USF Libraries
USF Digital Collections

Manatee sound localization

MISSING IMAGE

Material Information

Title:
Manatee sound localization performance abilities, interaural level cues, and usage of auditory evoked potential techniques to determine sound conduction pathways
Physical Description:
Book
Language:
English
Creator:
Colbert, Debborah
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Trichechus
Audition
AEP
Head related transfer function
Binaural hearing
Dissertations, Academic -- Psychology -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Three experiments investigated the ability and means by which Florida manatees determine sound source directionality. An eight-choice discrimination paradigm determined the sound localization abilities of two manatees within a 360° array of speakers. Five conditions were tested including a 3,000 and 200 ms, 95 dB, 0.2-24 kHz signal, a 3,000 ms, 80 dB, 18-24 kHz signal, a 3000 ms, 110 dB, 0.2-1.5 kHz signal and a 200 ms, 101 dB, 4 kHz tonal signal. A sixth condition attenuated the level of the 3,000 ms, 95 dB, 0.2-24 kHz signal in 3 dB increments until accuracy reached 75%. Subjects performed above the 12.5% chance level for all broadband frequencies and were able to localize over a large level range. Errors were typically located to either side of the signal source location when presented in the front 180° but were more dispersed when presented from the 135°, 180° and 225° locations.Front-to-back confusions were few and accuracy was greater when signals originated from the front 180°. Head/body related transfer functions determined how different frequencies were filtered by the manatees' head/torso to create frequency-specific interaural level differences (ILDs). Hydrophones were suspended next to each manatee ear and Fast Fourier transform (FFT) ratios compared received signals with and without the subject's presence. ILD magnitudes were derived for all frequencies, as well as specific 0.2-1.5, 0.2-5, and 18-30 kHz bands of frequencies. ILDs were found for all frequencies as a function of source location, although they were largest with frequencies above 18 kHz and when signals originated at 90° and 270°. Larger ILDs were found when the signals originated behind the subjects as compared to in front of them. Auditory evoked potential (AEP) techniques were used to map manatee sound conduction pathways in-water and in-air using 15 and 24 kHz carriers.All subjects produced AEPs at each position the transducer was placed, however specific sound conduction pathway(s) were not identified. AEP amplitudes were usually greater with the 24 kHz carrier, however patterns between carriers at identical body positions were highly variable between subjects.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2008.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Debborah Colbert.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 206 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002000467
oclc - 318988350
usfldc doi - E14-SFE0002489
usfldc handle - e14.2489
System ID:
SFS0026806:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002000467
003 fts
005 20090422143913.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 090422s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002489
035
(OCoLC)318988350
040
FHM
c FHM
049
FHMM
090
BF121 (Online)
1 100
Colbert, Debborah.
0 245
Manatee sound localization :
b performance abilities, interaural level cues, and usage of auditory evoked potential techniques to determine sound conduction pathways
h [electronic resource] /
by Debborah Colbert.
260
[Tampa, Fla] :
University of South Florida,
2008.
500
Title from PDF of title page.
Document formatted into pages; contains 206 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2008.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
520
ABSTRACT: Three experiments investigated the ability and means by which Florida manatees determine sound source directionality. An eight-choice discrimination paradigm determined the sound localization abilities of two manatees within a 360¨ array of speakers. Five conditions were tested including a 3,000 and 200 ms, 95 dB, 0.2-24 kHz signal, a 3,000 ms, 80 dB, 18-24 kHz signal, a 3000 ms, 110 dB, 0.2-1.5 kHz signal and a 200 ms, 101 dB, 4 kHz tonal signal. A sixth condition attenuated the level of the 3,000 ms, 95 dB, 0.2-24 kHz signal in 3 dB increments until accuracy reached 75%. Subjects performed above the 12.5% chance level for all broadband frequencies and were able to localize over a large level range. Errors were typically located to either side of the signal source location when presented in the front 180¨ but were more dispersed when presented from the 135¨, 180¨ and 225¨ locations.Front-to-back confusions were few and accuracy was greater when signals originated from the front 180¨. Head/body related transfer functions determined how different frequencies were filtered by the manatees' head/torso to create frequency-specific interaural level differences (ILDs). Hydrophones were suspended next to each manatee ear and Fast Fourier transform (FFT) ratios compared received signals with and without the subject's presence. ILD magnitudes were derived for all frequencies, as well as specific 0.2-1.5, 0.2-5, and 18-30 kHz bands of frequencies. ILDs were found for all frequencies as a function of source location, although they were largest with frequencies above 18 kHz and when signals originated at 90¨ and 270¨. Larger ILDs were found when the signals originated behind the subjects as compared to in front of them. Auditory evoked potential (AEP) techniques were used to map manatee sound conduction pathways in-water and in-air using 15 and 24 kHz carriers.All subjects produced AEPs at each position the transducer was placed, however specific sound conduction pathway(s) were not identified. AEP amplitudes were usually greater with the 24 kHz carrier, however patterns between carriers at identical body positions were highly variable between subjects.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Co-advisor: Toru Shimizu, Ph.D.
Co-advisor: David Mann, Ph.D.
653
Trichechus
Audition
AEP
Head related transfer function
Binaural hearing
690
Dissertations, Academic
z USF
x Psychology
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2489



PAGE 1

Manatee Sound Localization: Performance Abilities, Interaural Level Cues and Usage of Auditory Evoked Potential Techniques to Determine Sound Conduction Pathways by Debborah Colbert A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of Arts and Sciences University of South Florida Co-Major Professor: Toru Shimizu, Ph.D. Co-Major Professor: David Mann, Ph.D. Steven Stark, Ph.D. Theresa Chisolm, Ph.D. Gordon B. Bauer, Ph.D. Date of Approval: April 15, 2008 Keywords: Trichechus, audition, AEP, head re lated transfer function, binaural hearing Copyright 2008, Debborah E. Colbert

PAGE 2

Dedication This dissertation is dedicated in loving me mory of my father-in-law, Lawrence William Colbert, Sr. In every heart he touched, his love lives on. In every life he changed, he continues to inspire. In every thought he shar ed, his voice still echoes. I am privileged to have been a part of his life.

