Logo image
Sulcal morphology and cognitive aging: posterior middle frontal sulci as structural markers of executive function and resting-state network organization in older adults
Dissertation

Sulcal morphology and cognitive aging: posterior middle frontal sulci as structural markers of executive function and resting-state network organization in older adults

Bryan Madero
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Spring 2026
DOI: 10.25820/etd.008383
pdf
Madero_Thesis_04172612.10 MB
Embargoed Access, Embargo ends: 06/29/2028

Abstract

Individual differences in cortical morphology, specifically tertiary sulci like the posterior middle frontal sulci (pmfs), are obscured by standard template-based neuroimaging approaches, yet may be key to explaining variability in age-related executive function (EF) decline. pmfs develop late in gestation, mature through early development, and in younger adults predict EF performance. Whether pmfs explain EF variability in aging, and how they structurally relate to functional network organization, remains unknown. The goal of this dissertation is to test the hypothesis that the variability in pmfs presence indicates structural differences that inform functional network spatial organization and in turn explain variability of EF performance in older adults. I predict that older adults with more pmfs will have faster reaction times on an EF task and show distinct patterns of network organization compared to those with fewer pmfs. To test this prediction, three separate Aims are carried out to characterize EF decline in older adults, to determine a brain structure and behavior relationship, and determine differences between pmfs presence and absence. In the first aim, age-related declines in EF were characterized by leveraging a task-switch paradigm whose measures characterized sub-processes of EF and was designed with two versions to measure increasing cognitive load in older adults. Results showed that older adults, compared to younger older adults, had slower reaction times on both task versions. Switch cost in the Overlap version increased in the older adults compared to younger adults. In contrast, in the Overlap versions, older adults showed a decrease in mixing cost compared to younger adults. These findings showed that older adults are slower and show increasing switch cost under higher cognitive demands. Mixing cost however shows a decrease though importantly the decrease does not account for the general slowing throughout the task for the older adults. In the second aim, pmfs presence was used to explain brain behavior relationships of EF. Findings showed that individuals with greater pmfs had smaller mixing cost and showed faster working memory indexed by trials in the mixed block that repeated a task. pmfs did not show a relationship with a processing speed EF task. Two additional post hoc analyses were conducted to investigate if pmfs showed better model performance than cortical thickness. A second post hoc tested if other sulci explained performance to see if pmfs showed specificity. Results for the post hoc test showed that pmfs models explained the variability better than cortical thickness and at times was complementary to cortical thickness. The central sulcus did not show a relationship with performance and only the inferior frontal sulcus on the right hemisphere showed a relationship with smaller switching cost in the right hemisphere. These findings highlight the brain behavior relationship of pmfs to explain individual differences in EF performance. In the third aim, pmfs structural contribution to association networks was characterized using a data driven approach to account for individual differences in spatial network boundaries. Results showed, only in the left hemisphere, that individuals with fewer pmfs showed amplified within-network signal magnitude across association, sensory, and motor network components. In contrast, those with more pmfs showed attenuated anticorrelation between association network regions, reflecting reduced between-network suppression. Both patterns are interpreted as distinct expressions of neural dedifferentiation within the r-STAC framework. Spatial alignment between structural pmfs and networks showed that left hemisphere pmfs allegiance to networks is more distributed across individuals while right hemisphere pmfs showed specific spatial overlap with association networks supporting EF. Together, this work extends our knowledge of the importance of identifying and accounting for individual differences in brain structure and its impact on networks and behavior. It tests for the first time the role pmfs play in older adults and age-related declines in EF performance and characterizes how they affect network organization. This research identifies the potential for pmfs to be treated as sensitive age-related brain biomarkers that captures individual differences template-based approaches obscure, offering a more sensitive tool for tracking cognitive aging trajectories, with direct implications for early identification of at-risk individuals in ADRD.
Cerebral Cortex Cognitive Aging Executive Function Individual Differences

Details

Metrics

1 Record Views
Logo image