual attention (Cabeza et al., 2004; Madden et al., 2007), and selective attention tasks (Geerligs et al., 2014). This age-related increase of

TE D

activation in frontal and parietal areas has also often been accompanied with weaker occipital activity (Cabeza et al., 2004; Grady et al., 1994, Madden et al., 2007). The increased activation in the frontal lobe has been suggested to reflect increased top-down attentional

EP

(endogenous) processes in order to compensate for the age-related decline in bottom-up (exogeneous) sensory processing (Madden et al., 2007; Spreng et al., 2010a). While most of

AC C

the studies reported here have focused on the effects of aging on the dorsal fronto-parietal network, much less is known about the effects of aging on the ventral fronto-parietal network. Moreover, even if some studies have reported changes in functional connectivity (See Goh, 2011 for a recent review), the issue of age-related effects on network interactions, and more particularly on the functional balance between the DAN and the VAN, remains underexplored.

5

ACCEPTED MANUSCRIPT

The current study explored the effects of aging on the interaction between the DAN and VAN during a verbal STM task based on the paradigm developed by Majerus et al. (2012). Twenty older and twenty young healthy participants were recruited. Two STM load levels (2 and 5

RI PT

items, allowing to assess DAN involvement and VAN deactivation) were used and a DS was presented in some trials in order to study activation of the VAN. Increased activation of the DAN in high versus low memory load conditions was expected in the young and the older

SC

group, based on previous demonstrations of this effect in the literature in young populations (Majerus et al., 2012). Since STM-load related increases of activation have been found in

M AN U

frontal and parietal areas of the DAN in older relative to young participants (Davis et al., 2008; Madden et al., 2007), we further expected increased activation in the DAN in the older group as compared to the young one. With regard to the VAN, a progressive deactivation was expected with increasing memory load in young participants, as well as activation in response to the DS, particularly in the low load condition (Majerus et al., 2012). In older adults, we

TE D

aimed at determining whether the expected STM-load-related decrease and DS-related increase of VAN activity was also present in this participant group. We may expect diminished reactivity of the VAN given previous reports of impaired sensory bottom-up or

EP

stimulus-driven processing in older participant groups (Madden et al., 2007). Finally, in order to further explore the interactions between DAN and VAN activity, we examined functional

AC C

connectivity patterns between the DAN and VAN, as a function of age group. Since changes in functional connectivity have been reported in previous aging studies within networks such as within the fronto-parietal control network (Geerligs et al., 2015) but also between different networks (Geerligs et al., 2015; Goh, 2011), one might expect differences in the functional connectivity between regions of the DAN and the VAN in older adults as compared to younger adults.

6

ACCEPTED MANUSCRIPT

2. Methods 2.1. Participants Twenty young (9 males, mean age: 23.4 ± 8.7) and twenty healthy older right handed

RI PT

participants (9 males, mean age: 74.4 ± 5.6, Dementia Rating Scale (Mattis, 1973): 138.5 ± 3.2) participated in this study. The groups were matched for educational level (mean years of education, young: 15.0 ± 2.0; older: 15.0 ± 2.8; t(38) = -.38, p=.13). Young participants were

SC

recruited from the university community and older participants were recruited from seniors’ organizations and among acquaintances of the experimenters’ relatives and friends. Young

M AN U

and older participants had no history of neurological or psychiatric disorders. The study was approved by the Ethics Committee of the Faculty of Medicine of the University of Liège and was performed in accordance with the ethical standards described in the Declaration of Helsinki (2008). All participants gave their written informed consent prior to their inclusion in

2.2. Task description

TE D

the study.

We used a similar STM probe recognition task as used by Majerus et al. 2012 (See Figure 1).

EP

The encoding phase consisted of a horizontally organized sequence of 2 or 5 consonants (fixed duration: 3250 ms) followed by the appearance of a star in the center of the screen

AC C

indicating the maintenance phase (variable duration: random Gaussian distribution centered on a mean duration of 4000 ± 500 ms). In the retrieval phase (4000 ms), an array of lines (the number of lines corresponded to the number of consonants presented previously) was presented to the participant. A consonant was displayed in one of the positions of the lines and the participant had to decide, by pressing a yes/no button whether this consonant was presented previously and had occurred in the indicated position. In all conditions, there were an equal number of positive and negative probe trials, probing equally all serial positions. In

7

ACCEPTED MANUSCRIPT

half of the trials and for each load, a DS was presented briefly during 60 ms at random time points and at variable locations during the maintenance phase in order to diminish stimulus expectancy. The latter was further reduced by the use of a variable duration of the

RI PT

maintenance phase during which the DS occurred. The stimulus occurred within 9° of fixation. The stimulus was a consonant of a size 67% smaller than consonants of the memory list and the font color was gray. The specific size, font color, and duration parameters of the

SC

DS were chosen so to make the stimulus just noticeable but without further possibility for the participant to check or reanalyze the stimulus, based on similar parameters as in Todd and

M AN U

collaborators (2005). Furthermore, a consonant was chosen since the ventral attention pathway has been shown to react most strongly for unexpected stimuli, which share some features with the target stimulus set (Serences et al., 2005; Anticevic et al., 2010). We furthermore ensured that the DS was never part of the current memory set or the subsequent probe stimulus in order to avoid any priming or interference effect between the DS and the

TE D

target/probe stimuli. A baseline condition, controlling for letter identification, motor response and decision processes, consisted of the presentation of a sequence containing 2—5 identical vowels ordered horizontally, followed by a delay interval (a fixation star of variable duration)

EP

and a response display showing the same letter in one of the 2 or 5 positions. The probe letter was presented in either upper or lower case, and the participants had to decide whether the

AC C

case was the same as in the target list by pressing the button under the index or not by pressing the button under the right finger. The 4 STM conditions (load 2 with DS, load 2 without DS, load 5 with DS, load 5 without DS) and the baseline condition were presented in a single session using an event-related design. There were 26 trials for each STM condition and 20 trials for the baseline condition. The different trials were presented in pseudorandom order, with the restriction that 2 successive trials of the same STM load condition could not be separated by more than 5 trials of a different condition in order to keep blood oxygen level-

8

ACCEPTED MANUSCRIPT

dependent (BOLD) signals for same condition epochs away from the lowest frequencies in the time series. Before the start of a new trial, an exclamation mark appeared on the center of the screen during 1000 ms, informing the participant about the imminent start of a new trial. The

RI PT

duration of the inter-trial interval was also variable (random Gaussian distribution centered on a mean duration of 2000 ± 200 ms) and further varied as a function of the participants’ response times: the probe array disappeared immediately after pressing the response button

SC

followed by the presentation of the next trial. If the participant did not respond within 4000 ms, a ‘‘no response’’ was recorded and the next trial began. Both response accuracy and

M AN U

response times were collected. Participants were instructed to respond as accurately and quickly as possible and to stare on the fixation star. Finally, a practice session outside the magnetic resonance environment, prior to the start of the experiment, familiarized the participants with the specific task requirements and included the administration of 12 practice

2.3. Imaging procedure

TE D

trials; no DS was presented during the practice trials.

