Caltech Library System

About | Browse | Search | Caltech Student Instructions

Type of Document	Dissertation
Author	Jayaraman, Vivek
URN	etd-05192007-195030
Persistent URL	http://resolver.caltech.edu/CaltechETD:etd-05192007-195030
Title	Neural circuit dynamics and ensemble coding in the locust and fruit fly olfactory system
Degree	PhD
Option	Computation and Neural Systems
Advisory Committee	Advisor Name Title Erik Winfree Committee Chair Christof Koch Committee Member Erin Schuman Committee Member Gilles Laurent Committee Member Masakazu Konishi Committee Member Michael Dickinson Committee Member
Keywords	dimensionality reduction neuroscience neural coding sensory systems drosophila systems neuroscience
Date of Defense	2006-10-23
Availability	unrestricted
Abstract Raw sensory information is usually processed and reformatted by an organism’s brain to carry out tasks like identification, discrimination, tracking and storage. The work presented in this dissertation focuses on the processing strategies of neural circuits in the early olfactory system in two insects, the locust and the fruit fly. Projection neurons (PNs) in the antennal lobe (AL) respond to an odor presented to the locust’s antennae by firing in slow information-carrying temporal patterns, consistent across trials. Their downstream targets, the Kenyon cells (KCs) of the mushroom body (MB), receive input from large ensembles of transiently synchronous PNs at a time. The information arrives in slices of time corresponding to cycles of oscillatory activity originating in the AL. In the first part of the thesis, ensemble-level analysis techniques are used to understand how the AL-MB system deals with the problem of identifying odors across different concentrations. Individual PN odor responses can vary dramatically with concentration, but invariant patterns in PN ensemble responses are shown to allow odor identity to be extracted across a wide range of intensities by the KCs. Second, the sensitivity of the early olfactory system to stimulus history is examined. The PN ensemble and the KCs are found capable of tracking an odor in most conditions where it is pulsed or overlapping with another, but they occasionally fail (are masked) or reach intermediate states distinct from those seen for the odors presented alone or in a static mixture. The last part of the thesis focuses on the development of new recording techniques in the fruit fly, an organism with well-studied genetics and behavior. Genetically expressed fluorescent sensors of calcium offer the best available option to study ensemble activity in the fly. Here, simultaneous electrophysiology and two-photon imaging are used to estimate the correlation between G-CaMP, a popular genetically expressible calcium sensor, and electrical activity in PNs. The sensor is found to have poor temporal resolution and to miss significant spiking activity. More generally, this combination of electrophysiology and imaging enables explorations of functional connectivity and calibrated imaging of ensemble activity in the fruit fly.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    VivekJayaramanPhDThesis.pdf 5.41 Mb 00:25:02 00:12:52 00:11:16 00:05:38 00:00:28