The purpose of any sensory system is to filter and transform environmental energies into CNS information which is the basis of behavioral decisions about environmental conditions. Sensory systems have filters that perform a variety of complex tasks, such as contrast enhancement, intensity amplification, spectral coding, and frequency detection, that allow for the extraction of information from environmental signals. Sensory filters can either be physiological filters, i.e., physiological properties of sensory cells, odorant binding proteins, enzymatic degradation of signals, secondary inhibitory or excitatory neural connections, or biomechanical filters, i.e., the physical shape and properties of lenses for vision, arrangement of microvilli of arthropod reticular cells, the basilar and tectorial membrane in mammalian audition, or boundary layers around chemosensory appendages. The purpose of this area of research for our lab is to begin a comprehensive series of studies designed to characterize biomechanical filters and the role that they play in filtering stimulus energies in chemoreception.

Once the structure of natural stimulus patterns has been quantified, the next step is to analyze how the biomechanic filters of a sensory system alter the environmental stimulus patterns that actually arrive at receptor cell surfaces. This idea is easily understood with analogies from other sensory systems. For vision, once the physics of light propagation is known, we can examine how the shape and construction of the lens serve to focus and filter the light arriving at the photoreceptors. For arthropods, the perpendicular placement of microvilli in the reticular cells allows the sensory system to detect plane polarized light. In both of these situations, it is not the physiological properties of receptor cells that give rise to the system properties, but the physical structure of the receptor organ that serves to filter sensory signals or add additional sensory properties to the system. For chemoreception, the biomechanical or physical filters are the boundary layers and microscale fluid transport processes associated with structurally distinct chemoreceptive organs in a moving fluid.

The relationship between fluid flow and chemoreception

The chemical senses are unique among all of the senses in that the physical process of transmission through a medium (such as fluid flow) is independent of any inherent excitatory properties of the receptor-activating signal. This is not true for light, as the quality of light (spectral frequency or wavelength) directly influences different transmission impediments such as scattering, absorption, and attenuation. In addition, the frequency (or wavelength) of a sound is a critical factor in determining propagation through various media and in the reflection and diffraction involved in echolocation. For chemoreception, there are only two physical processes of transmission; fluid flow and molecular diffusion. The relative roles that each of these two processes play in dispersing chemicals can be quantified by the Peclet number (ul / D_m), where u is fluid velocity, l is the characteristic length scale taken along the direction of fluid flow, and D_m is the molecular diffusion coefficient. The Peclet number is the ratio of transport by convection (ul) to that by diffusion (D_m); values larger than 1 indicate that convection is the dominant process, while values less than 1 indicate that molecular diffusion is dominant. Diffusion coefficients for amino acids in water, which are typical aquatic feeding signals, range from 0.5 to 2 x 10^-5 cm²/s. Using the greater value for D_m and solving for a Peclet number of 1 allows us to quantify at what flow velocity or length scale diffusion becomes an important dispersion process. It is evident from this calculation that, for average flow situations in an aquatic environment (< 1 cm/s) and for macroscopic size scales (< 100 microns), flow is the dominant dispersal processes.

The optimal method for initially characterizing the forces involved in any flow situation is to calculate the Reynolds number of the flow field. The Reynolds number (Re = ul /n) is an indicator of the relative importance that viscous or inertial forces play in determining the type of fluid motion. When Re numbers are high, flows are turbulent and turbulent mixing (which is a rapid process) occurs. Conversely, low Re numbers indicate that either laminar or highly viscous flow is present and turbulence is absent. Thus, it is important to first quantify the fluid dynamics of a particular situation before we can begin to quantify the dynamics of dispersion, and hence, the spatial and temporal dynamics of the chemical signal arriving at receptor cells.

Finally, any solid object (such as a sensory appendage) placed within a moving fluid (either air or water) will have a boundary layer associated with it. This is due to the physical condition that any fluid in direct contact with the surface of a solid does not move relative to that surface. Since the dynamics of the chemical signal are due to fluid transport, there will also be a gradient of chemical dynamics surrounding a solid surface. Boundary layers are easily calculated for simple flows and morphologically simple structures such as flat planes or spheres. Unfortunately, chemosensory appendages are usually quite complex and often reside in unsteady laminar or turbulent flows. These two conditions make theoretical models of fluid flow quite complicated and often require empirical determination of boundary layer structures and the development of chemical transport models.

Research in our lab has focused on describing the biomechanical filters associated with various chemosensory appendages. As the table below demonstrates, we have done this for both terrestrial and aquatic organisms. Our goal is to understand the physical processes involved in the perception of chemical signals.