All imaging data were collected on a 3-Tesla head-only Siemens Allegra scanner (Siemens, Allegra, Erlangen, Germany) with the standard transmit-receive quadrature head

EP

coil. T2*-weighted functional images were acquired using a gradient-echo EPI sequence with TR = 2040 ms, TE = 30 ms, flip angle = 90°, FoV = 192 x 192 mm2, matrix size = 64 x 64,

AC C

voxel size = 3 x 3 x 3 mm3. Thirty-four 3-mm thick transverse slices were acquired, with an interslice gap of 25%, covering the whole brain. The first three volumes were discarded to allow for magnetization equilibrium. Gradient-recalled sequences were applied directly after the functional sequences to acquire two complex images with different echo times (TE = 4.92 and 7.38 ms respectively) and generate field maps for distortion correction of the EPI images. A high-resolution T1-weighted image was acquired for each participant with the T1-weighted 3D MPRAGE sequence with TE = 4.35 ms, TR = 1960 ms, TI = 1100 ms, field of view = 230 9

ACCEPTED MANUSCRIPT

x 173 mm2, resolution = 256 x 192 x 176, voxel size = 0.9 x 0.9 x 0.9 mm3. Head movement was minimized by restraining the participant’s head using foam pads. In each session, between 730 and 803 functional volumes were obtained. Stimuli were displayed on a screen

mirror mounted on the standard head coil. 2.4. Functional magnetic resonance imaging

SC

2.4.1. Preprocessing

RI PT

positioned at the rear of the scanner, which the participant could comfortably see through a

Imaging data were preprocessed and analyzed using SPM8 (Wellcome Department of

M AN U

Imaging Neuroscience, http//www.fil.ion.ucl.ac.uk/spm) implemented in MATLAB 7.12. (Mathworks Inc., Sherborn, MA). For each participant, EPI time series were corrected for motion and distortion using Realign and Unwarp (Andersson et al., 2001) together with the FieldMap toolbox (Hutton et al., 2002) in SPM8. Functional scans were realigned using rigid body transformations, iteratively optimized to minimize the residual sum of squares between

TE D

the first and each subsequent image separately and a mean realigned functional image was then calculated by averaging all the realigned functional scans. This mean functional image was coregistered to the structural T1-image using a rigid body transformation optimized to

EP

maximize the normalized mutual information between the two images. The resulting coregistration parameters were then applied to the realigned functional images. The mapping

AC C

from participant to MNI space was estimated from the structural image using the “unified segmentation” approach (Ashburner and Friston, 2005). The warping parameters were then separately applied to the functional and structural images to produce normalized images with resolutions of 2 x 2 x 2 mm3 and 1 x 1 x 1 mm3, respectively. Finally, the warped functional images were spatially smoothed using a Gaussian kernel of 8 mm full-width at half maximum (FWHM).

10

ACCEPTED MANUSCRIPT

2.4.2. Functional analyses For each participant, brain responses were estimated at each voxel using a general linear model with epoch and event-related regressors. The model assessed sustained activity over the

RI PT

whole STM trials as a function of STM load, and the epoch regressors ranged from the time of the onset of each trial until the participant’s response. This model also included eventrelated regressors assessing transient activity associated with the presentation of the DS as a function of the STM load of the trial within which the DS occurred. This model was used to

SC

explore overall effects of STM load on DAN and VAN, independently of the STM phase. In

M AN U

this model, the baseline condition was modeled implicitly meaning that any activation reported in this study is controlled for baseline activation. Boxcar functions representative for each regressor were convolved with the canonical hemodynamic response. The design matrix included four events assessing the effects of load and DS: load 2, load 5, presence of DS in load 2 and presence of DS in load 5. It also included the realignment parameters to account

TE D

for any residual movement-related effect. A high-pass filter was implemented using a cutoff period of 128 s in order to remove the low-frequency drifts from the time series. Serial autocorrelations were estimated with a restricted maximum likelihood algorithm with an

EP

autoregressive model of order 1 (+white noise). A series of linear contrasts were performed at the individual subject level: first, the brain areas activated in the high vs low load memory

AC C

condition and the brain areas activated in the low vs high load condition were performed in order to test the activation in the DAN and deactivation in the VAN networks. Secondly, the activity associated to the appearance of a DS in the low vs high load condition and in the high vs low load condition in order to test the VAN was assessed. The resulting set of voxel values constituted a map of t statistics (SPM[T]). These contrast images were then smoothed again (6-mm FWHM Gaussian kernel) in order to reduce remaining noise due to intersubject differences in anatomical variability in the individual contrast images. Smoothing by 8 mm (at

11

ACCEPTED MANUSCRIPT

the first level) and then by 6 mm leads to a single equivalent smoothing kernel of 10 mm (as 100 = 64 + 36), a common value for multiple subject analysis. Given the linear nature of the general linear model used here, smoothing can be applied at any stage of processing. The use

RI PT

of a 2-step smoothing procedure was justified by the fact that we used low levels of smoothing for the estimation of the data at the single-subject level; these data were used for the calculation of individual beta parameter estimates reported. The additional smoothing by 6

SC

mm then allowed us to attain the more common higher levels of smoothing for group-level analyses. In order to compare both groups, the contrast images were then entered in second-

M AN U

level analyses, corresponding to two sample T-test random effect models. As a rule, statistical inferences were performed at the voxel level at p< 0.05 corrected for multiple comparisons across the entire brain volume using random field theory (Worsley et al., 1996b). Moreover, in order to directly test hypotheses about the involvement of the DAN and VAN, region of interest analyses were conducted by selecting a 10-mm radius sphere around the averaged

TE D

published coordinates for location of interest (see table 1) and by conducting statistical analysis directly on these regions with additional small volume corrections (Worsley et al., 1996a).

