The Fluid Mechanics of Fish Hearing

---

Several studies have shown that fish can determine the range and direction of underwater sound at frequencies ranging from 0.1-1.0 kHz even in the presence of background noise. Humans and other land animals directionalize sound using the time of arrival differences between our two ears. Given that sound speed in water is about five times higher than that in air and the distances between the two ears in fish are no more than a few centimeters, however, fish must use a fundamentally different directionalization mechanism. Most fish have two "inner" ears with no direct fluid connection to their environment. The fish ear consists of three endolymph-filled semicircular canals, each of which contains a bony mass, the otolith, suspended <100 microns above the macular membrane densely covered with more than 100,000 hair cells (similar to those found in our own ears). Incident sound oscillates the otolith, with its greater inertia, with respect to its surroundings (Figure 1). These otoliths have amazingly varied geometries (Figure 2); marine biologists can identify the specific species and age of individual fish from just their otoliths. Given that biological systems are optimal, why do fish need complex otolith geometries and so many hair cells?

Figure 1 Otolith oscillating above the macular hair cells [1]

Figure 2 Typical saccular otoliths for goldfish (left) [2] and cod (right) [1]; note the cross-sectional views of the goldfish otolith

Current models of fish hearing assume that fish determine the direction of incident sound by detecting otolith motion along the direction of the acoustic wave--but these models fail to explain why fish need complex otolith geometries or so many hair cells. The incident sound also creates a flow between the otolith and the macula, however. We hypothesize instead that the fish ear functions as an "auditory retina". In this hypothesis, the densely packed hair cells visualize the flow patterns due to the acoustically induced flow in the complex three-dimensional geometry between the otolith and the macula, much like a tuft visualization. The complex geometry of fish otoliths may help to distinguish flow patterns for sound from different directions. By converting acoustic signals into spatial patterns sampled with extremely high spatial resolution by the macular hair cells, directionalizing sound becomes a pattern recognition problem, not unlike the visual patterns imaged by the retina.

The objectives of this research involving marine biology, acoustics and fluid mechanics are:

• To experimentally and numerically study these (stready streaming) flows (at nearly zero Reynolds numbers) in model fish ears using particle-image velocimetry and finite-element and/or finite-volume techniques, respectively. How do otolith geometry and incident sound direction (i.e., otolith oscillation) direction affect these flows? Could fish use these flows to directionalize sound?
• Use this knowledge to design and build a compact underwater acoustic sensor for detecting the direction of incident sound.

This project, a collaboration with P. Rogers and D. Trivett in Mechanical Engineering, is supported by the National Science Foundation and the Office of Naval Research.

References

[1] Platt, C. and Popper, A.N. (1981) "Fine structure and function of the ear" In Hearing and Sound Communication in Fishes (W.N. Tavolga, A.N. Popper, and R.R. Fay, editors), Springer-Verlag, Berlin, pp. 3–36
[2] Schellart, N.A.M. and. Wubbels, R.J (1998) "The auditory and mechanosensory lateral line system" In The Physiology of Fishes (D.H. Evans, editor) CRC Press, New York, pp. 283–312

Publications (contact M. Yoda for reprints)

1. C. W. Kotas, M. Yoda, and P. H. Rogers (2007) Visualization of steady streaming near oscillating spheroids. Experiments in Fluids 42: 111–121, DOI: 10.1007/s00348-006-0224-8
2. C. W. Kotas, M. Yoda, and P. H. Rogers (2006) Visualizations of steady streaming at moderate Reynolds numbers. Physics of Fluids 18:091102, DOI: 10.1063/1.2335902
3. M. Yoda, M., P. H. Rogers and K. E. Baxter (2002) Is the fish ear an auditory retina? Steady streaming in the otolith-macula gap. Bioacoustics 12: 131–134

Presentations

1. C. W. Kotas, M. Yoda, P. H. Rogers. (2007) Visualizing Sound in the Fish Ear. Poster presentation, Underwater Sensors Networks Workshop, Atlanta, GA
2. C. Kotas, P. H. Rogers, M. Yoda. (2006) Flows near model otoliths and their implications for fish hearing. 4th Joint Meeting of the Acoustical Society of America and the Acoustical Society of Japan, Honolulu, HI
3. C. Kotas, P. H. Rogers, M. Yoda. (2006) Flows around oscillating grooved spheroids. 59th Annual Meeting of the American Physical Society Division of Fluid Dynamics, Tampa, FL
4. C. Kotas, P. H. Rogers, M. Yoda. Are acoustically induced flows relevant to fish hearing? 151st Meeting of the Acoustical Society of America, Providence, RI
5. C. W. Kotas, M. Yoda, P. H. Rogers. (2006) Acoustically induced flows in the fish ear. International Symposium for Biologically Inspired Design and Engineering, Atlanta, GA [Invited presentation]
6. C. W. Kotas, M. Yoda, P. H. Rogers. (2006) Visualizations of steady streaming flows: Could fish ears be "auditory retinas"? Poster presentation, Ibid.
7. C. Kotas, P. H. Rogers, M. Yoda. (2005) Flows around oscillating bodies at low Reynolds numbers. 58th Annual Meeting of the American Physical Society Division of Fluid Dynamics, Chicago, IL
8. C. Kotas, M. Yoda, P. H. Rogers. (2005) Visualization of steady streaming at moderate Reynolds numbers. Poster presentation, Ibid
9. C. Kotas, P. H. Rogers, M. Yoda. (2005) Design of an oscillating flow test chamber for modeling the fish ear. 149th Meeting of the Acoustical Society of America , Vancouver, Canada
10. C. Kotas, M. Yoda, P. H. Rogers. (2003) Low Reynolds number steady streaming around a cylinder at various orientations. 56th Annual Meeting of the American Physical Society Division of Fluid Dynamics, East Rutherford, NJ