Aging is accompanied by cognitive declines, including in declarative (episodic) memory and in cognitive control, which refers to a set of cognitive functions that align actions with internal goals. Extant evidence indicates that older adults exhibit difficulty in proactively adjusting cognitive control, failing to effectively adjust control based on contextual cues that signal upcoming cognitive control demand. This difficulty leads to reduced flexibility, impedes goal-directed behavior, and can profoundly impact older people?s daily lives, as contexts are common and powerful predictive cues to signal adjustment of cognitive control. For instance, viewing a usually busy traffic circle can serve as a cue to increase control demand in driving before entering the traffic circle. While age-related deficits in proactive control likely stem from disruptions in memory system and control system processes, and their interaction, little is known about the neurocognitive mechanisms supporting the formation and expression of associations between contexts and cognitive control demand and how these mechanisms change with aging. To answer these questions, three experiments are proposed using a combination of virtual reality behavioral methods and human functional magnetic resonance imaging.
Specific Aim 1 will delineate the mechanisms supporting the encoding and retrieval of associations between context and cognitive control demand, enabling proactive control in young adults. This experiment will also lay the foundations for Specific Aims 2 and 3.
Specific Aim 2 will determine whether and how context-control demand associations can be generalized to objects appearing in the context via the mechanism of memory replay. Finally, Specific Aim 3 will characterize whether reductions in proactive control in older adults are partially due to dysfunctional mnemonic mechanisms that build and express associations between contexts and cognitive control demands.
Specific Aim 3 will further examine whether individual differences in hippocampal anatomical and cerebrospinal fluid biomarkers of elevated risk for Alzheimer?s disease relate to age-related differences in memory for context-control demand associations. In all proposed experiments, we will harness virtual reality techniques to present contextual stimuli (virtual rooms) in order to more closely simulate real life scenarios. FMRI univariate and multivariate pattern analyses will be combined to comprehensively interrogate age-related changes in neurocognitive mechanisms underlying associations between contexts and cognitive control demand. The proposed studies promise to advance understanding of memory?cognitive control interactions, including how memory of procedural processing features (e.g., past levels of control demand experienced in a context) can drive flexible adjustments of proactive control, and will provide valuable new insights into how decreased cognitive control in older adults may stem, in part, from multi-system (memory and control) dysfunction.
Older adults suffer deficits in cognitive control, mechanisms that align neural processing with internal goals. Age-related deficits in flexible adjustments of control may reflect disruptions in memory systems, control systems, and their interaction (e.g., older adults have difficulty using memory of past control demands experienced in a context to proactively adjust control when back in the context at some point in the future), but little is known about the neurocognitive mechanisms underlying the learning and memory of associations between contexts and cognitive control demand, and how aging affects these mechanisms. Using a combination of sophisticated behavioral methods and brain imaging, the proposed research program will advance understanding of memory?cognitive control interactions, and will provide valuable new insights into how decreased cognitive control in older adults may stem, in part, from multi-system (memory and control) dysfunction.
| Jiang, Jiefeng; Wagner, Anthony D; Egner, Tobias (2018) Integrated externally and internally generated task predictions jointly guide cognitive control in prefrontal cortex. Elife 7: |