EP

2.4.3. Psychophysiological interactions (PPI)

AC C

In order to directly explore functional connectivity and to assess group effects in the DAN and VAN interactions, we conducted psychophysiological interaction (PPI) analyses. Differential connectivity patterns as a function of high versus low load STM conditions in each group (which represent the main experimental manipulation of this study) were explored. One drawback of PPI analyses is their sensitivity to the choice of the source region. The seed region was therefore carefully selected. The logic underlying the choice of the right and the left IPS as region of interest (ROI) for these analyses was that it could be informative about functional connectivity between regions of the DAN but also about inverse relationship with 12

ACCEPTED MANUSCRIPT

regions belonging to the VAN. The regions were chosen from the results of the previous conjunction analysis which showed common activation to both groups in right and left IPS in the high versus low load contrast.

RI PT

The cerebral activity for the two ROIs (left and right IPS) was extracted using a spherical 10 mm-radius for each volunteer. A GLM was used to perform PPI analyses. At the first level of the analyses (fixed effect), three regressors were created. The first regressor represented the

SC

psychological condition (High versus low load contrast). The second one was the activity in the seed area. The third regressor represented the interaction of interest between the first

M AN U

(psychological) and the second (physiological) regressors. The contrast images obtained allowed us to determine, in each participant, the brain areas interacting significantly with the left and right IPS with respect to the psychological regressor. The contrast images were used at the second level (random-effect analysis) for between-group comparisons. All consistent PPI results are presented at a cluster level with a threshold of p<0.05 corrected for multiple

TE D

comparisons. When regions of interest were not significant at this level, a small volume correction (Worsley et al., 1996a) was computed on a 10-mm radius sphere around the averaged coordinates published for the corresponding location of interest (See Table 1). The

AC C

3. Results

EP

results obtained will be discussed in terms of functional connectivity.

3.1. Behavioral data

Response accuracy was assessed via a 2 (STM load) by 2 (DS, no-DS) by 2 (Group) ANOVA, with load and DS as repeated-measure factors. Results revealed a main effect of group (F(1,38)=13.91, p<.001, η2 P =.27) with young participants having a higher correct response rate than older ones (See Figure 2). No significant main effect of load (F(1,38)=0.58, (p=0.45, η2 P =.02), and DS (F(1,38)=0.86, (p=0.36, η2 P =.01) were observed. There was however a significant interaction between load and distractor (F(1,38)=5.86, p<.05, η2 P =.13). 13

ACCEPTED MANUSCRIPT

Planned comparisons showed that in the low load condition, performance was significantly better in the condition without distractor (F(1,38)=5.67, p<.05, η2 P =.13). There was no significant difference in the high load condition (F(1,38)=0.74, (p=0.30, η2 P =.03). There was

RI PT

no significant interaction with the group (F(1,38)=0.56, p=0.46, η2 P=.01). Response times on correct responses (see Figure 2) were submitted to the same analyses and revealed a main effect of group (F(1,38)=34.22, p<.001, η2 P =.47), young participants being

SC

significantly faster than older participants; a main effect of load (F(1,38)=361.68, p<.001, η2 P =.90) with low load items being processed faster than high-load ones; and a significant load P

=.11) with older participants being

M AN U

by group interaction (F(1,38)=5.17, p<.05, η2

significantly slower in the high load condition as compared to the low load condition than young participants. However, after controlling for age-related slowing by using a z-score transform of the raw reaction time data (Faust et al., 1999), this interaction was not significant anymore. There was no main effect of DS (F(1,38)=0.02, p=0.9, η2 P =.00), but there was a

TE D

significant load by DS interaction (F(1,38)=11.79, p<.05, η2

P

=.23). Participants were

significantly slower (F(1,38)=7.04, p<.05, η2 P =.16) in the low load condition with DS than without DS; the effect of the DS was not significant in the high load condition, although it

EP

approached significance (F(1,38)=3.47, p=0.07, η2 P =.08); if any, this marginal effect went in the opposite direction, with response times tending to be slower in the absence of a DS. There

AC C

was no significant interaction with group (F(1,38)=0.05, p=0.83, η2 P =.01). 3.2. Imaging data

3.2.1. Effects of STM load In order to test load effects common to both groups, a conjunction analysis was performed for the high (5) versus low (2) load contrast in the young and older groups. The results of the conjunction analysis showed common, positive effects of STM load in a large set of regions 14

ACCEPTED MANUSCRIPT

extending from the occipital lobe to the superior parietal lobule, and including bilateral IPS regions. Other regions typically activated during short term memory (SMA, IFG) were also evidenced (see table 2 and figure 3). A region of interest analysis confirmed activation in

RI PT

regions previously reported to belong to the DAN (bilateral IPS, MFG) for the high versus low load condition (see table 3).

Contrasts looking at group differences (see tables 4 and 5) showed that the pre-central cortex

SC

was recruited to a higher extent in the high load condition for older than for younger participants. This region has also been identified as the FEF region in other studies and is

M AN U

considered to be part of the DAN (Corbetta et al., 2002). Region-of-interest analyses did not reveal group effects in other regions of the DAN. In order to test the existence of a compensatory function of the DAN in older adults, correlations between performance (accuracy and reaction times) and the magnitude of event-related activation (β estimates) in the DAN (right and left IPS, FEF) were performed for the both groups at each load level.

TE D

These correlations elicited no significant result in any of the two groups. Strong group effects were also observed in occipital regions (see Table 4). Occipital regions,

EP

and predominantly the lingual gyrus (10 -86 0), were significantly less activated in older than in younger participants in the high versus low load condition. An analysis of beta parameters

AC C

for this lingual region showed that it was deactivated in the low load condition and activated in the high load condition, this effect being more important in young participants (See figure 4). A complementary ANOVA conducted on these parameters confirmed a significant load by group interaction (F(1,38)=21.79, p<.001, η2 P =.36). The reverse comparison (low load versus high load) aimed at exploring the progressive deactivation of regions of the VAN with increasing load. No significant results were observed for the conjunction analysis of the activations in this contrast over the two

15

ACCEPTED MANUSCRIPT

groups. When conducting region of interest analyses separately for each group, an expected negative effect of STM load in a region associated with the VAN (left TPJ) was observed in the young participant group (see table 6). In older participants, the low versus high memory

RI PT

load contrast did not elicit any significant result. Analyses of group effects confirmed differential load-dependent sensitivity of the left TPJ in young versus older participants. This effect was explored by an analysis of beta parameters confirming that the left TPJ region

SC

associated with the VAN was more deactivated with increasing memory load in young than older participants (See figure 5). A complementary ANOVA conducted on these parameters

M AN U

showed a significant load by group interaction (F(1,38)=5.25, p<.05, η2 P =.12). Planned comparisons showed no significant group effect in the low load but a significant group effect for the high load condition (F(1,38)=4.25, p=0.05, η2 P =0.10), confirming stronger TPJ deactivation in young than in older participants. In order to gain more insight into the functional significance of this latter result, correlations between performance (accuracy and

TE D

reaction times) at each load level (2 and 5) and the magnitude of event-related activation (β estimates) were performed in the left and right TPJ belonging to the VAN. These correlations elicited no significant result in any of the two groups.