PAGE 3

Acknowledgments This dissertation was made possible thr ough the assistance of numerous people. The faculty serving on my committee has been instrumental in the development and completion of the investigations included, and I would like to extend my gratitude to my major professors, Dr. Toru Shimizu and Dr. David Mann, as well as Dr. Stephen Stark, Dr. Theresa Chisolm and Dr. Gordon Baue r for their valuable assistance. The completion of the research included in this dissertation would not have been possible without the assistance of the staff at the two facilities data were acquired. Manatee care staff, Joseph Gaspard, Kimberly Dziuk and Adrienne Cardwell, provided vital assistance in the collec tion of all data with Hugh and Buffett at Mote Marine Laboratory and Aquarium. Mote staff member Jay Sp rinkle provided invaluable assistance in creating the localization polar plots. I would also like to thank Mote volunteer trainer Jann Warfield; Mote inte rns Emily Copeland, Nicki Shumway, Taryn Roberts, Anne Johnson and LaTashia Read, a nd New College students that assisted with the manatee training at Mote Anne Schm ieg, Amanda Vennare, Ammanda Stansbery, Kara Tyler, Marc Silpa, Christina Gambacorta, and Beverly Fortner for their assistance. Animal care staff, Dr. Andy Stamper, Patric k Berry, Kim Odell, David Feuerbach, Leslie Larsen, Cathy Goonen, Wendi Fellner and Barb Losch provided the crucial assistance needed to collect data with Mo and Bock at Walt Disney Worlds The Living Seas at EPCOT. My deepest gratitude is extended to all of you for making these investigations possible! I would like to acknowledge the United States Fish and Wildlife Service for granting the permits needed to conduct these investigations and express my gratitude to

PAGE 4

those that funded these studies includi ng the Avoidance Technology Grant from the Florida Wildlife Research Institute, the University of Florida, and the Thurell Family. Finally, I have to acknowledge that the completion of this dissertation would not have been possible without the love and suppor t of my incredible family. My husband Larry, daughters Katie, Alyssa and Lauren, and mother-inlaw Mary, have all provided unending support throughout the entire process. Each day I continue to find that I am humbled by their love and strength and consid er myself incredibly fortunate to be so blessed.

PAGE 5

i Table of Contents List of Tables iii List of Figures iv ABSTRACT viii Chapter One: The Importance of Understa nding the Auditory Sensory System of the Florida Manatee, Trichechus manatus latirostris : An Introduction 1 References Cited 9 Chapter Two: Eight-Choice Sound Loca lization Abilities of Two Florida Manatees, Trichechus manatus latirostris 13 Abstract 13 Introduction 15 Hypotheses 27 Materials and Methods 29 Subjects 29 Subject Training 30 Experimental Design 35 Acoustic Stimuli 37 Signal Generation & Programming 39 Data Recording 41 Experimental Controls 42 Results 45 Discussion 55 References Cited 64 Chapter Three: Head/Body Related Transf er Functions of the Florida Manatee, Trichechus manatus latirostris 73 Abstract 73 Introduction 75 Hypotheses 87 Materials and Methods 88 Subjects 88 Experimental Design 89 Signal Generation & Programming 92 Results 94 Discussion 106 References Cited 113 Chapter Four: Potential Sound Conducti on Pathways for the Florida Manatee, Trichechus manatus latirostris 123

PAGE 6

ii Abstract 123 Introduction 126 Hypotheses 138 Materials and Methods 140 Subjects 140 Experimental Design 142 Signal Generation and Programming 145 Results 147 Discussion 163 References Cited 171 Chapter Five: The Importance of Understa nding the Auditory Sensory System of the Florida Manatee, Trichechus mana tus latirostris: Concluding Remarks 176 References Cited 191 Appendices 195 Appendix A: RPvds language used to generate signals used in the manatee sound localization experiment. 196 Appendix B: Computer protocols used for setting up the manatee sound localization and head/body related transfer function experimental conditions. 199 Appendix C: Data recording protocols used to document each sound localization session. 202 Appendix D: MATLAB program used to determine and chart the manatee head/body related transfer functions. 204 About the Author End Page

PAGE 7

iii List of Tables Table 1.1. Number of manatee deaths and their causes from 2002 through 2007. 2 Table 2.1. Overall performance within and between location, duration and frequency conditions of Gerstein s (1999) four-choice localization experiment. 22 Table 2.2. Overall performance within and between the duration and frequency conditions of Colberts (2005) f our-choice localization experiment. 24 Table 2.3. Frequency, duration and level conditions of the ei ght-choice manatee localization experiment. 38 Table 2.4. Results for the conditions of the eight-choice loca lization experiment with chance level at 12.5%. 47 Table 2.5. Average percents correct by front, back and side regions. 54 Table 2.6. Number of confusions made between contralateral pairs for each subject. 60 Table 3.1. Level differences (in dB) be tween subjects of averaged animal absent minus animal present si gnals for left and right hydrophones with 0.2-1.5, 0.2-5 and 18-30 kHz bands of frequencies. 105 Table 4.1. AEP amplitudes (nV) obtained from the areas surrounding the external auditory meatus and zygomatic process for all subjects. 167 Table 4.2. AEP amplitudes (nV) obtained from points along the vertebral column and lateral ribs 169

PAGE 8

iv List of Figures Figure 2.1. Interaural dist ances vs. maximum frequenc y perceived at 60 dB SPL 20 Figure 2.2. Testing configuration for Co lberts (2005) four-choice localization experiment. 23 Figure 2.3. Percent correct and distri bution of sound localization errors by frequency collapsed across durati on in Colberts (2005) 4-choice sound localization study. 25 Figure 2.4. Diagram of the 265,000 L manat ee exhibit composed of a Medical Pool, Shelf Area, and Exhibit Area. 30 Figure 2.5. Training setup comparison for the Colberts four-choice and eightchoice sound localization experiments 31 Figure 2.6. Power spectra (top) and spectrograms (bottom) of the secondary reinforcement signals. 32 Figure 2.7. Stationing apparatus used in the eight-choice sound localization study 34 Figure 2.8. Testing setup for the eigh t-choice sound localization experiment. 35 Figure 2.10. Electronic button box used to run the sessions and automatically download each trial into a digital excel file. 41 Figure 2.11. Data entry screen used to en ter session information into the Access database. 42 Figure 2.12. Subject accuracy before and after speaker normalization calibration. 46 Figure 2.13. Selection distribution with the 0.2-24 kHz, 3,000 ms, 95 dB re 1 Pa test signal. 48 Figure 2.14. Selection distribution with the 0.2-24 kHz, 200 ms, 95 dB re 1 Pa test signal. 50 Figure 2.15. Selection distribution with the 18-24 kHz, 3,000 ms, 80 dB re 1 Pa test signal. 51

