A New Measure for Neural Compensation Is Positively Correlated With Working Memory and Gait Speed.

Lanxin Ji,David C Steffens,Keith A Hawkins,Godfrey D Pearlson,Hua Guo,Lihong Wang

doi:10.3389/fnagi.2018.00071

Abstract

Neuroimaging studies suggest that older adults may compensate for declines in brain function and cognition through reorganization of neural resources. A limitation of prior research is reliance on between-group comparisons of neural activation (e.g., younger vs. older), which cannot be used to assess compensatory ability quantitatively. It is also unclear about the relationship between compensatory ability with cognitive function or how other factors such as physical exercise modulates compensatory ability. Here, we proposed a data-driven method to semi-quantitatively measure neural compensation under a challenging cognitive task, and we then explored connections between neural compensation to cognitive engagement and cognitive reserve (CR). Functional and structural magnetic resonance imaging scans were acquired for 26 healthy older adults during a face-name memory task. Spatial independent component analysis (ICA) identified visual, attentional and left executive as core networks. Results show that the smaller the volumes of the gray matter (GM) structures within core networks, the more networks were needed to conduct the task (r = −0.408, p = 0.035). Therefore, the number of task-activated networks controlling for the GM volume within core networks was defined as a measure of neural compensatory ability. We found that compensatory ability correlated with working memory performance (r = 0.528, p = 0.035). Among subjects with good memory task performance, those with higher CR used fewer networks than subjects with lower CR. Among poor-performance subjects, those using more networks had higher CR. Our results indicated that using a high cognitive-demanding task to measure the number of activated neural networks could be a useful and sensitive measure of neural compensation in older adults.

Highlights

With the global increase in the aging population, there is an urgency to better understand neural mechanisms of age-related cognitive decline and resilience in order to promote healthy brain aging (Christensen et al, 2009)
In the present study, using an independent component analysis (ICA) approach, we found that among older cognitively intact adults performing a cognitively challenging task, there was a significant increase in the number of active neural networks when the task-related core networks were smaller in volume
We proposed the number of activated neural networks controlling for the volume of the core networks as a measure of neural compensatory capacity

Summary

INTRODUCTION

With the global increase in the aging population, there is an urgency to better understand neural mechanisms of age-related cognitive decline and resilience in order to promote healthy brain aging (Christensen et al, 2009). Functional neuroimaging studies have reported that, compared with young adults, neural activity increases in older adults in a variety of aging-vulnerable brain regions including the prefrontal cortex, posterior parietal cortex and right parahippocampal gyrus/lingual gyrus (Rajah and D’Esposito, 2005; Greenwood, 2007; Park and ReuterLorenz, 2009; Steffener et al, 2009; Lighthall et al, 2014) To accommodate these changes in activity, the brain may be able to reorganize its functioning to counteract neural decline and to maintain performance (Chanraud and Sullivan, 2014). Physical exercise improves executive function (Colcombe and Kramer, 2003; Lautenschlager et al, 2008; Ji et al, 2017), and gait ability has been shown to be associated with global and executive cognitive function (Atkinson et al, 2007), as well as the activation level of executive network in older adults (Jor’dan et al, 2017) These findings support studying the relationship between gait speed and neural compensatory capability. We hypothesized that, if the number of activated neural networks can reflect neural compensation, : (1) greater damage to core task-relate neural networks should evoke greater number of neural networks as a compensatory mechanism; (2) those individuals who have activated more networks should have better cognition than those who lost capacity to activate additional networks; and (3) factors such as physical activity and gait speed might influence brains’ compensatory function

MATERIALS AND METHODS

RESULTS

DISCUSSION