EP

3.2.2. Effects of DS

AC C

In young participants, the main effect of the DS elicited stronger occipital activations in the low load versus the high load condition; no significant activation was found in older participants for this contrast (see Table 7). The reverse comparison exploring the effect of DS in the high vs low load conditions did not elicit any significant result neither in the young nor in the old group. Analyses of group effects confirmed that in response to the DS in the low vs high load conditions, occipital regions were more activated in young than older participants (see Table 7). Analysis of beta parameters showed that for young participants, there was an important lingual deactivation in response to the DS in the high load condition (see Figure 6). 16

ACCEPTED MANUSCRIPT

A complementary ANOVA conducted on these parameters confirmed a significant load by group interaction (F(1,38)=21.03, p<.001, η2

P

=.36). Planned comparisons showed no

significant group effect in the low load but a significant group effect for the high load

RI PT

condition (F(1,38)=24,77, p=0.001, η2 P =0.39). To further explore VAN activation in response to the DS, we ran contrasts for both load levels separately. For young participants, when the DS appeared in the low load

SC

condition (see table 8), a large set of regions belonging to the VAN (bilateral TPJ and OFC) but also regions associated with STM maintenance processes (IFG, postcentral or bilateral

M AN U

anterior IPS) and thalamic regions were activated, whereas older participants only activated thalamic and midbrain regions. These group differences were however not statistically significant. In reaction to the DS in the load 5 condition, young participants activated the VAN (bilateral TPJ, left OFC), regions involved in STM maintenance (anterior IPS), and the posterior precuneus, whereas no significant activation was found in older participants in

TE D

response to the DS in the load 5 condition (see table 8). A VAN region (lTPJ) was significantly more activated in young participants than in older participants when the DS appeared in the load 5 condition (see Figure 7). A complementary ANOVA conducted on

EP

these parameters showed a marginally significant effect for the load by group interaction (F(1,38)=3,88, p=.056, η2 P =.0.09). Planned comparisons showed no significant group effect

AC C

in the low load but a significant group effect for the high load condition (F(1,38)=15,64, p=0.001, η2 P =0.29). The higher activation of VAN regions for load 5 in young compared to old participants was also supported by region of interest analyses (see Table 9). 3.2.3. Psychophysiological interactions (PPI) As we have shown previously, the DAN seems to be preserved while the VAN is less reactive in older adults. We wondered whether the functional balance between DAN and VAN

17

ACCEPTED MANUSCRIPT

observed in younger adults would still be present in older ones. We therefore performed PPI analyses in order to obtain functional connectivity measures between regions of the DAN (left and right IPS) and the rest of the brain, and more particularly with regions belonging to the

RI PT

VAN. We observed that there was no significant group difference in the connectivity with regions of the DAN. However, when exploring the results in each group separately (See Table 10), we observed that, in young adults, seed regions (more particularly the left IPS) were

SC

negatively correlated with regions of the VAN (left and right TPJ) but also with regions belonging to the default mode network-DMN (precuneus and posterior cingulate cortex). In

M AN U

older adults, only when using small volume corrections with regions of the VAN as ROIs, we found significant negative correlations between DAN (left and right IPS) and VAN (left TPJ) regions. 4. Discussion

The purpose of this study was to explore the effects of aging on two attentional networks, the

TE D

DAN and the VAN and their interactions during a STM task. At the behavioral level, older and young groups showed similar patterns of results, STM performance decreasing with increasing STM load, and STM performance being negatively impacted by the occurrence of

EP

a distractor (DS), specifically during the low load STM condition. This load by distractor interaction has been found in previous behavioral and fMRI studies (Majerus et al., 2012,

AC C

Todd et al., 2005). This effect has been interpreted as reflecting an inattentional blindness effect: high task-related attentional load leads to maximal recruitment of attentional resources by the on-going task, preventing DS to get consciously detected; when attentional load is lower, DS can be processed by the still available attentional resources. At the neuroimaging level, this has been shown to be associated by increased activation of the DAN – involved in task-related attentional processing – and relative deactivation of the VAN – involved in stimulus-driven processing of distractor information in tasks involving attentional load (Todd 18

ACCEPTED MANUSCRIPT

et al., 2004, 2005; Majerus et al., 2012). In line with these studies, DAN activation increased and VAN activation decreased as a function of increasing STM load in young participants. In older participants, increased DAN was also associated with increasing memory load but, at

RI PT

the same time, the VAN was significantly less deactivated than in young participants. When a DS occurred during the STM maintenance phase, the VAN was activated in young participants as it was expected according to previous studies (Todd et al., 2005, Majerus et al.,

SC

2012). Older participants differed from young participants by showing significantly less activation of the VAN as a result of the occurrence of the DS, and this specifically when the

M AN U

DS occurred in the higher STM load context. Results of PPI analyses showed no significant group difference in functional connectivity measures. In young participants, the DAN was inversely interacting with the VAN but also with regions belonging to the default mode network (DMN). This latter result is consistent with the often reported anticorrelation between a task positive and a task-negative network (DMN) during resting state but also during task

TE D

performance (Fox et al., 2005). In older participants, even if the VAN was less reactive, it was still inversely interacting with the DAN, suggesting that the functional interaction between both networks is still present with aging.

EP

4.1. Effects of aging on the DAN

AC C

We observed that in older as well as in younger participants, the DAN is progressively activated with increasing memory load. At first sight, these results do not seem to support aging effects on the engagement of the top-down, task-related attention network. However, we also observed that older participants tend to recruit parts of this network (FEF) to a larger extent than younger participants and this increased activation was concomitant, at the behavioral level, with less accurate and slower responses in the older participants. However, when controlling for age-related slowing, it appears that there were no group differences anymore, suggesting that young and older adults were similarly affected by STM load 19

ACCEPTED MANUSCRIPT

manipulations. The higher recruitment of parts of the DAN in older participants was observed specifically for the high load STM condition. Increased activation in regions of the DAN in older participants could reflect

RI PT

compensation, or at least an attempt to compensate, in order to maintain performance. According to the compensatory recruitment hypothesis, older adults exhibit enhanced activation during cognitive tasks in order to maintain previous performance levels and

SC

compensate for age-related decline (Cabeza et al. 2002). Compensation has been observed in fronto-parietal regions in older adults in previous studies (Cabeza et al. 2004). In our study,

M AN U

there was no correlation between activation in this region and behavioral data. It therefore cannot be concluded whether increased activation in this network reflects compensatory attempt or less efficient use of neural resources in older adults.

We further observed that the increased activity in frontal DAN regions, responsible for

TE D

top-down attentional goal-directed behavior, was paralleled by reduced activation in occipital and VAN regions (see below), related to bottom-up attention processes. These results are compatible with a recent theory of aging effects on visual selective attention (Madden et al.,

EP

2007). According to Madden et al., aging is associated with increased activation in the dorsal fronto-parietal network, reflecting stronger engagement of the top-down attentional control,

AC C

and compensating for lower activation in the occipital areas in older participants, related to a decline in the quality of the bottom-up sensory input. 4.2. Effects of aging on the VAN The most important finding of this study is the reduced VAN response during STM tasks in aging. The deactivation in regions (TPJ) of the VAN with increasing memory load, consistently observed in young participants (Majerus et al., 2012; Todd et al., 2005) was not observed in our group of healthy older participants. However, PPI analysis showed that the 20

ACCEPTED MANUSCRIPT

TPJ region is inversely correlated to IPS in the high versus low load contrast in young but also in older adults. This suggests that the DAN and the VAN are still interacting in the older participants and further suggests that a functional balance between these networks is still

RI PT

present in aging. It could be that the VAN network, considered to be involved in the detection of task irrelevant information, does not sufficiently deactivate in older participants when task demands are high. This could prevent efficient deployment of the DAN and associated task-

SC

related attention processes, as STM task performance may be disrupted by distracting information. In that case, we should observe a particularly strong tendency in older

M AN U

participants to be distracted by the occurrence of DS. This was however not the case. First, older participants’ STM behavioral performance was impacted by the occurrence of a DS only in the low STM condition, as in young participants. Second, the older group showed the expected lower activation of the VAN when the DS occurred in a high load STM context, relative to the young participant group 1 . Hence, we have no evidence for a higher

TE D

distractibility during STM in the older group. Another possibility is that, although the VAN network activity is reduced but still functionally connected to the DAN in the older group during STM performance, VAN does not directly relate to the general decrease of STM

EP

performance observed in this group. A more general implication is that the VAN may not be determinant for STM performance. In support of this assumption, recent studies emphasize

1

AC C

the critical role of the dorsal part of the fronto-parietal network in STM performance. Ceko et

The young group showed similar activation of the VAN to the DS in the low and high STM load condition probably because the high load

condition (5 letters) was still fairly easy for this group. In a previous study in young participants, Majerus et al. (2012) had used three memory loads (2-4-6) and they observed DS-related VAN activation for the 2 and 4 STM load condition, but not for the 6-load condition. It thus appears that the high load condition in the present study (5 items) led to similar results as the 4-load condition in the Majerus et al. (2012) study. This is also supported by the behavioral results, with performance levels for the 5-load condition in the young participant sample of the present study being equivalent to performance levels for the 4-load condition in the Majerus et al. (2012) study.

21

ACCEPTED MANUSCRIPT

al. (2015) showed that a responsive external task positive network may be sufficient for successful performance in a working memory task. Furthermore, in contradiction to theories attributing a circuit-breaker role to the TPJ, the

RI PT

study of Leitao et al. (2015) suggests a causal influence from IPS to TPJ regions, but not the reverse. Leitao et al. showed that target-evoked TPJ activation is dependent on inputs from IPS, which suggests that the IPS and more generally the DAN may be the critical network for

SC

accurate behavioral performance.

M AN U

4.3 Conclusion

Despite less accurate and slower responses in older participants, we observed similar task and stimulus-related behavioral effects in young and older participants. These behavioral effects rely on different cerebral mechanisms even if interactions between DAN and VAN are still present with aging. The over-recruitment in regions of the DAN in older adults associated to

TE D

less accurate and slower responses could reflect attempted compensation to maintain performance or less efficient use of neural resources in older adults. Increased activity in frontal regions of the DAN could also reflect reliance on top-down attention in an attempt to

EP

compensate for bottom-up decline associated with aging, reflected by less reactivity in VAN and occipital regions. In older adults, we observed that VAN deactivation was no more

AC C

present with increasing STM load. Furthermore, in older participants, we were not able to observe stimulus-driven activity in the VAN in response to DS in low or high load, suggesting a less responsive VAN network in older adults. In terms of functional balance between networks, this study however suggests that the networks are still interacting but in a slightly modified way. Those results also support the critical role of the DAN for good achievement in our STM task and for accurate behavioral performance more generally. Acknowledgements 22

ACCEPTED MANUSCRIPT

This work was supported by the SAO-FRA grant S#12005, ARC 12/17-01 REST, InterUniversity Attraction Pole P7/11, the F.R.S.-FNRS, SM is Senior Research Associate at F.R.S.-FNRS, CB is Research Associate at F.R.S.-FNRS, FC is Research Director at F.R.S.-

RI PT

FNRS, and the University of Liège.

SC

References

Andersson, J.L., Hutton, C., Ashburner, J., Turner, R., Friston, K., 2001. Modeling geometric

M AN U

deformations in EPI time series. Neuroimage 13, 903–919.

Anticevic, A., Repovs, G., Shulman, G.L., Barch, D.M., 2010. When less is more: TPJ and default network deactivation during encoding predicts working memory performance. Neuroimage 49, 2638–2648.

Ashburner, J., Friston, K.J., 2005. Unified segmentation. Neuroimage 26, 839–851.

TE D

Asplund, C.L., Todd, J.J., Snyder, A.P., Marois, R., 2010. A central role for the lateral prefrontal cortex in goal-directed and stimulus-driven attention. Nat. Neurosci. 13, 507–512.

EP

Cabeza, R., Anderson, N.D., Locantore, J.K., McIntosh, A.R. 2002. Aging gracefully: compensatory brain activity in high-performing older adults. Neuroimage 17, 1394-

AC C

1402.

Cabeza, R., Daselaar, S.M., Dolcos, F., Prince, S.E., Budde, M., Nyberg, L., 2004. Taskindependent and task-specific age effects on brain activity during working memory,

visual attention and episodic retrieval. Cereb. Cortex 14, 364–375.