PAGE 9

v Figure 2.16. Selection distribution with the 0.2-1.5 kHz, 3,000 ms, 110 dB re 1 Pa test signal. 52 Figure 2.17. Selection distribution with the 4 kHz, 200 ms, 101 dB re 1 Pa test signal. 53 Figure 3.1. Interaural time (ITD), phase (IPD), and level (ILD ) cues used for sound localization. 78 Figure 3.2. Azimuth, elevati on and medial planes used to integrate the vertical, horizontal and distance dimensions with a sounds temporal, phase and level cues. 79 Figure 3.3 Cone of confusion caused from sounds originating in different locations. 82 Figure 3.4. Interpretation of how signals presented from the 135o and 225o locations reflect off the manatees elliptically shaped body. 87 Figure 3.5. Testing setup for the ma natee body related transfer function experiment. 89 Figure 3.6. Stationing apparatus used to measure manatee bo dy related transfer functions 90 Figure 3.8. Comparisons of left vs. right received signals as a function of signal source location and the presence (red line) or absence (blue line) with Hugh. 95 Figure 3.9. Comparisons of left vs. right received signals as a function of signal source location and the presence (red line) or absence (blue line) with Buffett. 96 Figure 3.10. Left (270o) vs. right (90o) head/body related tr ansfer functions for Buffett. 97 Figure 3.11. Head/body related tr ansfer functions for Hugh 98 Figure 3.12. Head/body related tran sfer functions for Buffett. 99 Figure 3.13. Interaural level difference ma gnitudes between the left and right hydrophones 101 Figure 3.14. Interaural level difference ma gnitudes between the left and right hydrophones 102

PAGE 10

vi Figure 3.15. Interaural level difference ma gnitudes between the left and right hydrophones 104 Figure 3.16. Simulated shadow effects created from the manatee head and body. 108 Figure 4.1. Diagrammatic illustration of manatee auditory anatomy based on multiple cross-sections th rough the transverse plane. 128 Figure 4.2. Manatee ossicl es from the right ear. 130 Figure 4.3. Right lateral view of a th ree dimensional reconstruction of a CT scanned manatee head 132 Figure 4.4. Potential sound pathways where a sound wave will experience the least amount of reflection. 133 Figure 4.5. Schematic illustrations of the manatee diaphragm and lung. 136 Figure 4.6. Testing setup for voluntary auditory evoked potential measurements used to map sound conduction pathways with subjects at Mote Marine Laboratory & Aquarium. 141 Figure 4.6. Testing setup for restrained auditory evoked potential measurements used to map sound conduction pathways with subjects at 142 Figure 4.7. Voluntary AEP measurements with Hugh. 143 Figure 4.8. Restrained AEP measurements with Bock. 144 Figure 4.9. A typical au ditory evoked potential found at the 600 Hz AM frequency using the 15 kHz carrier. 147 Figure 4.10. A typical auditory evoked potential found at the 600 Hz AM frequency using the 24 kHz carrier. 148 Figure 4.11. In-water auditory evoked poten tial response measurements for Hugh. 149 Figure 4.12. Estimated in-water sound pr essure level measurements for Hugh. 150 Figure 4.13. In-water auditory evoked pot ential response measurements for Buffett. 151 Figure 4.14. Estimated in-water sound pre ssure level measurements for Buffett. 151 Figure 4.15. In-water auditory evoke d potential response measurement comparison of common positions for Hugh and Buffett. 152

PAGE 11

vii Figure 4.16. In-air auditory evoked poten tial response measurements for Bock. 153 Figure 4.17. Estimated in-air sound pre ssure level measurements for Bock. 154 Figure 4.18. In-air auditory evoked poten tial response measurements for Hugh. 155 Figure 4.19. Estimated in-air sound pre ssure level measurements for Hugh. 156 Figure 4.20. In-air auditory evoked poten tial response measurements for Mo. 158 Figure 4.21. Estimated in-air sound pre ssure level measurements for Mo. 159 Figure 4.22. In-air auditory evoked poten tial response measurement comparison of common positions for Hugh, Mo, and Bock. 161 Figure 4.23. In-water vs.in-air auditory evoked potential response measurement comparison of common positions for Hugh, Buffett, Mo and Bock. 162 Figure 5.1. Contralateral comparis on between ILD magnitudes and sound localization selection distributions at 0o and 180o. 183 Figure 5.2. Contralateral comparis on between ILD magnitudes and sound localization selection distributions at 45o and 225o. 184 Figure 5.3. Contralateral comparis on between ILD magnitudes and sound localization selection distributions at 90o and 270o. 185 Figure 5.4. Contralateral comparis on between ILD magnitudes and sound localization selection distributions at 315o and 135o. 186 Figure A-1. RPvds language used to gene rate each subject s call to station. 196 Figure A-2. RPvds language used to generate the initiation of each trial. 197 Figure A-3. RPvds language used to gene rate each subjects secondary bridge when correct. 197 Figure A-4. RPvds language used to docum ent when incorrect selections were made. 198 Figure B-1. The graphical user interface scr een (programmed in Visual C) used to setup the experimental conditions and automatically download the results into an Excel file during the testing sessions. 201 Figure C-1. The tank-side data-recordi ng sheet used to document each session. 203

