Maria Neimark Geffen, Rockefeller University
Thursday, 14 May 2009, 10:30am
Center for Neural Science, Meyer Hall, Room 815
4 Washington Place, NY 10003
ABSTRACT:
I will explore in my talk how two sensory systems encode
information about complex signals, whose dynamics are derived from
analysis of the natural environment.
Olfactory processing in the insect antennal lobe is a highly dynamic process, yet it has been studied primarily with static step stimuli. To approximate the rapid odor fluctuations encountered in nature, we presented flickering "white-noise "odor stimuli to the antenna of the locust and recorded spike trains from antennal lobe projection neurons. The responses varied greatly across neurons, and across odors for the same neuron. Surprisingly, this diversity across the population was highly constrained, and most responses were captured by a quantitative model with a small number of parameters. Our simple model for the encoding of dynamic odor stimuli accounts for many prior observations on the population response.
Many natural sounds, such as rustling leaves, or running water, are broad-band sounds, which are correlated across frequencies. We developed a generative model of a library of textured sounds, based on a very small number of parameters. Changes in these parameters evoked large difference in perception of naturalness of the synthetic sounds, as measured through psychophysical experiments in human subjects. We presented these stimuli to awake freely-moving rodents and measured neural responses through multi-tetrode recordings in the primary auditory cortex. I will discuss the preliminary results from the analysis of psychophysical and electrophysiological experiments.