Cabeza, R., Ciaramelli, E., Moscovitch, M., 2012. Cognitive contributions of the ventral parietal cortex: an integrative theoretical account. Trends Cogn. Sci. 16, 338-352. Čeko, M., Gracely, J.L., Fitzcharles, M.-A., Seminowicz, D.A., Schweinhardt, P., Bushnell,

23

ACCEPTED MANUSCRIPT

M.C., 2015. Is a Responsive Default Mode Network Required for Successful Working Memory Task Performance? J. Neurosci. 35, 11595–11605. Chica, A.B., Bartolomeo, P., Valero-Cabré, A., 2011. Dorsal and ventral parietal

RI PT

contributions to spatial orienting in the human brain. J. Neurosci. 31, 8143–8149. Chiu, Y.-C., Yantis, S., 2009. A domain-independent source of cognitive control for task sets: shifting spatial attention and switching categorization rules. J. Neurosci. 29, 3930–

SC

3938.

Corbetta, M., Kincade, J.M., Shulman, G.L., 2002. Neural systems for visual orienting and

M AN U

their relationships to spatial working memory. J. Cogn. Neursci. 14, 508–523. Corbetta, M., Shulman, G.L., 2002. Control of goal-directed and stimulus-driven attention in the brain. Nat. Rev. Neurosci. 3, 201–215.

Corbetta, M., Patel, G., Shulman, G.L., 2008. The reorienting system of the human brain: from environment to theory of mind. Neuron 58, 306–324.

TE D

Cowan, N., Elliott, E. M., Saults, J. S., Morey, C. C., Mattox, S., Hismjatullina, A., Conway, A. R., 2005. On the capacity of attention: Its estimation and its role in working memory and cognitive aptitudes. Cogn. psychol. 51, 42-100.

EP

Cowan, N., 2011. The focus of attention as observed in visual working memory tasks: Making sense of competing claims. Neuropsychologia 49, 1401-1406.

AC C

Craik FIM, Jennings JM: Human memory. In Craik, FIM., Salthouse, TA. Editors. The Handbook of Aging and Cognition. Hillsdale, NJ: Lawrence Erlbaum; 1992. p. 51110.

Davis, S.W., Dennis, N.A., Daselaar, S.M., Fleck, M.S., Cabeza, R., 2008. Que PASA? The posterior–anterior shift in aging. Cereb. Cortex 18, 1201–1209. Engle, R.W., Kane, M.J. 2004. Executive attention, working memory capacity, and a twofactor theory of cognitive control. Psychol. Learn. Motiv. 44, 145-200.

24

ACCEPTED MANUSCRIPT

Faust, M.E., Balota, D.A., Spieler, D.H., Ferraro, F.R. 1999. Individual differences in information-processing rate and amount: implications for group differences in response latency. Psychol. Bull. 125, 777-799.

RI PT

Fox, M.D., Snyder, A.Z., Vincent, J.L., Corbetta, M., Van Essen, D.C., Raichle, M.E. 2005. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. P. Natl. Acad. Sci. USA 102, 9673–9678.

SC

Geerligs, L., Maurits, N.M., Renken, R.J., Lorist, M.M., 2014. Reduced specificity of functional connectivity in the aging brain during task performance. Hum. Brain Mapp.

M AN U

35, 319–330.

Geerligs, L., Renken, R. J., Saliasi, E., Maurits, N. M., Lorist, M. M., 2015. A brain-wide study of age-related changes in functional connectivity. Cereb. Cortex 25, 1987-1999. Grady, C.L., Maisog, J.M., Horwitz, B., Ungerleider, L.G., Mentis, M., Salerno, J., Pietrini, P., Wagner, E., Haxby, J., 1994. Age-related changes in cortical blood flow activation

TE D

during visual processing of faces and location. J. Neurosci. 14, 1450–1462. Hutton, C., Bork, A., Josephs, O., Deichmann, R., Ashburner, J., Turner, R., 2002. Image distortion correction in fMRI: a quantitative evaluation. Neuroimage 16, 217–240.

EP

Goh, J.O., 2011. Functional dedifferentiation and altered connectivity in older adults: neural accounts of cognitive aging. Aging Disease 2, 30–48.

AC C

Lavie, N., 2005. Distracted and confused? : Selective attention under load. Trends Cogn. Sci. 9, 75-82.

Leitão, J., Thielscher, A., Tünnerhoff, J., Noppeney, U., 2015. Concurrent TMS-fMRI Reveals Interactions between Dorsal and Ventral Attentional Systems. J. Neurosci. 35, 11445–11457. Linden, D. E., Bittner, R. A., Muckli, L., Waltz, J. A., Kriegeskorte, N., Goebel, R., Singer, W., Munk, M. H., 2003. Cortical capacity constraints for visual working memory:

25

ACCEPTED MANUSCRIPT

dissociation of fMRI load effects in a fronto-parietal network. Neuroimage 20, 15181530. Lorenzo-López, L., Doallo, S., Vizoso, C., Amenedo, E., Holguín, S.R., Cadaveira, F., 2002.

RI PT

Covert orienting of visuospatial attention in the early stages of aging. Neuroreport 13, 1459–1462.

Madden, D.J., 2007. Aging and visual attention. Curr. Dir. Psycho. Sci. 16, 70–74.

SC

Madden, D.J., Spaniol, J., Whiting, W.L., Bucur, B., Provenzale, J.M., Cabeza, R., White, L.E., Huettel, S.A., 2007. Adult age differences in the functional neuroanatomy of

M AN U

visual attention: a combined fMRI and DTI study. Neurobiol. Aging 28, 459–476. Majerus, S., D’Argembeau, A., Perez, T.M., Belayachi, S., Van der Linden, M., Collette, F., Salmon, E., Seurinck, R., Fias, W., Maquet, P., 2010. The commonality of neural networks for verbal and visual short-term memory. J. Cogn. Neurosci. 22, 2570–2593. Majerus, S., Attout, L., D’Argembeau, A., Degueldre, C., Fias, W., Maquet, P., Perez, T.M.,

TE D

Stawarczyk, D., Salmon, E., Van der Linden, M., 2012. Attention supports verbal short-term memory via competition between dorsal and ventral attention networks. Cereb. Cortex 22, 1086–1097.

EP

Mattis, S. Mental status examination for organic mental syndrome in the elderly patient. In Bellak, L., Karasu T.B., editors. Geriatric Psychiatry. New York: Grune & Stratton;

AC C

1976. p. 77–121.

Maylor, E.A., Lavie, N., 1998. The influence of perceptual load on age differences in selective attention. Psychol. Aging 13, 563.

Paulesu, E., Frith, C.D., Frackowiak, R.S., 1993. The neural correlates of the verbal component of working memory. Nature 362, 342-345. Ravizza, S.M., Delgado, M.R., Chein, J.M., Becker, J.T., Fiez, J.A., 2004. Functional dissociations within the inferior parietal cortex in verbal working memory.

