Please use this identifier to cite or link to this item:
Type: Thesis
Title: Neural responses to moving natural scenes.
Author: Straw, Andrew D.
Issue Date: 2004
School/Discipline: School of Molecular and Biomedical Science : Physiology
Abstract: Visual movement is important to most animals that move quickly, and even some that do not. What neural computations do animals use to see visual motion in their natural environment? The visual stimulus used to perform experiments on such questions is critical, and has historically limited the ability to perform experiments asking critical questions about responses to naturalistic moving scenes. The ability to display, at high frame rates, moving natural panoramas and other stimuli distorted to compensate for projection onto a flat screen was important to the experiments described here. I therefore created a software library called the 'Vision Egg' that allows creation of motion stimuli with recent, inexpensive computer hardware, and was used for the experiments described here. Additionally, I developed a mathematical model to determine the quality of motion simulation possible with computer displays. This model was applied to reach an understanding of the 'ghosting' artifact sometimes perceived on such apparent motion displays. Psychophysical experiments on human observers confirmed model predictions and allowed testing of synthetic motion blur for simulation of smooth motion and elimination of the ghosting artifact. I show this synthetic motion blur is optimal in the sense of creating the closest perception possible to that of smooth motion experienced in natural settings. Experiments on humans and flies show that such synthetic 'motion blur' has no effect on motion detection per se. However, ghosting in sampled displays results in information not present in smooth motion at high velocities, permitting inappropriate discrimination of rapidly moving features. I performed experiments measuring the responses of hoverfly wide-field motion detecting neurons (HS cells) in adapted and unadapted states to the velocity of natural scenes. Responses to natural images of varied intrinsic contrast depend little on the choice of image. Artificially reducing contrast, however, does reduce response magnitudes. Finally, the greatest component of response variation to natural scenes is directly related to local structure in the scenes, and could thus be called 'pattern noise.' The large receptive field of HS cells arises from a (non-linear) spatial summation of numerous elementary motion detectors. I measured spatial and temporal contrast sensitivity of small patches in the large receptive field. As predicted from the presence of a frontal optical acute zone, spatial tuning is highest frontally. A sexually dimorphic 'bright zone' in the frontodorsal eye is correlated with enhanced contrast sensitivity and faster temporal tuning in HS cells with receptive fields in this region of male flies.
Dissertation Note: Thesis (Ph.D.) -- University of Adelaide, School of Molecular and Biomedical Science, 2004
Keywords: visual motion; stimuli; neural responses
Provenance: Copyright material removed from digital thesis. See print copy in University of Adelaide Library for full text.
This electronic version is made publicly available by the University of Adelaide in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. This thesis may incorporate third party material which has been used by the author pursuant to Fair Dealing exception. If you are the author of this thesis and do not wish it to be made publicly available or If you are the owner of any included third party copyright material you wish to be removed from this electronic version, please complete the take down form located at:
Appears in Collections:Research Theses

Files in This Item:
File Description SizeFormat 
01front.pdf183.26 kBAdobe PDFView/Open
02whole.pdf12.72 MBAdobe PDFView/Open
  Restricted Access
Library staff access only564.29 kBAdobe PDFView/Open
  Restricted Access
Library staff access only13.13 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.