Developing New Experimental Techniques to Understand Neuronal Networks

Thumbnail Image
Aghayee_umd_0117E_22194.pdf(3.13 MB)
No. of downloads:
Publication or External Link
Aghayee, Samira Sadat
Losert, Wolfgang
Studying the propagation of action potentials across neuronal networks and how information is stored and accessed in their dynamic firing patterns has always been the essence of neuroscience. Emerging evidence shows that information in the brain is encoded in the simultaneous or avalanche-like firing of multiple, spatially separated groups of neurons. Thus, understanding the collective behavior of neurons is essential for understanding how the brain processes information and encodes memory. Since its discovery, the advent of optogenetics has brought upon a revolution in neuroscience, where individual neuronal circuits are able to be selectively probed and their connections decoded. This ability has been used by many groups to great effect, with some groups even using optogenetic stimulation to create phantom sensations, which are typically encoded in the functional activity of distinct neuronal populations. However, in-vivo optogenetic excitation relies inherently on the quality and accuracy of the stimulation method, with many problems arising due to biological effects such as animal motion, the scattering nature of brain tissue, and cell health. Typically, groups either use digital micromirror arrays or spatial light modulators, with the former lacking transmission efficiency and the latter having a high technical skill barrier due to its propensity to induce artifacts into intended patterns of light. This dissertation attempts to reduce the barrier towards the use of spatial light modulators in optogenetics by improving targeting accuracy, reducing the effects of unmodulated light and related artifacts, and developing new methods of stimulation which reduce the power density directed at neurons. To accomplish the first step, improving targeting accuracy, I created and demonstrated a real-time capable particle-based motion tracking algorithm to correct for animal motion. To reduce the effects of optical artifacts, I developed and patented a method of using Fresnel lenses convolved with intended light patterns to project higher orders of diffraction and un-diffracted light axially away from the object plane. To improve cell health during stimulation, I researched the use of optical vortices to stimulate neurons, allowing for ion channel activation with reduced power per unit area. Finally, I set the stage for new science by creating neuroimaging platforms integrating these techniques and capable of imaging activity across multiple scales. Other avenues for improvement are outlined as well in this dissertation, as well as new scientific questions that can be asked, leveraging these developments contained within.