Dorian S. Houser, PhD
Marine mammals are important cultural and natural resources, and, because many of them are top-level predators, they also serve as important sentinels of ocean health. Marine mammals have evolved to capitalize on acoustic cues in the ocean, largely because sound travels much more efficiently through water than does light. For the odontocete cetaceans, which are the echolocating dolphins, porpoises, belugas, and narwhals, hearing is the primary sensory modality. Echolocation is used for foraging and navigation, often in turbid waters or at great depth where diminished light penetration limits vision. The echolocating marine mammals rely on the transmission of sound and the reception of echoes to exploit the environment in which they live.
The past two decades have seen a dramatic increase in concern regarding the impact of anthropogenic (human-made) noise on ocean life, particularly that of the marine mammals. Shipping, seismic exploration, naval sonar activity, and recreational boating and fishing all contribute to ocean noise, and ocean noise has significantly increased over the past century. Determining how anthropogenic sound impacts marine mammals has been a daunting task. There are concerns regarding the potential for high amplitude sources to cause permanent or temporary shifts in hearing sensitivity. Other concerns exist over the potential for masking to limit the range over which marine mammals can communicate or detect predators. Disruption of natural behaviors in response to hearing anthropogenic sounds is also of interest, as it can affect social interactions, reproduction, and foraging. Little is known about the potential for impact and its relationship to the myriad of anthropogenic sound types in the ocean. Indeed, knowledge of the range of hearing and hearing sensitivity is unknown for most marine mammal species, meaning the fundamental information for assessing the potential impact of sound exposure is missing.
Traditionally, audiometry in marine mammals has been performed through behavioral methods; a subject (e.g., a dolphin) is played a sound, and it reports detection of the sound by vocalizing or touching a paddle. Such psychophysical procedures, though broadly accepted as the most accurate way to assess hearing ability, are time consuming and require long-term access to research subjects. For this reason, behavioral audiograms representing a marine mammal species' hearing capabilities are determined from studies with one to a handful of subjects. Furthermore, because few marine mammal species are maintained by humans, the hearing of a small percentage of the 130+ species of marine mammal has been tested.
Auditory evoked potentials (AEPs), which are small voltages generated by the brain when an animal hears a sound, can be monitored for the rapid assessment of hearing sensitivity and are commonly applied to the evaluation of hearing in humans. Over the past decade, the United States Navy Marine Mammal Program (MMP) and the National Marine Mammal Foundation (NMMF) have collaboratively developed and applied AEP methods for use in marine mammal audiometry. The rationale for the effort was threefold: to improve the clinical care of dolphins and sea lions employed by the MMP by enabling routine hearing assessments, to test a larger number of marine mammals for which some audiometric information exists in order to understand population-level variability in hearing, and to make techniques and hardware available that can be applied to stranded marine mammals so that the hearing of untested species and those unlikely to ever be held by humans can be assessed.
The Evoked Response Study Tool (EVREST) was developed by Dr. James Finneran of the MMP to address these objectives (Finneran, 2009). The EVREST system is a fully configurable, portable AEP system that permits real-time data collection and analysis via objective signal detection techniques. Although methods are drawn from those used in human audiology, they have necessarily been adapted to the unique characteristics of the auditory systems of species adapted to underwater hearing. For example, the hearing range of many marine mammals exceeds that of humans; a typical hearing range for a dolphin is from <1 kHz to 160 kHz, which is broader than the human hearing range by several octaves. Adaptation for this situation requires consideration of appropriate sound projection hardware that can accommodate the full range of dolphin hearing. The cetaceans (whales and dolphins) also have derived auditory systems and pathways that require methodological approaches that differ from those used in human evoked potential audiometry. They lack an external pinnae and the primary pathway for sound reception for frequencies above ~20 kHz is through the lower jaw in odontocetes. It is not the external auditory meatus. This situation requires either testing of animals while in the water with sound projected from an underwater transducer or testing in air/underwater with the sound projector coupled to the lower jaw via a medium that has properties similar to water (e.g., a "jawphone" [Figure 1]).