PAGE 12

viii Manatee Sound Localization: Performance Abilities, Interaural Level Cues and Usage of Auditory Evoked Potential Techniques to Determine Sound Conduction Pathways Debborah Colbert ABSTRACT Three experiments investigated the abili ty and means by which Florida manatees determine sound source direc tionality. An eight-choice discrimination paradigm determined the sound localization abi lities of two manatees within a 360o array of speakers. Five conditions were tested including a 3,000 and 200 ms, 95 dB, 0.2-24 kHz signal, a 3,000 ms, 80 dB, 18-24 kHz signal, a 3000 ms, 110 dB, 0.2-1.5 kHz signal and a 200 ms, 101 dB, 4 kHz tonal signal. A sixth condition attenuated the level of the 3,000 ms, 95 dB, 0.2-24 kHz signal in 3 dB increments until accuracy reached 75%. Subjects performed above the 12.5% chance level for al l broadband frequencies and were able to localize over a large level range. Errors were typically located to eith er side of the signal source location when presented in the front 180o but were more dispersed when presented from the 135o, 180o and 225o locations. Front-to-back confusions were few and accuracy was greater when signals or iginated from the front 180o. Head/body related transfer functions dete rmined how different frequencies were filtered by the manatees head/torso to create frequency-specific interaural level differences (ILDs). Hydrophones were suspended next to each manatee ear and Fast Fourier transform (FFT) ratios compared recei ved signals with and without the subjects

PAGE 13

ix presence. ILD magnitudes were derived for all frequencies, as well as specific 0.2-1.5, 0.2-5, and 18-30 kHz bands of frequencies. ILDs were found for all frequencies as a function of source location, although they we re largest with frequencies above 18 kHz and when signals originated at 90o and 270o. Larger ILDs were found when the signals originated behind the subjects as comp ared to in front of them. Auditory evoked potential (AEP) techni ques were used to map manatee sound conduction pathways in-water and in-air usi ng 15 and 24 kHz carriers. All subjects produced AEPs at each position the trans ducer was placed, however specific sound conduction pathway(s) were not identified. AEP amplitudes were usually greater with the 24 kHz carrier, however patterns between carriers at identical body positions were highly variable between subjects.

PAGE 14

Chapter One: The Importance of Understa nding the Auditory Sensory System of the Florida Manatee, Trichechus manatus latirostris : An Introduction The Florida manatee ( Trichechus manatus latirostris ) is a sub-species of the West Indian manatee ( Trichechus manatus) that is typically found in the coastal waterways surrounding the peninsula of Florida, but can ra nge as far north as Massachusetts and as far west as Louisiana. In the summer months it lives in turbid saltwater habitats, grazing primarily on sea grass (Reynolds & Odell, 1991). In colder months, it migrates to freshwater springs or power plant discharge sites where water remains at a warmer temperature, feeding primarily on water hyacinth, hydrilla, and other freshwater vegetation (Reynolds & Wilcox, 1986). It is considered a semi-social species, often grazing or traveling alone, although females with calves will often congregate together and males will frequently mass around estrous females for mating purposes (Reynolds, 1979). Although manatee ecology and population biology field studies validate that these behavior patterns are typical (Hartman, 1979; U.S. Fish and Wildlife Service, 2001), the means by which they navigate and locate one another within their vast habitat remains unclear. The Florida manatee is an endangered species, currently protected by both the Marine Mammal Protection Ac t (1972) and the Endangere d Species Act (1973). The February 2007 synoptic survey estimated the Florida manatee population to be approximately 2,817 animals (Florida Fish and Wildlife Research Institute, 2007a). It is 1

PAGE 15

known to be threatened by naturally occurring ev ents such as cold stress and red tide and by human-influenced events such as boat strikes, canal lock compression, and habitat degradation (Odell & Reynolds, 1979). Over 1,027,000 boats were registered in the state of Florida in 2007 (Florida Department of Highway Safety and Motor Vehicles, 2007) and many manatees are hit by vessels numerous times throughout their lives as evidenced by a multitude of scar patterns on their bodies (Beck & Reid, 1995). The frequency of deaths caused specifically by watercraft re mains relatively stable ranging between 1931% of the annual mortalities (Florida Fish and Wildlife Re search Institute, 2007b; Table 1.1). Because the annual number of undete rmined causes of death and unrecovered carcasses is high, the annual pe rcentage of deaths caused by watercraft is likely underestimated. Table 1.1. Number of manatee deaths and their causes from 2002 through 2007. Year Watercraft Gates/ Locks Other Human Perinatal Cold Stress Natural Undetermined Unrecovered Total 2002 95 5 9 53 17 59 65 2 305 2003 73 3 7 71 47 102 67 10 380 2004 69 3 4 72 50 24 51 3 276 2005 80 6 8 89 31 88 90 4 396 2006 92 3 6 70 22 81 116 27 417 2007 73 2 5 59 18 81 67 12 317 Since the Florida manatee lives in a habitat where boats are found in high numbers and conspecifics are often out of visu al range, it is important to gain a detailed understanding of how the manat ee perceives its environment Although no information has been published regarding the manatees gustatory and olfactory sensory systems, 2

PAGE 16

several anatomical and behavioral studies ha ve provided considerab le insight into the manatees visual, tactile, and auditory sensory processes. Anatomical investigations of the small 18 mm diameter manatee eye reveal that it possesses two types of cones (Cohen et al ., 1982; Ahnelt & Kolb, 2000; Ahnelt & Bauer, 2000), has relatively few retinal ganglion ce lls, lacks an accommodation mechanism, and has limited resolution with a minimum angle of 20 minutes of visu al arc (Walls, 1963; Piggins et al., 1983; West et al., 1991; Mass et al., 1997). Behavioral investigations of the manatees visual sensory system using discrimination testing paradigms found that subjects were able to distingui sh blue and green from a series of comparably bright grays (Griebel & Schmid, 1996) and differentiate brightness with a Weber fraction of 0.35 (Griebel & Schmid, 1997). A visual acuity st udy using gratings of various widths found that one subject possessed a minimum angle of 21 minutes of visual arc, while the second subjects was over a degree (Bauer et al., 2003). Results from this behavioral study in confluence with ganglion cell density anatomi cal data suggest that the 21 minutes of visual arc is probably ty pical for manatees. Anatomical investigations of the manate es facial vibrissae show that each is composed of a dense connec tive tissue capsule with a prominent blood sinus complex and substantial innervation. Six fields of peri oral bristles have been identified (Reep et al., 2001) and those located on the upper lip are used in a prehensile manner during feeding (Marshall et al., 1998 & 2003). Each po stcranial body vibriss ae also contains a blood sinus and is innervated by 20-50 axons (Reep et al., 2002). Behavioral investigations of the manatees facial tactil e sensory system using discrimination testing paradigms found that an Antillean manatee possessed good sensitivity with a Weber 3