26

ACCEPTED MANUSCRIPT

Neuroimage 22, 562–573. Salthouse, T.A., 1996. The processing-speed theory of adult age differences in cognition. Psychol. Rev. 103, 403-428.

RI PT

Serences, J.T., Shomstein, S., Leber, A.B., Golay, X., Egeth, H.E., Yantis, S., 2005. Coordination of voluntary and stimulus-driven attentional control in human cortex. Psychol. Sci. 16, 114–122.

SC

Spreng, R.N., Wojtowicz, M., Grady, C.L., 2010. Reliable differences in brain activity between young and old adults: a quantitative meta-analysis across multiple cognitive

M AN U

domains. Neurosci. Biobehav. R. 34, 1178–1194.

Todd, J.J., Marois, R., 2004. Capacity limit of visual short-term memory in human posterior parietal cortex. Nature 428, 751–754.

Todd, J.J., Fougnie, D., Marois, R., 2005. Visual short-term memory load suppresses temporo-parietal junction activity and induces inattentional blindness. Psychol. Sci.

TE D

16, 965–972.

Verhaeghen, P., Cerella, J., 2002. Aging, executive control, and attention: a review of metaanalyses. Neurosci. Biobehav. R. 26, 849–857.

EP

Vossel, S., Weidner, R., Driver, J., Friston, K.J., Fink, G.R., 2012. Deconstructing the architecture of dorsal and ventral attention systems with dynamic causal modeling. J.

AC C

Neurosci. 32, 10637–10648. Vossel, S., Geng, J.J., Fink, G.R., 2014. Dorsal and ventral attention systems distinct neural circuits but collaborative roles. The Neuroscientist 20, 150–159.

Worsley, K.J., Marrett, S., Neelin, P., Evans, A., 1996a. Searching scale space for activation in PET images. Hum. Brain Mapp. 4, 74–90. Worsley, K.J., Marrett, S., Neelin, P., Vandal, A.C., Friston, K.J., Evans, A.C., 1996b. A unified statistical approach for determining significant signals in images of cerebral

27

ACCEPTED MANUSCRIPT

activation. Hum. Brain Mapp. 4, 58–73. Zanto, T.P., Gazzaley, A. Attention and ageing. In Nobre, A.C., Kastner, S., editors. The Oxford Handbook of Attention. Oxford, UK: Oxford University Press; 2014. p. 927-

AC C

EP

TE D

M AN U

SC

RI PT

971.

28

ACCEPTED MANUSCRIPT

Figure 1: Schematic drawing of the short term memory task design and timing. Figure 2: Accuracy (left) and reaction times-RT (right) for the behavioral data in both groups. Error bars represent standard errors of the mean.

RI PT

Figure 3: Activation in regions of the DAN common to young and older participants in the high versus low STM load contrast.

Figure 4: Results of the 5>2 load contrast in young>old. Lingual gyrus (10 -86 0) is

being more important in young participants.

SC

deactivated in the low load condition and activated in the high load condition; this effect

M AN U

Figure 5: Results of the 2>5 load contrast in young > old. Left TPJ (-52 -53 23) is more strongly deactivated in the high versus low load condition in young participants, but not in older participants.

Figure 6: Results of the DS2>DS5 contrast in young>old. Lingual gyrus (-12 -88 -4) is more strongly deactivated in the high versus low load condition in young participants, but not in

TE D

older participants.

Figure 7: Results of the DS5 contrast in young>old. Left TPJ (-54 -54 20) shows relatively less activation in older participants in response to DS in the high load condition, but there was

AC C

EP

no significant group difference in the low load condition

ACCEPTED MANUSCRIPT

Table 1: Regions of interest for the DAN and the VAN Y

Z

References

24 -25

-56 -57

46 46

rSFG lSFG

26 -22

-2 -3

47 49

right and left intraparietal sulcus (Corbetta and Shulman 2002; Serences et al,. 2005; Chiu and Yantis 2009; Asplund et al., 2010) right and left superior frontal gyrus (Serences et al., 2005)

rMFG

48

7

34

rTPJ lTPJ

52 -52

-52 -53

25 23

rOFC lOFC

34 -37

27 27

-10 -8

RI PT

VAN

X rIPS lIPS

right middle frontal gyrus (Serences et al., 2005 ; Chiu and Yantis 2009) right and left temporo-parietal junction (Todd et al., 2005; Asplund et al., 2010) right and left orbitofrontal gyrus (Asplund et al., 2010)

SC

DAN

Table 2: Results of conjunction analysis for the high vs low STM load contrast over the young and old groups (p<.05 FWE corrected). Cluster

Z-Value

Young and old 5>2 Pre-central Gyrus

L

718

5.84

X -46

Y -2

Z 54

MFG MFG

R R

65 52

4.65 4.73

40 54

42 2

32 40

SFG

L

252

5.17

-6

4

56

Superior temporal gyrus (STG)

L

12

4.67

-54

-36

12

Lingual gyrus

R

12990

7.19

10

-84

-8

L R R L

642 222 343 203

5.40 5.15 5.10 5.38

-18 20 28 -4

-6 -4 20 -32

8 6 6 -6

EP

Globus pallidus Globus pallidus Insular gyrus Accumbens

MNI stereotaxic coordinates

M AN U

Side

TE D

Contrast

AC C

Table 3: Region of interest analysis for the high vs low STM load contrast. Dorsal attention network: right IPS: 24 -56 46; left IPS: -25 -57 46; right SFG: 26 -2 47; left SFG: -22 -3 49; MFG: 48 7 34). Contrast

Side

Cluster

Z-Value

R R R L L

24 313 313 238 238

4.67 5.55 5.25 5.94 5.25

Young and old 5>2 MFG rIPS rIPS lIPS lIPS

MNI stereotaxic coordinates X

Y

Z

52 24 20 -22 -25

2 -56 -54 -62 -50

40 50 42 52 44

ACCEPTED MANUSCRIPT

Table 4: Between-group effects for the high vs low STM load contrast (p< .05 FWE corrected) Side

Cluster

Z-Value

R L

658 658

4.67 4.34

X 10 -6

R L R

969 504 296

4.25 4.60 4.32

48 -42 50

Young>Old 5>2 Lingual gyrus Lingual gyrus

MNI stereotaxic coordinates

Old>young 5>2 Pre-central Middle temporal gyrus (MTG) MTG

Y -86 -86

Z 0 -2

RI PT

Contrast

-10 -56 -62

22 6 -2

Table 5: Between group effects for the high STM load contrast only (p< .05 FWE corrected) Side