On the other end of the spectrum, certain seals, such as the northern elephant seal ( Mirounga angustirostris) have proven difficult to test with airborne acoustic stimuli (Houser, Crocker, & Finneran, 2008; Houser, Crocker, Kastak, Mulsow, & Finneran, 2007), a problem that exists because they likely rely on bone conduction for sound transmission to the inner ear while diving and have reduced hearing sensitivity while on land (Figure 2).
Numerous advances in the use of AEP methods for testing marine mammal hearing have been made over the past decade. The AEP approaches are enabling the acquisition of basic audiometric information from marine mammals at rates that exceed by orders of magnitude what can be achieved by traditional behavioral approaches. These advances have made critical contributions to both marine mammal health care and environmental stewardship, some of which are discussed below.
The EVREST system was developed and benchmarked at the MMP in San Diego, California. The system is now used for the routine testing of both bottlenose dolphin (Tursiops truncatus) and California sea lion (Zalophus californianus) hearing. The information is used to screen animals for adequate hearing to support their mission, which includes the detection of underwater mines, the detection and interdiction of swimmers wishing to cause harm to Navy assets, or the retrieval of valuable equipment lost at sea. In addition, repetitive testing of MMP dolphins permits the onset of presbycusis to be monitored. Presbycusis typically occurs in dolphins older than 20 years of age. Since hearing loss affects performance at certain tasks, the detection and quantification of presbycusis can be used by animal management personnel in determining appropriate tasking.
The measurement of AEPs has also become a cornerstone of much of the auditory research that occurs within the MMP. For example, AEP methods are used to monitor the onset of temporary threshold shifts resulting from exposure to anthropogenic sources. They are also used to study dolphin echolocation, specifically the generation of evoked responses produced by a dolphin's own outgoing echolocation clicks and that produced by echoes returning from ensonified objects. Recent research projects include the exploration for AEP correlates of loudness, which have the potential to be used in the development of equal loudness contours and auditory weighting functions.
Since the development of the EVREST system, electrophysiological hearing tests have been performed on more than 150 bottlenose dolphins, compared to the handful of dolphins for which behavioral audiograms were reported from 1960 to 2000. The increase in the sample size of bottlenose dolphin hearing sensitivity curves has permitted demographic relationships to variation in hearing capabilities to be explored. We now know that dolphins tend to lose their upper range of hearing in the mid-20s and that males tend to lose their high frequency hearing at a slightly younger age than females.
Not all of the work on population level audiometry has been performed by the NMMF or MMP. The NMMF and MMP have collaborative efforts with stranding networks. As part of the collaboration, the stranding networks are provided with EVREST systems and training such that they can perform hearing assessments on stranded animals. Researchers involved in developing the methods and hardware for evoked potential audiometry in marine mammals are not the first responders to a stranding event. They often live and work at distances far from the stranding locations, and they are incapable of traveling to the stranding site in sufficient time to perform audiometric testing. The arrangement with the stranding networks is therefore advantageous for acquiring audiometric data in stranded dolphins and porpoises. Currently, five EVREST systems with a simplified user interface for testing dolphins and porpoises are deployed to various stranding networks within the coastal regions of the continental United States. In some regions, stranding networks have been able to acquire decent sample sizes from stranded species for which there was previously little to no audiometric information. For example, stranding responders from the International Fund for Animal Welfare have performed audiometric assessments on nearly 20 common dolphins (Delphinus delphis) over the past 3 years. The growth in marine mammal audiometric information is anticipated to increase dramatically over the next decade as a result of the partnerships with the national stranding networks.
The partnership with stranding networks is not only beneficial to collecting audiometric information from marine mammals for basic scientific purposes, but it is also useful to improving the care and disposition of animals that are in the process of rehabilitation. It has become increasingly common in recent years for AEP methods to be used to test the hearing of a stranded odontocete prior to a determination of whether it can be released back into the wild. Hearing is critical to the ability of dolphins and porpoises to forage and navigate, and a dolphin with a severe hearing deficit will likely fare poorly following release. Indeed, a number of stranded dolphins now kept at marine mammal facilities had severe hearing deficits, which may have contributed to the animals' stranding in the first place. Those stranded dolphins and porpoises that are found with severe hearing deficits, but which are otherwise healthy, can safely be placed under human care at one of the many marine mammal facilities around the country. Furthermore, hearing assessments conducted shortly after stranding and prior to release can be used to monitor for the potential impact of the rehabilitation. Specifically, ototoxic drugs may occasionally need to be given to save an animal's life (e.g., aminoglycoside antibiotics). Assessments prior to and following treatment can thus be compared to determine if any detriment to animal hearing has resulted from treatment with ototoxic compounds. Additionally, the performance of these types of assessments will increase our understanding of the susceptibility of marine mammals to these compounds.
