A Model of a Temporal Filter in Chemoreception to Extract Directional Information From a Turbulent Odor Plume

Open Access
Publisher Website
Google Scholar
Abstract
We ask whether animals can derive spatial information from temporal patterns contained in turbulant odor plumes under realistic biological constraints of receptor properties (size and physiological responses) and behavioral requirements (time averaging). We modeled an appropriately scaled aquatic odor plume with a salt tracer to serve as the input to two different lobster chemoreceptor organs. We then constructed a computer model based on some of the currently known temporal filtering characteristics of lobster chemoreceptor cells in situ. The output of this model represents the supra-threshold stimulus intensity fluctuations "seen" by realistically adapting cells. The input and output of the model were evaluated for directional information. We focused on four parameters that characterize concentration peaks within the plume: height, length, maximum rising slope, and off time (time between peaks). These characteristics were analyzed under two biologically important sampling strategies: one corresponding to a continuous-sampling receptor organ (e.g., lobster leg, catfish nose) and the other to a discrete-sampling receptor organ (e.g., lobster nose, tuna nose). We let the discrete-sampling model analyze at a frequency of four sniffs per second, each averaging over 100 ms. The continuous-sampling model used an historic exponential average of 25, 100, or 1000 ms based on disadaptation rates of receptor cells in situ. In this preliminary study, filtered odor spectra contained less biologically useful information than the unfiltered input spectra. Discrete and continuous models were not different. In all cases, the probability distribution of maximum rising slopes of stimulus concentration contained the most reliable directional information.

This publication has 0 references indexed in Scilit: