Elsevier

NeuroImage

Volume 86, 1 February 2014, Pages 150-163
NeuroImage

A multivariate analysis of age-related differences in functional networks supporting conflict resolution

https://doi.org/10.1016/j.neuroimage.2013.08.002Get rights and content

Highlights

  • Old adults were less accurate than young adults during the interference condition.

  • Greater activity during interference in the old was observed in DLPFC and ACC.

  • Connectivity networks differentially related to performance across age groups.

  • Age-related GM losses were seen in regions facilitated performance in the young.

Abstract

Functional neuroimaging studies demonstrate age-related differences in recruitment of a large-scale attentional network during interference resolution, especially within dorsolateral prefrontal cortex (DLPFC) and anterior cingulate cortex (ACC). These alterations in functional responses have been frequently observed despite equivalent task performance, suggesting age-related reallocation of neural resources, although direct evidence for a facilitating effect in aging is sparse. We used the multi-source interference task and multivariate partial-least-squares to investigate age-related differences in the neuronal signature of conflict resolution, and their behavioral implications in younger and older adults. There were interference-related increases in activity, involving fronto-parietal and basal ganglia networks that generalized across age. In addition an age-by-task interaction was observed within a distributed network, including DLPFC and ACC, with greater activity during interference in the old. Next, we combined brain–behavior and functional connectivity analyses to investigate whether compensatory brain changes were present in older adults, using DLPFC and ACC as regions of interest (i.e. seed regions). This analysis revealed two networks differentially related to performance across age groups. A structural analysis revealed age-related gray-matter losses in regions facilitating performance in the young, suggesting that functional reorganization may partly reflect structural alterations in aging. Collectively, these findings suggest that age-related structural changes contribute to reductions in the efficient recruitment of a youth-like interference network, which cascades into instantiation of a different network facilitating conflict resolution in elderly people.

Introduction

Humans have access to information that originates from available external stimuli or can be retrieved from stored experiences. However, prepotent but task-irrelevant information may interfere with relevant information and impair performance. The ability to ignore irrelevant information has been investigated in several paradigms, with participants responding to stimuli in the presence or absence of distracting items (Bush et al., 2003, Eriksen and Eriksen, 1974, Simon and Berbaum, 1990, Stroop, 1935). Each paradigm relies on different methods for inducing cognitive conflict (for review, see Nee et al., 2007). For instance, the Stroop effect represents cognitive interference produced by color incongruence between a word depicting a color with the color of the ink (i.e. stimulus conflict), whereas the Simon effect denotes cognitive interference through spatial incongruence between the target and response (i.e. response conflict). The type of conflict resolution targeted in this research declines with age. Elderly persons have fewer resources available during a task conflict to inhibit irrelevant information (Gazzaley et al., 2008, Madden et al., 2004). Imaging studies have linked age-related deficits in conflict resolution to alterations in the anterior control network that involves DLPFC (Langenecker et al., 2004, Thomsen et al., 2004) and dorsal ACC (Milham et al., 2002), as well as the posterior attention network, including superior parietal cortex (Schulte et al., 2009). For example, Langenecker and colleagues reported similar brain activation patterns for younger and older adults during a Stroop task, along with age-related over-recruitment in frontal regions. The authors suggested that greater frontal activity among elderly people promotes successful inhibition. By contrast, Milham et al. found that older adults exhibited reduced DLPFC activity and greater ACC activity compared to younger adults during interference. Reduced DLPFC activity may reflect age-related impairment in attentional control, whereas increased ACC activity may indicate heightened potential for error.

Complex cognitive processes, such as interference resolution, may not be localized to discrete brain regions such as ACC and DLPFC, but rather be mediated by interactions among a set of functionally related areas (McIntosh, 1999). Functional connectivity (Friston et al., 1993, McIntosh, 1999) is an approach to directly assess interactions among network nodes for a specific cognitive process and their alterations in aging, which in turn might affect behavior (e.g. Clapp et al., 2011, Grady et al., 2010, Nagel et al., 2011). For example, Clapp and colleagues reported age-related working memory deficits for scenes during a delayed matching-to-sample task along with disruption of functional connectivity between PFC and the parahippocampal place area. Despite increasing interest in understanding covarying activity within a network for determining the neural underpinnings of cognitive functions (Bressler and Menon, 2010, McIntosh, 1999), direct evidence regarding age-related differences in connectivity of critical nodes for conflict resolution (i.e., ACC, DLPFC) and concomitant behavioral implications is lacking. To date, few studies have reported how brain responses during interference are modulated by performance (Zysset et al., 2007), and no study has examined whole-brain patterns of activity in relation to performance (for review, see Grady, 2012). Similarly, past research on this topic has examined regional effects of age on conflict resolution and not reported data at the network level. Thus, although past research suggests age-related differences in functional activation of critical nodes for conflict resolution, they do not provide direct evidence as to whether unique brain networks promote interference resolution across different age groups.

