Memory Representations in Aging

点击翻译按钮获取中文摘要

Project Summary/Abstract Episodic memory shows a robust decline in aging. Decades of neuroimaging research has yielded a wealth of information on the neural mechanisms of age-related episodic memory decline, but a particular limitation of previous studies is that they focused on memory processes (operations performed on memories) and not memory representations (the content of memories). It is likely that memory representations also decline with age, making the investigation of memory representations essential to providing novel insight into mechanisms underlying age-related episodic memory decline, which is important for the development of interventions for memory rehabilitation. Particularly hindering future intervention development, it is unclear (1) how memory representations are affected by age-related differences in the contributions of perceptual vs. conceptual processes, which change across the lifespan and may be targeted in an intervention, and (2) how the structural basis of memory representations changes across the lifespan, which is necessary to understand for an intervention ameliorating age-related memory representation deficits to have long term effects. Here, the applicant proposes an innovative two-prong approach that examines age-related differences in the influence on memory representations of perceptual vs. conceptual processes and brain structural networks. The proposed project will provide mentored training in (1) the assessment of brain/neurocognitive changes in younger and older adults, (2) advanced MRI analysis (e.g., graph theory, deep convolutional neural networks) and (3) multimodal imaging analysis (functional and structural), which are necessary for utilizing the two-prong approach, and (4) career development. The training plan includes coursework, workshops, one- on-one mentorship, participation in seminars, directed readings, and attendance at national and international conferences. In Study 1, while undergoing fMRI, 36 younger adults and 36 older adults study pictures of complex scenes and later retrieve their memories of the scenes. Participants’ use of perceptual and conceptual processes are experimentally manipulated. The goal is to link age-related memory representational quality deficits to age effects in perceptual vs. conceptual processing (Aim 1). The applicant investigates this aim with representational similarity analyses combined with cutting-edge deep convolutional neural networks. In Study 2, the applicant also collects diffusion weighted images to link age-related memory representation quality deficits to age effects on structural brain networks (Aim 2). The applicant investigates this aim with novel graph theory and multimodal imaging (functional and structural) analyses. This investigation using a two- prong approach will provide a novel viewpoint on mechanisms underlying age-related episodic memory decline, which is essential for intervention development. Furthermore, at the culmination of this fellowship, the applicant will have a specialized skill set and be well prepared to begin the next phase of training.