Working memory (WM) is a fundamental cognitive ability that declines steadily over the course of normal aging. Mounting evidence suggests however, that with training there is potential for considerable improvement in WM even in older adults (OA). Performance on the WM training tasks can improve markedly and reliably especially when adaptive methods continually adjust task difficulty to changing performance level. While younger adults (YA) typically achieve higher performance than OA and show greater preservation of WM gains over time, OA benefit from WM training, and sustain training benefits even months after training. Critically, the mechanisms underlying such plasticity and the effects of age are largely unknown. Thus, the significant problem addressed by the proposed research is the need to identify the neural markers of working memory plasticity and how they differ due to age. The proposed research uses functional magnetic resonance imaging (fMRI) to investigate the neural bases of improvement on a WM training task in three aims, framed within the Compensation Related Utilization of Neural Circuits Hypothesis (CRUNCH), and other hypothesized mechanisms of plasticity. CRUNCH proposes that the relationship between brain activation and memory load is an inverted-U function regardless of age, as depicted in the specific aims. To compensate for age-related declines, the OA function is shifted left relative to the YA function and consequently OA reach a resource limit at lower loads. This model predicts that training should increase neural efficiency thereby shifting the functions rightward, for both YA and OA, and increasing the range of task demands to which the brain can respond. The CRUNCH framework also makes predictions about how distinctiveness (as measured with multi-voxel pattern analyses, MVPA) and functional connectivity should change due to training. MVPA can provide new evidence about plasticity of domain-specific resources, and connectivity analyses can reveal the plasticity of network circuitry associated WM training. We propose one training experiment (with 20 OA and 20 YA) to conduct a preliminary assessment of (1) univariate, (2) multivariate, and (3) connectivity predictions of CRUNCH in three aims (respectively). The proposed research will test the sensitivity of these measures to WM training in YA and OA in order to demonstrate the feasibility of our approach and establish a strong empirical base to inform power estimates, design and data analysis decisions for a future R01. By identifying the neural correlates of the plasticity underlying training gains, future research cn determine whether such markers can predict the magnitude and maintenance of training gains, the extent and limits of transfer and potentially other cognitive outcomes in YA and OA. The development of interventions that achieve even modest improvements of WM function, especially in the elderly, can potentially maintain effective levels of functioning and stave off further decline.

Public Health Relevance

There is mounting evidence that some computerized working memory (WM) interventions really do improve working memory even in limited ways. Such evidence raises the need to understand why these programs work, especially in older adults. Using brain imaging we aim to identify promising signatures of neurocognitive plasticity of WM training and how they differ due to age. The proposed research will advance our understanding of the benefits and limits of WM training to improve and direct future intervention research which is especially vital to the nation's aging population who confront declining plasticity and cognitiv losses.

Agency
National Institute of Health (NIH)
Type
Exploratory/Developmental Grants (R21)
Project #
1R21AG045460-01A1
Application #
8699998
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Wagster, Molly V
Project Start
Project End
Budget Start
Budget End
Support Year
1
Fiscal Year
2014
Total Cost
Indirect Cost
Name
University of Michigan Ann Arbor
Department
Psychology
Type
Schools of Arts and Sciences
DUNS #
City
Ann Arbor
State
MI
Country
United States
Zip Code
48109
Reuter-Lorenz, Patricia A; Park, Denise C (2014) How does it STAC up? Revisiting the scaffolding theory of aging and cognition. Neuropsychol Rev 24:355-70