PAGE 17

fraction of 0.14 (Bachteler & Dehnhardt, 1999) and two Florida manatees had excellent sensitivity with a Weber fraction of 0.025 fo r one subject and 0.075 fo r the other (Bauer et al., 2005), sensitivity comparable to that of a human index finge r (Weber fraction of 0.028) (Gaydos, 1958). Anatomical investigations of the manate es ear demonstrate that the external pinna flange is absent and that the external auditory meatus is of minute size, occluded with cellular debris that reach es a blind end separated from the tympanic membrane, and is an unlikely channel for s ound transmission (Ketten et al., 1992; Chapla et al., 2007). Behavioral investigations of the manatees auditory sensory system have been conducted using discrimination and auditory evoked poten tial testing techniques. Gerstein et al. (1999) obtained a behavioral audiogram fo r two manatees and found that hearing thresholds ranged from 0.5 kHz for one s ubject and 0.4 kHz for the other. The frequency range of best hearing was betw een 10 kHz and maximum sensitivity was ~50 dB re: 1 Pa at 16 and 18 kHz, decreasing by ~20 dB re: 1 Pa per octave from 0.8 to 0.4 kHz and 40 dB re: 1 Pa per octave above 26 kHz. Auditory evoked potential investigations found the frequency range of detection reached up to 35 kHz when tested in air (Bullock et al., 1980; 1982; Popov & Supin, 1990) and 60 kHz when tested in water (Klishen et al., 1990). More recently, Mann et al. (2005) found an upper limit of detection at 40 kHz when tested in water. The information gained from sensory invest igations with animals in a controlled setting offer indications about how their se nsory systems function in natural settings. These results suggest that manatee vision is built for sensitivity in dim light conditions with the ability to distinguish brightness diffe rences and differentiate blues from greens, 4

PAGE 18

but that acuity is poor and not useful for fine details. Tact ile sensitivity appears to be superior, but like vision, is probably designed to function with nearby tasks. Audition is excellent and spans a wide range of frequencies seemingly fulfilling a crucial role for functioning in both nearby and distant scenarios. This capability likely facilitates the capacity for sound localization which would be of great importance for tasks such as navigation, finding conspecifics and boat avoidance. Therefor e, the localiza tion abilities of the manatee warrant further investigation. Initial estimations of manatee sound local ization abilities were determined by comparing manatee interaural time delays (the distance sound travels from one ear to the other divided by the speed of sound) to thos e of other species. Heffner and Heffner (1992) generated a regression equation that de scribed the relationship between interaural time delays and the upper frequency hearing limits for a variety of species. Animals with narrower heads had smaller interaural time de lays and typically needed higher frequency sensitivity to be able to localize sounds. Ketten et al. (1992) calculated the manatee intermeatal distance as 278 mm with a maximum acoustic travel time of 258 sec, and the intercochlear distance as 82 mm w ith a maximum acoustic travel time of 58 sec. When these time delays were plotted on He ffner and Heffners regression line, it appeared that manatees would need a 50 kHz upper frequency limit to be able to localize sound. Given that behavioral inves tigations indicated that the upper limit of manatees hearing likely lies between 40 -60 kHz, below or bordering the 50 kHz interaural upper frequency limit estimates ne eded for localization, it was predicted that manatees may not possess good sound localiz ation abilities (Ke tten et al., 1992). 5

PAGE 19

This calculation however, was not suppor ted by the results of two distinct behavioral sound localization studies. Gerste in (1999) tested one manatees ability to localize in a four-choice (45o, 90o, 270o and 315o) testing paradigm. Stimuli included 0.5, 1.6, 3, 6, or 12 kHz tonal signals, pulsed for e ither 200 or 500 ms, paused for 400 ms, and then repeated. Results indicated that the su bject was capable of localizing all signals but accuracy increased with the higher frequencies and at the 90o angles. Given the subjects poorer performance with the low frequency stimuli, Gerstein suggested that low frequency sounds typical of recreational boat engine noises might be difficult to localize. Colbert (2005) also conducted a f our-choice localization experiment (45o, 90o, 270o and 315o) with two manatees. Stimuli included three broad-band noises of 0.2-20, 6-20, and 0.2 kHz tested at four durations (3,000, 1,000, 500 and 200 ms) and two tonal signals of 4 and 16 kHz tested at 3,000 ms. Results in dicated that the subj ects were able to localize all of the broad-band stimuli at each duration and location, including the lowest frequencies which conflicted with Gersteins predictions. Both subjects also performed above chance levels with the tonal signals but with lower accuracy. As often happens when conducting research, although one question may be answered, many more arise. These localization investigations demonstrated that subjects were able to localize high, medium and low frequency test signals from four speakers located at 45o, 90o, 270o and 315o. The goal of this dissertation is to expand upon these studies and determine the manatees ability to determine sound source directionality within all 360o of the azimuth plane and identify the possible means by which they do so. Chapter Two investigates the manatees ability to localize test signals at 45o angles within the 360o of the azimuth plane. In this study, two male captive-born 6

PAGE 20

manatees at Mote Marine Laboratory in Sa rasota Florida, Hugh and Buffett, were conditioned to participate in an eight-choice localization project. The experimental design of this study expands upon a previous manatee sound localization study (Colbert, 2005) by incorporating a broadband stimulus th at spanned a wider range of frequencies (from 0.2-20 kHz to 0.2-24 kHz), one restricted to higher frequencies (from 6-20 kHz to 18-24 kHz), and one limited to lower frequenc ies (from 0.2-2 kHz to 0.2-1.5 kHz). In addition, a 4 kHz tonal signal was tested at a shorter duration (f rom 3,000 ms to 200 ms) and the level of the 0.2-24 kHz signal was in crementally reduced to investigate the effects of decreased amplitude. Chapter Three considers how different frequencies of a test signal, presented at 45o angles within the 360o of the azimuth plane, are filtered by the manatees head and body to provide interaural level difference cu es that may aid sound localization. The same two male captive-born manatees at Mote Marine Laboratory in Sarasota Florida, Hugh and Buffett, participated in this study. These are the first body related transfer function data collected for any Sirenian species. Chapter Four investigates how auditory evoked potential techniques can be used to evaluate the potential existence sound c onduction pathways, outside of the traditional pinna-to-cochlea pathway, which might be used by the manatee since the external auditory meatus is occluded with cellular debris and is separated from the tympanic membrane (Chapla et al., 2007). Four male ma natees participated in this study including Hugh and Buffett at Mote Marine Laboratory in Sarasota Florida, and Mo and Bock at Walt Disney Worlds The Living Seas at EP COT in Lake Buena Vista, Florida. 7

