How much power does neural signal propagation need?footnotePresented at theISPIEConf. on BioMEMS and Smart Nanostructures (Adelaide., 2001)/I; the conferenceversion was published in its proceedings./footnote

摘要

Two well known, biologically inspired non-dynamical models of stochasticresonance, the threshold-crossing model and the fluctuating rate model,are analyzed in terms of channel information capacity and dissipation ofenergy necessary for small-signal transduction. Using analogies to spikepropagation in neurons we postulate the average output pulse rate as a measureof dissipation. The dissipation increases monotonically with the inputnoise. We find that for small dissipation both models give a close to lineardependence of the channel information capacity on dissipation. In bothmodels the channel information capacity, as a function of dissipation, has amaximum at input noise amplitude that is different from that in the standardsignal-to-noise ratio versus input noise plot. Though a direct comparison is notstraightforward, for small signals the threshold model gives appreciably higherdensity of information per dissipation than the exponential fluctuatingrate model. We show that a formal introduction of cooperativity in therate fluctuating model permits us to imitate the response function of thethreshold model and to enhance performance. This finding may havedirect relevance to real neural spike generation where, due to a strongpositive feedback, the ion channel currents add up in a synchronized way.