In the context of possible functional alterations in aging, an important question is how age-related functional alterations relate to other factors that are also affected by aging and influence brain function, such as brain structure. There is robust evidence for age-related decline in brain volumes, particularly in the frontal lobes (Good et al., 2001, Raz et al., 2005). Age-related differences in gray-matter (GM) volume may locally account for regional alterations in functional responses (e.g. Kalpouzos et al., 2011, Salami et al., 2012). However, an examination of the full set of brain regions also reflecting potential distal associations is lacking (e.g. Calhoun et al., 2006, Stevens et al., 2009).

To assess age differences in conflict resolution and associated brain systems, multivariate spatial–temporal partial-least-squares (PLS; McIntosh et al., 1996, McIntosh et al., 2004) was applied to data from younger and older adults. Participants were scanned with functional magnetic resonance imaging (fMRI) while they performed the Multi-Source Interference Task (MSIT; Bush and Shin, 2006, Bush et al., 2003), which involves control and interference trials and the critical contrast concerns differences in accuracy and/or response latency between the two conditions. This task combines multiple dimensions of cognitive interference. Specifically, the MSIT contains elements of flanker interference (i.e. distracting items flanking the target; Eriksen and Eriksen, 1974) and Simon interference (i.e. incongruity between the position of the target and the position of the response; Simon and Berbaum, 1990). The MSIT engages the cingulo-fronto-parietal attentional network (Bush and Shin, 2006, Bush et al., 2003). To identify whole-brain activity during the two MSIT conditions for each age group, we initially applied task PLS analysis. As opposed to traditional analysis which largely rests on cognitive subtraction, PLS is able to use all conditions in an experiment at once, and thus provides an additional dimension to data by simultaneously considering indices of both similarities and differences across all grouping/experimental variables. If young and older adults engage many similar brain regions differentiating the two task conditions, task PLS analysis reveals an age-common network. Alternatively and/or in addition, if some brain regions exhibit age-differential activation between conditions, PLS analysis reveals group-specific networks. Another way in which interference resolution might be compromised in aging is that regions supporting interference resolution may remain relatively similar, although the functional connectivity within the network is altered. On this view, less proficient interference resolution in elderly people may be related to less efficient recruitment of the network promoting interference resolution in the young. Specifically, interference resolution might be compromised in older adults due to less efficient interaction within the anterior control network or the posterior attention network. Using seed PLS, we first examined whether functional connectivity among those network nodes reflecting the most reliable group differences varied between age groups and was modulated by performance. If younger and older adults alike engage the same network to support performance, the seed PLS should reveal a common circuitry with possible quantitative differences across age. Alternatively, if younger and older persons recruit distinct networks to support performance, the seed PLS should reveal networks differentially correlated with performance across age. This analysis allows us to verify whether the functional network supporting interference resolution in old age is similar to that recruited by the young, or constitutes a different network that may not facilitate interference resolution in younger adults. Finally, relationships between brain activity and GM volume were explored to investigate whether age-related differences in structural integrity are associated with age-related alterations in functional networks during interference resolution.

Section snippets

Participants

29 young (20–31 years of age, 16 females) and 29 old (65–74 years of age, 16 females) participants from Stockholm, Sweden participated. There were no significant age differences in years of education (Young: 14.7 ± 2.1; Old: 14.3 ± 3.7) or on a test of mental status (Mini Mental Status Examination, Folstein et al., 1975) (Young: 29.2 ± 0.7; Old: 28.9 ± 0.8). From the initial sample of 58 participants, four (three old, one young) were excluded due to low task performance (below chance level). Two older

Behavioral findings

2 (Age) × 2 (Condition) ANOVAs were conducted on the data for accuracy and latency for correct trials. Where appropriate, t-tests with Bonferroni correction were carried out. For accuracy (Fig. 1C), there was a main effect of condition (F (3, 98) = 24.98, p < 0.0001), a marginally significant effect of age (F (3, 98) = 3.36, p = 0.068), and a significant age × condition interaction (F (3, 98) = 4.40, p < 0.05). Accuracy was lower during interference than during control in older adults (p < 0.0001), but only

Discussion

We explored age differences in the neural correlates of interference resolution, an operation by which the brain attempts to limit processing to task-relevant information while ignoring irrelevant information. We characterized the behavioral implications of neural activity patterns during the MSIT in young and older adults, and provide novel evidence of functional reorganization of neural networks to support interference resolution in aging.

Older adults were slower during both interference and

Acknowledgments

This research was conducted at the Karolinska Institute MR-Center, Huddinge Hospital, Stockholm, Sweden. The study was supported by grants from the Swedish Research Council and Swedish Brain Power, an Alexander von Humboldt Research Award, and a donation from the af Jochnick Foundation to Lars Bäckman, and by a grant from the Swedish Research Council to Håkan Fischer. The authors wish to thank Joakim Svärd for assistance in data collection and entry, and Lars Nyberg for comments on a previous

References (88)

  • J.O. Goh et al.

    Frontal function and executive processing in older adults: process and region specific age-related longitudinal functional changes

    NeuroImage

    (2013)
  • C.D. Good et al.

    Cerebral asymmetry and the effects of sex and handedness on brain structure: a voxel-based morphometric analysis of 465 normal adult human brains

    NeuroImage

    (2001)
  • B.J. Harrison et al.

    Functional connectivity during Stroop task performance

    NeuroImage

    (2005)
  • C.M. Huang et al.

    Both left and right posterior parietal activations contribute to compensatory processes in normal aging

    Neuropsychologia

    (2012)
  • G. Kalpouzos et al.

    Voxel-based mapping of brain gray matter volume and glucose metabolism profiles in normal aging

    Neurobiol. Aging

    (2009)
  • A. Krishnan et al.

    Partial Least Squares (PLS) methods for neuroimaging: a tutorial and review

    NeuroImage

    (2011)
  • S.A. Langenecker et al.

    fMRI of healthy older adults during Stroop interference

    NeuroImage

    (2004)
  • A.R. McIntosh et al.

    Partial least squares analysis of neuroimaging data: applications and advances

    NeuroImage

    (2004)
  • A.R. McIntosh et al.

    Spatial pattern analysis of functional brain images using partial least squares

    NeuroImage

    (1996)
  • A.R. McIntosh et al.

    Spatiotemporal analysis of event-related fMRI data using partial least squares

    NeuroImage

    (2004)
  • M.P. Milham et al.

    Attentional control in the aging brain: insights from an fMRI study of the Stroop task

    Brain Cogn.

    (2002)
  • R.S. Prakash et al.

    Age-related differences in the involvement of the prefrontal cortex in attentional control

    Brain Cogn.

    (2009)
  • R.S. Prakash et al.

    Age-related differences in cortical recruitment and suppression: implications for cognitive performance

    Behav. Brain Res.

    (2012)
  • M.E. Raichle et al.

    A default mode of brain function: a brief history of an evolving idea

    NeuroImage

    (2007)
  • P.A. Reuter-Lorenz et al.

    Brain aging: reorganizing discoveries about the aging mind

    Curr. Opin. Neurobiol.

    (2005)
  • A. Salami et al.

    Characterizing the neural correlates of modality-specific and modality-independent accessibility and availability signals in memory using partial-least squares

    NeuroImage

    (2010)
  • J.R. Simon et al.

    Effect of conflicting cues on information processing: the ‘Stroop effect’ vs. the ‘Simon effect’

    Acta Psychol. (Amst)

    (1990)
  • Y. Stern

    Cognitive reserve

    Neuropsychologia

    (2009)
  • M.C. Stevens et al.

    Age-related cognitive gains are mediated by the effects of white matter development on brain network integration

    NeuroImage

    (2009)
  • S.F. Taylor et al.

    Isolation of specific interference processing in the Stroop task: PET activation studies

    NeuroImage

    (1997)
  • T. Thomsen et al.

    Brain localization of attentional control in different age groups by combining functional and structural MRI

    NeuroImage

    (2004)
  • D.J. Tisserand et al.

    Age-related reorganization of encoding networks directly influences subsequent recognition memory

    Brain Res. Cogn. Brain Res.

    (2005)
  • M.W. Voss et al.

    Dedifferentiation in the visual cortex: an fMRI investigation of individual differences in older adults

    Brain Res.

    (2008)
  • E. Zarahn et al.

    Age-related changes in brain activation during a delayed item recognition task

    Neurobiol. Aging

    (2007)
  • S. Zysset et al.

    Stroop interference, hemodynamic response and aging: an event-related fMRI study

    Neurobiol. Aging

    (2007)
  • M.T. Banich et al.

    fMRI studies of Stroop tasks reveal unique roles of anterior and posterior brain systems in attentional selection

    J. Cogn. Neurosci.

    (2000)
  • R.L. Buckner et al.

    The brain's default network: anatomy, function, and relevance to disease

    Ann. N. Y. Acad. Sci.

    (2008)
  • G. Bush et al.

    The Multi-Source Interference Task: an fMRI task that reliably activates the cingulo-frontal-parietal cognitive/attention network

    Nat. Protoc.

    (2006)
  • G. Bush et al.

    Dorsal anterior cingulate cortex: a role in reward-based decision making

    Proc. Natl. Acad. Sci. U. S. A.

    (2002)
  • G. Bush et al.

    The Multi-Source Interference Task: validation study with fMRI in individual subjects

    Mol. Psychiatry

    (2003)
  • V.D. Calhoun et al.

    Extracting intrinsic functional networks with feature-based group independent component analysis

    Psychometrika

    (2013)
  • V.D. Calhoun et al.

    A method for making group inferences from functional MRI data using independent component analysis

    Hum. Brain Mapp.

    (2001)
  • V.D. Calhoun et al.

    Method for multimodal analysis of independent source differences in schizophrenia: combining gray matter structural and auditory oddball functional data

    Hum. Brain Mapp.

    (2006)
  • W.C. Clapp et al.

    Deficit in switching between functional brain networks underlies the impact of multitasking on working memory in older adults

    Proc. Natl. Acad. Sci. U. S. A.

    (2011)
  • Cited by (28)

    • Combination of structural and functional connectivity explains unique variation in specific domains of cognitive function

      2022, NeuroImage
      Citation Excerpt :

      In recent years, efforts have been made to investigate how SC and FC relate to each other and how this relationship may affect cognitive function and health (Bullmore & Sporns, 2009; Rykhlevskaia et al., 2008). A key question is whether SC and FC provide complementary or, alternatively, overlapping information for explaining cognition and behaviour (de Kwaasteniet et al., 2013; Guye et al., 2010; Hahn et al., 2013; Salami et al., 2014; van den Heuvel & Fornito, 2014; Wang et al., 2016). In general, the research that relates brain structure and function is motivated by the fact that the human brain activates and operates on the scaffold of neurons and neuronal connections.

    • Functional maintenance in the multiple demand network characterizes superior fluid intelligence in aging

      2020, Neurobiology of Aging
      Citation Excerpt :

      Thus, it is reasonable to expect that older individuals with youth-like brain measures are also those with intact gF. Previous studies have shown age-related alterations in the functional connectivity of large-scale networks, including the default mode network (DMN) (Salami et al., 2014, 2016), frontoparietal network (FPN) (Andrews-Hanna et al., 2007; Avelar-Pereira et al., 2017), and salience network (Ng et al., 2016). Moreover, alterations in global functional architecture of the brain have been observed such that older individuals show lower modularity (i.e., less segregation among large-scale networks) than their younger counterparts (Cao et al., 2014; Geerligs et al., 2015).

    • Cognitive decline and increased hippocampal p-tau expression in mice with hearing loss

      2018, Behavioural Brain Research
      Citation Excerpt :

      The manifestation of cognitive impairment depends on the ability of an individual to use neural reserve or to recruit alternative neural networks to actively compensate for pathology [24]. Functional neuroimaging studies have demonstrated age-related differences in recruitment of a neural network, suggesting that successful aging is associated with functional reorganization of the neural systems [50,51]. It may be that the NIHL mice were successful in compensating for working memory impairment.

    View all citing articles on Scopus
    View full text