PAGE 21

8 Chapter Five provides a brief summary of the experiments detailed in Chapters Two, Three, and Four, each of which are fo rmatted for individual journal publication. Concluding remarks tie the three Chapters toge ther and address the questions of how well manatees are able to localize sound sources, ho w interaural intensity cues may facilitate sound localization, and what sound conduction pa thways may be used for hearing.

PAGE 22

References Cited Ahnelt, P.K. & Bauer, G.B. (2000). Unpublished data reported in Ahnelt, P. K. & Kolb, H. (2000). The mammalian photore ceptor mosaic-adaptive design. Progress in Retinal and Eye Research 19, 711-777. Ahnelt, P.K. & Kolb, H. (2000). The mamm alian photoreceptor mosaic-adaptive design. Progress in Retinal and Eye Research 19, 711-777. Bachteler, D. & Dehnhardt, G. (1999). Ac tive touch performance in the Antillean manatee: Evidence for a functional differen tiation of the facial tactile hairs. Zoology, 102, 61-69. Bauer, G.B., Colbert, D.E., Gaspard, J.C. & Littlefield, B. & Fellner, W. (2003). Underwater visual acuity of Florida manatees ( Trichechus manatus latirostris.). International Journal of Comparative Psychology 16, 130-142. Bauer, G.B., Gaspard III, J.C., Colbert, D.E., Leach, J. B. & Reep, R. (2005, March). Tactile Discrimination of Textures by Florida Manatees, Trichechus manatus latirostris. Paper presented at the 12th annua l International Conference on Comparative Cognition, Melbourne, Florida. Beck, C.A. & Reid, J.P. (1995). An automate d photo-identification ca talog for studies of the life history of the Florida manatee. In T. J. O'Shea, B. B. Ackerman, & H. E Percival (Eds.) Population Biology of the Florida Manatee U.S. Department of the Interior, National Biological Servi ce, Information and Technology Report 1, pp. 120-134. Bullock, T.H., Domning, D.P. & Best, R.C. (1980). Evoked potentials demonstrate hearing in a manatee ( Trichechus inunguis ). Journal of Mammalogy 61 (1), 130133. Bullock, T.H., OShea, T.J. & McClune, M.C. (1982). Auditory evoked potentials in the West Indian manatee (Sirenia: Trichechus manatus). Journal of Comparative Physiology, 148, 547-554. Chapla, M., Nowacek, D., Rommel, S. & Sadler, V. (2007). CT scans and 3D reconstructions of Florida manatee ( Trichechus manatus latirostris ) heads and ear bones. Hearing Research 228, 123-135. Cohen, J.L., Tucker, G.S., & Odell, D.K. ( 1982). The photoreceptors of the West Indian 9

PAGE 23

manatee. Journal of Morphology, 173, 197-202. Colbert, D. (2005). Sound Localization Abili ties of Two Florida Manatees, Trichechus manatus latirostris. Unpublished Master s Thesis, University of South Florida, Tampa, FL. Florida Department of Highway Safety and Motor Vehicles (2007). Annual vessel statistics http://www.hsmv.state.fl.us/dmv/vslfacts.html Florida Fish & Wildlife Res earch Institute (2007a). Annual synoptic surveys http://research.myfwc.com/featur es/view_article.asp?id=15246 Florida Fish & Wildlife Res earch Institute (2007b). Annual mortality rates http://research.myfwc.com/featur es/view_article.asp?id=12084 Gaydos, H.F. (1958). Sensitivity in the judgement of size by finger span, The American Journal of Psychology 71, 557-562. Gerstein, E. (1999). Psychoacoustic Eval uations of the West Indian manatee ( Trichechus manatus latirostris ). Unpublished Doctoral Dissertation, Florida Atlantic University, Boca Raton, FL. Gerstein, E., Gerstein, L., Forsythe, S. & Bl ue, J. (1999). The unde rwater audiogram of the West Indian manatee (Trichechus manatus ). Journal of the Acoustical Society of America 105, 3575-3583. Griebel, U. & Schmid, A. (1996) Color vision in the manatee ( Trichechus manatus). Vision Research, 36, 2747-2757. Griebel, U. & Schmid, A. ( 1997). Brightness discrimination ab ility in the West Indian manatee ( Trichechus manatus). Journal of Experimental Biology, 200, 15871592. Hartman, D.S. (1979). Ecology and Behavior of the manatee (Trichechus Manatus) in Florida. Special publication No. 5 (153 pp.) Th e American Society of Marine Mammologists. Heffner, R.S. & Heffner, E.H. (1992). Evoluti on of sound localization in Mammals. In D. Webster, R. Fay & A. Popper (Eds.) The Biology of Hearing New York, Springer-Verlag, pp. 691-715. Ketten, D.R., Odell, D.K. & Domning, D.P. (1992). Structure, function, and adaptation of the manatee ear. In J.A. Thomas, R.A. Kastelein, & A.Y. Supin (Eds.) Marine Mammal Sensory Systems (pp.77-79). NY: Plenum Press. 10