Cluster

Z-Value

R R R L L L L

2314 2314 2314 922 546 573 862

4.74 4.60 4.55 4.50 4.35 4.26 5.47

Pre-central Pre-central Pre-central Pre-central STG (TPJ) STG (TPJ) MTG

X 34 26 48 -32 -58 -54 -40

M AN U

Old>young 5

MNI stereotaxic coordinates

SC

Contrast

Y -8 2 -2 -10 -46 -50 -54

Z 32 24 24 28 18 24 8

Table 6: Region of interest analysis for low vs high STM load contrast. Ventral attention network: Right TPJ: 52 -52 25; left TPJ: -52 -53 23; right OFC: 34 27 -10; left OFC: -37 27 8. Side

Cluster

TE D

Contrast Young 2>5 lTPJ

Young>Old 2>5

MNI stereotaxic coordinates X

Y

Z

L

3

3.15

-48

-60

20

L

75

3.88

-56

-46

24

EP

lTPJ

Z-Value

AC C

Table 7: Within and between group effects for DS-related activity in the low vs high load and high vs low load contrasts. Contrast

Side

Cluster

Z-Value

X

Y

Z

L L L R R R

1444 1444 1444 238 238 400

6.59 5.94 5.50 5.38 4.49 5.73

-12 -30 -22 34 34 10

-88 -80 -86 -86 -74 -86

-4 -10 10 -10 -8 0

L

32

4.68

-12

-88

-4

Young DS2>DS5

Lingual gyrus Inferior occipital gyrus Middle or lateral occipital Middle or lateral occipital Inferior occipital Cuneus

MNI stereotaxic coordinates

Young>old DS2>5 Lingual gyrus

Note: No significant results were found for the following contrasts: Young DS5>DS2; Old DS2>DS5; OLD DS5>DS2

ACCEPTED MANUSCRIPT

Table 8: Within and between group effects for DS-related activity in low and high load conditions separately (p< .05 FWE corrected) Side

Cluster

Z-Value

R R L L L L R R R R R L

45 89 65 546 546 573 862 862 432 432 208 166

4.68 5.14 4.95 5.69 5.48 5.46 5,47 4.91 5.65 4.54 5.88 5.54

R L

43 42

4.84 4.79

X

Y

Z

L L R R L R L

195 596 308 308 95 95 30

5.72 5.53 5.11 4.86 5.19 4.80 5.17

-36 -46 46 54 -8 2 -38

-26 -62 -62 -58 -60 -60 -80

50 12 16 4 40 40 34

L L

50 37

4.83 4.77

-46 -52

-62 -26

14 52

L L L R

533 533 533 110

5.61 5.05 4.77 5.08

-14 -28 -30 20

-92 -82 -86 -86

-8 -14 0 -10

Inferior frontal gyrus (IFG) rOFC lOFC Post-central (anterior IPS) Post-central (anterior IPS) MTG (TPJ) MTG (TPJ) MTG (TPJ) Angular gyrus (rTPJ) Angular gyrus (rTPJ) Thalamus Thalamus

Thalamus Midbrain

Old DS2>young DS2 Nihil

Young>old DS5 MTG (TPJ) Post-central

Old>young DS5

AC C

Lingual Inferior occipital Lateral occipital Occipital

EP

Post-central (anterior IPS) MTG (TPJ) MTG (TPJ) MTG (TPJ) Precuneus Precuneus Lateral occipital

TE D

Young DS5

Y

Z

12 24 20 -28 -22 -52 -46 -66 -46 -66 -16 -20

24 -6 2 44 54 10 8 16 40 56 0 -4

-20 -26

4 -2

16 -8

M AN U

Old DS2

X 40 34 -30 -36 -48 -44 52 50 40 30 12 -10

RI PT

Young DS2

MNI stereotaxic coordinates

SC

Contrast

ACCEPTED MANUSCRIPT

Table 9: Results of region of interest analysis for DS-related activity in low and high load conditions. Ventral attention network: Right TPJ: 52 -52 25; left TPJ: -52 -53 23; right OFC: 34 27 -10; left OFC: -37 27 -8). Side

Cluster

Z-Value

X

Y

Z

R L R R L

220 50 168 6 172

5.14 5.20 4.85 3.29 5.05

34 -30 48 48 -48

24 24 -50 -46 -50

-6 -4 18 32 16

R L R R L

70 40 91 91 292

3.69 4.42 4.55 3.48 5.10

34 -30 48 50 -52

24 22 -58 -46 -60

-6 -8 20 24 18

L

173

4.00

-54

20

Young DS2 rOFC LOFC rTPJ rTPJ lTPJ

MNI stereotaxic coordinates

rOFC lOFC rTPJ rTPJ lTPJ

lTPJ

-54

M AN U

Young >old DS5

SC

Young DS5

RI PT

Contrast

Table 10: Psycho-physiological interactions in the high vs low load contrast. Regions displaying negative interactions with the left or right IPS seeds are presented. Results from small volume corrections are presented in bold type.

Seed

Young

lIPS

EP

lTPJ rTPJ Precuneus Precuneus Post. Cingulate lTPJ rTPJ Precuneus Post. Cingulate

AC C

rIPS

Region

Older

lIPS rIPS

Cluster

TE D

Group

lTPJ lTPJ

Z-Value

MNI stereotaxic coordinates X

Y

Z

222 481 1041 1041 1041 42 1 525 525

4.02 3.76 4.22 4.06 4.11 3.38 3.15 3.93 3.54

-56 54 -6 4 -10 -48 54 -4 10

-54 -54 -56 -54 -46 -52 -54 -54 -52

24 18 44 36 40 22 26 36 36

10 31

3.29 3.72

-48 -52

-50 -52

32 32

AC C EP TE D

M AN US C

RI PT

AC C EP TE D

M AN US C

RI PT

AC C EP TE D

M AN US C

RI PT

AC C EP TE D

M AN US C

RI PT

AC C EP TE D

M AN US C

RI PT

AC C EP TE D

M AN US C

RI PT

AC C EP TE D

M AN US C

RI PT

ACCEPTED MANUSCRIPT

Investigation of age effect on attention networks interaction during STM Old and young participants display the same behavioral task and stimulus-driven effects DAN activity is preserved in aging and essential for accurate STM VAN activation/deactivation is disrupted in aging during STM VAN is still functionally connected to the DAN in aging

AC C

EP

TE D

M AN U

SC

RI PT

• • • • •