Loading...
Journal Title
Keywords
Readers/Advisors
Fenton, Andre A.
Journal Title
Term and Year
Spring 2015
Publication Date
2015-06-19
Book Title
Publication Volume
Publication Issue
Publication Begin
Publication End
Number of pages
Files
Loading...
Doctoral Dissertation
Adobe PDF, 35.33 MB
Research Projects
Organizational Units
Journal Issue
Abstract
Each brain is a living computational system. It was shaped by evolution in order to predict the dynamics of events in its environment, caused by both living and non-‐living objects. A predicting machine is only effective when it can build and apply
its model of reality on the fly. This requires mechanisms for sensory input selection, routing the flow of information between neural networks, organization of available computational resources, storage and recall of memories and mechanisms for
switching between various modes of operations. Neural oscillations are the best known instrument of effective self-‐organization of neural networks. While slow oscillations such as delta (1-‐4 Hz) and theta (5-‐12 Hz) are typically involved in
global control of the network state, sensory input selection and fragmentation of processed information, fast oscillations such as gamma (30-‐90 Hz) appear to be involved in the recruitment of computational assemblies, supporting memory processes and communication within and between networks. Neural oscillations define the cognitive state. The robust quantification of the cognitive state is of major importance because it may serve as a diagnostic tool for the evaluation of abnormal cognitive processing as results from disorders like schizophrenia and autism. It is therefore important to develop robust numerical analytical tools to quantify oscillatory activity.
In this biomedical engineering series of studies we first define the framework for the analysis of neural oscillations based on phase locking and cross-‐frequency coupling and we define algorithmic boundary conditions for robust assessment of these measures of synchrony in neural oscillations. Second, we evaluate the utility of phase locking in the analysis of inter‐ and intra‐hemispheric neural discoordination of local field potentials (LFP) in the neurodevelopmental, neonatal ventral hippocampus lesion (NVHL) rat model that is relevant to schizophrenia. Third, we evaluate the utility of cross-‐frequency coupling in the analysis of local discoordination of LFP signals in the acute phencyclidine (PCP) rat model that is relevant to schizophrenia. Fourth, we evaluate the utility of cross-‐frequency coupling in the analysis of input-‐specific LFP signals in the mouse Fmr1
genetic knockout model that is relevant to intellectual disability in Fragile X Syndrome and autism. Fifth, we propose a new data-‐driven approach for unbiased parameter estimation of cross-‐frequency coupling and a method for high-‐resolution analysis of single oscillatory events and evaluate the method in the place avoidance task as a test for cognitive flexibility
in both wild-‐type and Fmr1 knockout mice.
The ability to quantify neural oscillations as a physiological indicator of cognitive
processing has the potential to enable the future use of such quantitative
physiological measures in the diagnosis and characterization of human patients
suffering from mental disorders, such as schizophrenia and autism.
In my thesis defense lecture, I will focus on the Fmr1 KO mouse model and
describe the relevance of neural discoordination of local field potentials to
cognitive dysfunction. I then present a novel analytical framework for
assessing neural coordination using discreet oscillatory events and then use
the framework to describe memory recall events on the rotating arena in both
wild type and Fmr1 KO mice.
Citation
Dvorak, D. (2015) Measuring Neural Discoordination. [Doctoral dissertation, SUNY Downstate Health Sciences University]. SUNY Open Access Repository. https://soar.suny.edu/handle/20.500.12648/15889
DOI
Description
Doctoral Dissertation