Another benefit of the EVREST system is the ability to respond to rare stranding events. These events provide the opportunity to obtain audiometric information from species that have not been tested and which are unlikely to be maintained by humans at a marine mammal facility. Within the past several years, audiograms have been obtained from two species of beaked whale. These cryptic species are shy and difficult to spot at sea, and we know relatively little about them. However, they have been demonstrated to strand in response to certain naval sonar activities, which fact has elevated the need to know more about their hearing sensitivity. Other species on which AEP testing has been performed include a sperm whale (Physeter macrocephalus) calf and a gray whale (Eschrichtius robustus) calf. The calves are members of species that grow to immense size. It will be difficult to apply AEP methods to adults of these species because of an unfavorable brain-to-body mass ratio, thick skulls, and deep blubber layers between the sites of AEP production and recording. However, there may be potential in the testing of stranded calves, which are much smaller in size. The testing of the large whales, particularly the baleen whales, is a "holy grail" of those interested in marine mammal hearing and for environmentalists concerned with ocean noise. Simply put, there are no direct measures of the hearing range or hearing sensitivity of these animals, yet the frequencies at which ocean noise has increased the most are arguably within the frequency range that baleen whales most utilize. Speculation about their hearing abilities is based upon anatomy of their auditory system, frequencies at which vocalizations occur, and observations of behavioral reactions to sound. These types of approaches leave much to be desired, because they are not direct measures of hearing and may never accurately allow prediction of the frequency range of hearing and hearing sensitivity. Because of their immense size, it is unlikely that any baleen whale will ever be held in captivity for a sufficient duration to attempt behavioral audiometry. The use of electrophysiological approaches may be insufficient, but they may also provide the only opportunity for assessing hearing in these giants of the sea.
The use of AEPs to test hearing in marine mammals has dramatically increased over the past decade. This has occurred because of technological advancements in hardware, a reduction in hardware footprints enabling the portability of testing systems, and the culmination of decades of research leading to modern methods of testing marine mammal hearing via evoked potential audiometry. In addition, partnerships with stranding organizations are facilitating the acquisition of data from wild species. The collaboration permits testing of rare species and is increasing the sample size for more common species, allowing us to understand for the first time population-level variability in hearing. The information obtained from the collective efforts of marine mammal audiologists and stranding networks will ultimately contribute to better care of captive, rehabilitating, and stranded marine mammals and improve environmental stewardship by filling knowledge gaps about marine mammal sensitivities to sound.
For more information about the National Marine Mammal Foundation and its various research programs, please visit their website.
Dorian S. Houser is a biologist with research interests in marine mammal bioacoustics and physiology. He serves as the director of conservation and biological research for the National Marine Mammal Foundation, a non-profit that conducts research relevant to marine mammal conservation and the crossover between marine mammal and human medicine. He is a member of the Acoustical Society of America Committee on Standards and is currently the vice chair of the Accredited Standards Committee S3/SC1 (Animal Bioacoustics). Contact him at firstname.lastname@example.org.
Finneran, J. J. (2009). Evoked Response Study Tool: A portable, rugged system for single and multiple auditory evoked potential measurements. Journal of the Acoustical Society of America, 126(1), 491–500. doi: 10.1121/1.3148214.
Houser, D. S., Crocker, D. E., & Finneran, J. J. (2008). Click-evoked potentials in a large marine mammal, the adult male northern elephant seal (Mirounga angustirostris). Journal of the Acoustical Society of America, 124(1), 44–47.
Houser, D. S., Crocker, D. E., Kastak, C., Mulsow, J., & Finneran, J. J. (2007). Auditory evoked potentials in northern elephant seals Mirounga angustirostris). Aquatic Mammals, 33(1), 110–121.