LOCAL MOTION DETECTION: COMPARISON OF HUMAN AND MODEL OBSERVERS A Dissertation for the Department of Neuroscience at the University of Pennsylvania

In the last chapter we introduced a model for local translation detection we called a ’planar power detector’. The purpose of this chapter is to experimentally test several qualitative properties of the model. The experiments involve comparing detection performance across a novel set of stochastic stimuli buried in white noise. The chapter is divided into several sections. In the first section, the basic properties of the model to be tested are outlined. The stochastic stimuli are then introduced, and the ideal observer for their detection is discussed. In the next section the experimental logic is explained and predictions are generated. The remaining sections present the results of the experiments and a general discussion.