Date of Award

Spring 1-1-2017

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

First Advisor

Naomi P. Friedman

Second Advisor

John K. Hewitt

Third Advisor

Marie T. Banich

Fourth Advisor

Kenneth P. Wright

Fifth Advisor

Matthew C. Keller

Abstract

This dissertation presents four studies that examined how executive functions (EFs) relate to problematic behaviors, such as psychopathology and atypical sleep, as well as how EFs are characterized in healthy individuals. The first study examined whether genetic risk for five different forms of psychopathology predicted EFs at the latent variable level in a population sample. The second study characterized the neural activation patterns in health individuals in response to 3 different types of EF tasks. This study (a) assessed the overlapping activation elicited across the different EF task types, and (b) used neural activation to predict high or low EF ability. The third study examined whether individual differences in sleep duration influenced either depression or EFs (a) within a single time of assessment, (b) across time, and (c) how sleep duration, depression, and EFs influence each other in the same model. Study 4 decomposed the relationships of sleep duration with depression and then EF into their genetic and environmental factors in order to better understand the underlying architecture for these relationships.

In the first study, publically available datasets from the Psychiatric Genomics Consortia were used to generate polygenic risk scores for 5 different psychiatric disorders: Autism, Attention Deficit Hyperactivity Disorder (ADHD), Bipolar Disorder, Major Depressive Disorder (MDD), and Schizophrenia. I then used a deeply phenotyped (and genotyped) subset of 354 twins in the Colorado Longitudinal Twin Study (LTS) from the University of Colorado Boulder to test whether or not genetic risk in these individuals predicted EF abilities. I also examined whether the appropriate risk scores were associated with ADHD and MDD symptoms or lifetime diagnoses to the same relative extent as the EF scores. Results indicated polygenic risk for psychopathology did not significantly predict EFs after controlling for multiple testing. Results also suggested that effect sizes for EFs were comparable to those for ADHD and MDD symptoms and lifetime diagnoses.

The second study was a pilot study that included 30 subjects from the Colorado Twin Study from the University of Colorado Boulder at approximately age 28. These subjects were chosen because they were either high or low in Common EF ability as measured in a previous wave of data collection 7 years prior, until we had 15 of each. Each subject completed 3 EF tasks in a functional magnetic resonance imaging scanner. Results indicated that common brain activation in response to these tasks both overlapped with a frontoparietal network typically associated with cognitive tasks, and extended beyond this network. The common areas associated with individual differences in EF ability fell outside of the frontoparietal network.

The third and fourth studies utilized data from the same group of 857 twins from the LTS sample. These studies examined sleep, depression, and EF when available at approximately ages 12, 17, 21, and 23. Study three looked at the phenotypic relationships between these variables and found linear and nonlinear relationships between sleep duration and EFs and depression across age. When put together in the same model, depression seems to suppress the relationship between EF and sleep duration in adolescence. Study four results showed that sleep duration is moderately heritable, and that the phenotypic relationships between these variables is typically attributable to non-shared environmental influences.

Share

COinS