Stanley Klein
School of Optometry, UC Berkeley and Vision Science and Neuroscience Graduate Groups

Limits of Vision and psychophysical methods

Tuesday 29th of November 2005 at 04:00pm
5101 Tolman

I was asked to present our psychophysics research that has relevance to computational neuroscience. Since I'm local I thought it best to give a brief overview of several of our past projects. People can easily contact me or look at the publications for more details. My goal is to show how psychophysics can be a powerful neuroscience tool.

1) Perception vs. psychophysics. How to measure subjectivity objectively. I'll present a brief introduction to signal detection theory and how one can remove subjective biases from experimental results. I'll introduce the topic of optimal approaches for measuring the psychometric function. http://cornea.berkeley.edu/pubs/01-11_PERCEPTION_AND_PSYCHOPHYSICS-Measuring_estimating_and_understanding_the_psychometric_function.pdf

2) Double Judgment Psychophysics provides a powerful, but easily misused tool for exploring interactions among mechanisms. Exotic mechanisms vs exotic decisions will be discussed. http://cornea.berkeley.edu/pubs/34.pdf

3) The Guinness record for vision involves a position task with a close reference. Spatial vision modeling indicates that mechanisms localized in space and spatial frequency are used with surprisingly minimal gain control. Scale-space maps are presented showing vision may be similar to audition. http://cornea.berkeley.edu/pubs/33.pdf

4) Minimizing space-spatial frequency uncertainty. Why Gabor was wrong! http://cornea.berkeley.edu/pubs/63.pdf

5) Temporal dynamics of machine-like ganglion cells with context dependence of firing time. The stimulus was a pair of binary white noise stimuli, flashing on the center and the surround of a ganglion cell's receptive field. The state of the stimulus more than 90 msec in the past reliably shifted the timing of spikes by a few msec. http://cornea.berkeley.edu/pubs/OptimizingEstimationNonlinearKernelsSKlein.zip

6) Rate coding vs. time coding of information. Does the 1 msec temporal asynchrony of moving vernier acuity rule out rate coding? These experiments on moving 3-dot vernier stimuli pull together insights from above items #1, 3 and 5. http://cornea.berkeley.edu/pubs/89.pdf


Join Email List

You can subscribe to our weekly seminar email list by sending an email to majordomo@lists.berkeley.edu that contains the words subscribe redwood in the body of the message.
(Note: The subject line can be arbitrary and will be ignored)