PAGE 24

Klishen, V.O., Diaz, R.P., Popov, V.V. & Supin, A.Y. (1990). Some characteristics of hearing of the Brazilian manatee, Trichechus inunguis. Aquatic Mammals 16 (3), 139-144. Mann, D., Colbert, D.E., Gaspard, J.C. III, Casper, B., Cook, M.L.H., Reep, R.L. & Bauer, G.B. (2005). Temporal resolution of the Florida manatee ( Trichechus manatus latirostris ) auditory system. Journal of Comparative Physiology 191, 903-908. Marshall, C.D., Huth, G.D., Edmonds, V.M. Halin, D.L. & Reep, R.L. (1998). Prehensile use of perioral bristles during feeding and associated behaviors of the Florida manatee ( Trichechus manatus latirostris ). Marine Mammal Science, 14, 274-289. Marshall, C.D., Maeda, H., Iwata, M., Furuta, M., Asano, A., Rosas, F. & Reep, R.L. (2003). Orofacial morphology and feedi ng behavior of the dugong, Amazonian, West African, and Antillean manat ees (Mammalia: Sirenia): Functional morphology of the muscular-vibrissal complex. Journal of Zoology 259, 245260. Mass, A.M., Odell, D.K., Ketten, D.R. & Supin, A.Y. (1997). Ganglion layer topography and retinal resolution of the Caribbean manatee Trichechus manatus latirostris Doklady Biological Sciences, 355, 392-394. Odell, D. K., & Reynolds, J. E. III. (1979). Observations on manatee mortality in South Florida. Journal of Wildlife Management 28 (3), 572-577. Piggins, D.J., Muntz, W.R.A. & Best, R.C. (1983). Physical and mo rphological aspects of the eye of the manatee, Trichechus inunguis Marine Behaviour and Physiology 9, 111-130. Popov, V.V. & Supin, A.Y. (1990). Electrophys iological studies on hearing in some cetaceans and a manatee. In J.A. Thomas, & R. A. Kastelein (Eds.), Sensory Abilities of Cetaceans: Laboratory and Field Evidence NY: Plenum Press. Reep, R.L ., Marshall, C.D., Stoll, M.L., Homer, B.L. & Samuelson, D.A. (2001). Microanatomy of facial vibrissae in the Florida manatee: the basis for specialized sensory function and oripulation. Brain Behavior and Evolution 58, 1-14. Reep, R.L., Marshall, C.D. & Stoll, M.L. (2002) Tactile hairs on the postcranial body in Florida manatees: a mammalian lateral line? Brain Behavior and Evolution 59, 141-154. Reynolds, J.E. III. (1979). The semisocial manatee. Natural History 88 (2), 44-53. Reynolds, J.E. & Odell, D.K. (1991). Manatees and Dugongs (192 pp.). NY: Checkmark Books. 11

PAGE 25

Reynolds, J.E. III, & Wilcox, J.R. (1986). Distribution and abundance of the West Indian manatee, Trichechus manatus around selected Florida power plants following winter cold fronts: 1984-1985. Biological Conservation 38, 103-113. U.S. Fish and Wildlife Service (2001). Florida Manatee Recovery Plan, (Trichechus manatus latirostris), Third Revision. U.S. Fish and Wildlife Service, Atlanta, Georgia. 144 pp. Walls, G.L. (1963). The vertebrate eye and it s adaptive radiation NY: Hafner. West, J.A., Sivak, J.G., Murphy, C.J. & K ovacs (1991). A comparative study of the anatomy of the iris and ciliary body in aquatic mammals. Canadian Journal of Zoology, 69, 2594-2607. 12

PAGE 26

Chapter Two: Eight-Choice Sound Localization Abilities of Two Florida Manatees, Trichechus manatus latirostris Abstract An eight-choice discrimination paradigm was used to determine the sound localization abilities of two Florida manatees (Trichechus manatus latirostris) within a 360o array of speakers positioned 45o apart. Five conditions were tested including a 3,000 ms and 200 ms, 95 dB, 0.2-24 kHz br oadband signal, a 3,000 ms, 80 dB, 18-24 kHz broadband signal that was restricted to frequencies with wavele ngths shorter than a manatees interaural time distances, a 3000 ms, 110 dB, 0.2-1.5 kHz broadband signal that was limited to frequencies with wave lengths longer than their interaural time distances, and a 200 ms, 101 dB, 4 kHz tonal st imulus thats an approximate midpoint of the fundamental frequency range of manatee vocalizations. A sixth condition attenuated the spectrum level of the 3,000 ms, 95 dB, 0.2-24 kHz signal in 3 dB increments until accuracy reached 75%. Both subjects performed well above the 12.5% chance level for all broadband frequencies tested. They also were able to localize over a fairly large sound level range with Hughs accuracy at 48% and Buffetts at 56% when the signal was presented at 80 dB re 1 Pa. Accuracy deteriorated to 14 % for Hugh and 20 % for Buffett when the 4 kHz, 200 ms, 101 dB re 1 Pa signal was tested. Errors were primarily located at the 13

PAGE 27

14 nearest neighbor locations on either side of the signal source location when presented in the front 180o but became more dispersed when signals originated from the 135o, 180o and 225o locations. Very few front to back c onfusions were made and accuracy was greater when test signals or iginated from the front 180o and with the longer 3,000 ms duration of the 0.2-24 kHz signal. Results from this study demonstrate that the subjects could lo calize short duration and low intensity test signals within the fr equency ranges of recrea tional boat engines and conspecifics in all 360o of the azimuth plane at distances of at least 3 meters.

PAGE 28

Introduction The Florida manatee ( Trichechus manatus latirostris ) is a sub-species of the West Indian manatee ( Trichechus manatus) protected by both the Marine Mammal Protection Act (1972) and the Endangered Species Act (1973) It is the only marine mammal that is euryhaline, living in both saltwater and fres hwater habitats depending on the time of year (Reynolds & Wilcox, 1986; Reynolds & Odell, 1991) It has been described as a semisocial species, often grazing or traveling alone, although females with calves will often congregate together and males will frequen tly mass around an estrous female for mating purposes (Reynolds, 1979). It is threatened by naturally occurring events such as cold stress and red tide, as well as by human-influenced events such as canal lock compression, habitat degradation, and boa t strikes (Odell & Reynolds, 1979). The frequency of deaths caused specifically by wa tercraft remains relatively stable ranging between 19-31% of the annual mo rtalities (Florida Fish and Wildlife Research Institute, 2007; Table 1-1). Although field research has provided cruc ial information about the manatees social structure, habitat usage and annual migratory behaviors, the means by which they are able to find one another, determine dire ctionality, and avoid danger in their vast habitat is unclear. Research has not been published regarding the manatees gustatory and olfactory sensory systems, however anatom ical and behavioral studies have gained considerable insight into the manatees visual and tactile sensory processes. The manatee visual sensory system is built for sensitivity in dim light conditions with the ability to 15

