Dissertation
Uncovering the latent structure of visuospatial ability and predicting visuospatial dysfunction from lesion location
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Summer 2023
DOI: 10.25820/etd.007201
Abstract
Visuospatial ability is the ability to perceive, understand, and utilize visual sensory data to construct a spatial map of one’s environment. Visuospatial dysfunction commonly occurs following brain lesions and often improves in the first few months post-injury; however, some patients continue to have chronic visuospatial dysfunction which can have a devastating impact on recovery trajectory, the ability to live independently, and quality of life. Prognoses for these patients are often made based on clinical intuition, but an evidence-based approach for predicting long-term visuospatial dysfunction in the acute epoch of recovery remains elusive. This study sought to identify the latent structure of visuospatial ability, investigate brain regions and networks involved in visuospatial ability independent of the tests used to measure it, and develop a model to predict long-term visuospatial dysfunction in two out-of-sample cohorts. Exploratory factor analysis was used to identify potential latent variables representing domain-general visuospatial ability across twelve neuropsychological tests in patients from the Iowa Neurological Patient Registry (n = 480). One latent variable, “domain-general visuospatial ability”, described variance in this sample best. Brain regions and networks associated with domain-general visuospatial ability were identified using multivariate lesion-symptom mapping and lesion network mapping. Finally, models were compared to determine the best predictors of long-term visuospatial dysfunction in the Benton Neuropsychology Clinic cohort (n = 80) and the Washington University cohort (n = 104). Domain-general visuospatial ability was associated with right hemisphere damage to the putamen and bilateral damage to posterior white matter tracts constituting the dorsal visual stream in the lesion-symptom map. After controlling for general cognitive ability, only right white matter damage was implicated. Damage to regions in the dorsal attention network and visual network along with posterior inter- and intra-hemispheric white matter was also associated with visuospatial dysfunction. A model including functional lesion network mapping, structural lesion network mapping, and lesion volume best predicted outcomes in both validation cohorts. This study provides novel evidence for the right putamen’s importance in visuospatial ability and confirms the previously identified involvement of the dorsal visual stream, dorsal attention network, and visual network in visuospatial ability. Furthermore, my predictive model is one step towards the broader goal of predicting long-term outcomes of brain damage in the acute epoch.
Details
- Title: Subtitle
- Uncovering the latent structure of visuospatial ability and predicting visuospatial dysfunction from lesion location
- Creators
- Jax Dimitri Skye
- Contributors
- Aaron Boes (Advisor)Daniel Tranel (Advisor)James Traer (Committee Member)Justin Sipla (Committee Member)Jeffrey Long (Committee Member)
- Resource Type
- Dissertation
- Degree Awarded
- Doctor of Philosophy (PhD), University of Iowa
- Degree in
- Neuroscience
- Date degree season
- Summer 2023
- Publisher
- University of Iowa
- DOI
- 10.25820/etd.007201
- Number of pages
- xvii, 125 pages
- Copyright
- Copyright 2023 Jax Skye
- Language
- English
- Date submitted
- 07/21/2023
- Description illustrations
- Illustrations, tables, graphs, charts
- Description bibliographic
- Includes bibliographical references (pages 106-125).
- Public Abstract (ETD)
The brain processes information from the eyes to create a map of spatial relationships that is used for things like navigation, drawing, rotating objects in one’s mind, and guiding movement to avoid running into objects. Damage to the brain can affect someone’s ability to do this, a deficit called visuospatial dysfunction. After brain damage, this problem can resolve on its own in some, but not all, people. Visuospatial dysfunction can impact multiple areas of everyday life, and predicting long-term recovery in the days following brain damage is challenging. This study had three main goals: 1) identify which sub-processes comprise visuospatial ability, 2) identify regions and networks in the brain that support visuospatial ability, and 3) develop mathematical models that can predict long-term visuospatial dysfunction. Analyses were performed in 664 people with brain damage. I identified regions in the brain, the right putamen and the dorsal visual stream, and networks in the brain, the dorsal attention network and the visual network that are important for visuospatial ability. The significance of the dorsal visual stream for visuospatial ability has been well described, but the involvement of the right putamen in visuospatial ability has not been reported in humans. These results were used to create a mathematical model that predicts visuospatial dysfunction using brain networks and the size of a brain lesion as predictors. My findings contribute to the broader goals of predicting outcomes from brain injury using mathematical models and better understanding the functional neuroanatomy of visuospatial ability.
- Academic Unit
- Interdisciplinary Graduate Program in Neuroscience
- Record Identifier
- 9984454319202771
Metrics
4 File views/ downloads
28 Record Views