Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


36365.pdf (5.59 MB)

ETD Abstract Container

Abstract Header

Towards Connectionist Neuroimaging: Brain Connector Hubs for Expressive Language

Williamson, Brady
http://orcid.org/0000-0002-1161-402X
http://rave.ohiolink.edu/etdc/view?acc_num=ucin157106095114712

Abstract Details

2019, PhD, University of Cincinnati, Arts and Sciences: Psychology.
Many models attempting to characterize the neural correlates of cognition focus on localizing functions, but few address the question of how the localized regions take part in the network as a whole. Models of language in the brain have existed for nearly 150 years and most fall into this category. While early research was limited to postmortem studies of pathology, the advent of neuroimaging has allowed researchers to investigate brain function in vivo and in healthy populations as well as those with pathology. Even with this advance, many researchers have still largely focused on localization of function. Recently, there has been a paradigmatic shift from activation-based to connectivity-based analyses of neuroimaging data, emphasizing investigation of how different brain regions interact. When conceptualizing the brain as a network, some regions emerge as important for communication between functionally specialized neural populations. These areas are known as connector hubs. While connector hubs have been acknowledged for several years, they are rarely characterized, especially in the context of a specific cognitive function. The current study aims to characterize connector hubs in the pediatric language network, observe their developmental trajectory, and assess their relationship with age-corrected language performance. Data from an existing study were used that included neuroimaging (MRI, fMRI, diffusion MRI or dMRI, MEG) and standardized language assessment (EVT) data for 15 young children (4-6 years) and 15 adolescents (16-18 years). A data-driven, connectivity-based pipeline was developed to characterize connector hubs during an expressive language task. After initial preprocessing, connectivity matrices were calculated for MEG (functional) and dMRI (structural) data. Functional connectivity was calculated using amplitude-amplitude correlation (AAC) within and between 4 different frequency bins (2-4, 4-12, 12-30, 30-100 Hz) to capture intra- and inter- frequency connectivity. Each functional connectivity matrix was then multiplied (entry-wise) by a threholded and binarized structural connectivity matrix to keep only biologically-plausible connections. Finally, the matrices were combined into a multilayer network and graph analyses were performed to identify connector hubs. Results showed good overlap of connector hubs with vetted fMRI-derived maps, though the current pipeline was more consistent across individuals. Additionally, connector hub distribution became more left lateralized, cortical, and focal, with age. Connector hubs in regions thought to be involved in expressive language were significantly correlated with EVT performance in adolescents but not young children. Overall, the results suggest that the expressive language network becomes more focal and efficient during development, a process potentially mediated by central and subcortical regions. Comprehensive characterization of connector hubs and how they change during childhood has important clinical and computational applications. Cutting-edge neuroimaging analyses combined with new developments in connectionist modeling may lead to a better understanding of how neurophysiology leads to cognition.
Michael Riley, Ph.D. (Committee Chair)
Darren Kadis, Ph.D. (Committee Member)
Dieter Vanderelst, Ph.D. (Committee Member)
102 p.
Psychology
Neuroimaging; Network Analysis; Connectionism

Recommended Citations

Citations

  • Williamson, B. (2019). Towards Connectionist Neuroimaging: Brain Connector Hubs for Expressive Language [Doctoral dissertation, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin157106095114712

    APA Style (7th edition)

  • Williamson, Brady. Towards Connectionist Neuroimaging: Brain Connector Hubs for Expressive Language. 2019. University of Cincinnati, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin157106095114712.

    MLA Style (8th edition)

  • Williamson, Brady. "Towards Connectionist Neuroimaging: Brain Connector Hubs for Expressive Language." Doctoral dissertation, University of Cincinnati, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=ucin157106095114712

    Chicago Manual of Style (17th edition)

ucin157106095114712
271
© 2019, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.