PAGE 29

differentiate brightness differences (Grieb el & Schmid, 1997) and blues from greens (Cohen et al., 1982; Griebel & Schmid, 1996; Ahnelt & Kolb, 2000; Ahnelt & Bauer, 2000), but acuity is poor (Walls, 1963; Piggins et al., 1983; West et al., 1991; Mass et al., 1997; Bauer et al., 2003) and not useful for fine details. Tactile sensitivity appears to be excellent (Bachteler & Dehnhardt, 1999; Reep et al., 2002; Bauer et al., 2005), but like vision, is probably designed to function with nearby tasks. It seems likely that the manatees audito ry sensory system plays a crucial role with functioning not only in clos e proximity, but also in more distant scenarios and that the ability to determine conspecific and boat engine sound source dire ctionality would be of great importance. Mana tee vocalizations are charac terized as short harmonic complexes that range from almost pure tones to broad-band noise and have a fundamental frequency that ranges between 2.5 5.9 kHz but can extend to 15 kHz (Nowacek, et al., 2003). The dominant recreational boat engi ne frequency ranges between 0.01 2 kHz, but can reach over 20 kHz with the estimated 1/3-octave source leve ls at 120-160 dB re 1 Pa at 1 m for small motorboats (Gerst ein, 2002; Richardson et al., 1995) and at approximately 9 dB quieter for personal wate rcraft, such as jet-skis (Buckstaff, 2004). Boats traveling at rapid speeds typically pr oduce higher frequency cavitating noise, while those traveling at idle and slow speeds produce lower frequency non-cavitating noise (Ross, 1976; Miksis-Olds, 2006). Sound localization is the a uditory systems ability to process the frequency, level, and phase of a sound and associate it with the spatial locat ion of that sounds source (Yost, 2000). Sound can be localized from the vertical, distance and azimuth (horizontal) planes using inte raural time, phase, and/or leve l difference cues. Interaural 16

PAGE 30

time differences compare the time the sound arri ves at each ear; since the speed of sound is relatively constant and not effected by frequency wavele ngth, frequency variations do not have an effect on the perception of inte raural time differences. Interaural phase differences compare the period of the sound as it arrives at each ear and is affected by frequency wavelength. Intera ural level differences compare the level or amplitude of a sound as it reaches each ear a nd is also affected by wavele ngth with higher frequencies having shorter wavelengths and greater sound shadows (Yost, 2000). The ability to localize sounds is considered a primary source of selective pressure in the evolution of mammalian hearing (Mas terson et al., 1969) and is vital for many species ability to find food a nd conspecifics while avoidi ng predation. Behavioral testing of sound localization ab ilities has typically been i nvestigated by measuring the species minimum audible angle (MAA) (B rown, 1994; Brown & May, 1990). This method determines the smallest detectable angular difference between two sound source locations positioned in front of the subject in the azimuth plane (Mills, 1958). Numerous in-air auditory localization studi es have been conduc ted with terrestrial mammals including humans (Stevens & Newman, 1936; Mills, 1972), monkeys (Don & Star, 1972; Houben & Gourevitch, 1979; Brown et al., 1980), the domestic cat (Casseday & Neff, 1973; Wakeford & Robinson, 1974; Heffner & Heffner, 1988b), red fox (Isley & Gysel, 1975), hedgehog (Masterson et al., 1975), el ephant, horse, Norway rat, pig, gerbil, Northern grasshopper mouse, pocket gopher, goat and cattle (Heffner & Heffner, 1982; 1984; 1985; 1988a; 1988c; 1989; Heffner & Masterson, 1990; Heffner & Heffner, 1992b respectively). Results from these studies suggest that some combination of interaural time, level and phase difference cues are us ed to localize sounds although some species 17

PAGE 31

have reduced or lost the ability to use one or two of them, and one species (pocket gopher) cannot use any of them and is incapable of sound localization. While in-air localization may be difficult or impossible for some terrestrial species, the ability to localize sounds underwater presents additional challenges to marine mammals. The speed of sound in water ( 1500 m/second) travels approximately five times faster than in air (340m/second) (U rick, 1996) requiring marine mammal auditory systems to process interaural time, phase a nd level differences much more rapidly than terrestrial mammals. Although most acoustic en ergy propagates more efficiently in water than light, thermal or electromagnetic energy (Au, 1993), in shallow waters, higher frequencies become more directional, refl ecting off the surface and bottom and hindering sound wave travel efficiency and very low frequencies may not propagate well (Medwin & Clay, 1998). Marine mammals likely utilize underwater acoustic information for reproduction and territorial purposes (Watki ns & Schevill, 1979; Cleator & Stirling, 1990; Bartsh et al., 1992; Hanggi & Schusterm an, 1994; Rogers et al., 1996; Smolker & Pepper, 1996; Van Parijs et al., 1999; Van Pa rijs et al., 2000a; Va n Parijs et al., 2000b; Serrano & Terhune, 2002; Van Parijs et al., 2003 ; Bjrgesaeter et al ., 2004; Hayes et al., 2004), individual identificati on (Caldwell & Caldwell, 1965; Sayigh et al., 1990; Sayigh et al., 1995), prey detecti on (Barrett-Lennard et al., 1996; Tyack & Clark, 2000; Gannon et al., 2005), predator avoidance (Deecke et al., 2002) and navigation (Norris, 1967). Minimum audible angle measurements have also been assessed for some marine mammals including pinnipeds (Gentry, 1967; Anderson, 1970; Moore, 1974; Terhune, 1974; Moore & Au, 1975; Babushina and Poliakov, 2004; Holt et al., 2004) and cetaceans (Renaud & Popper, 1975; Moore & Pawloski, 1993; Moore and Brill, 2001). 18

PAGE 32

More recently, some pinniped sound localization i nvestigations have re quired subjects to identify sound sources relative to different locations surrounding the subjects body. This has been done by presenting sign als in the frontal 180 or co mplete 360 of the horizontal plane surrounding a stationary subject (Kaste lein et al., 2007) or by having the subject swim along a half circle diameter and orie nt towards a sound source when presented (Bodson et al., 2006). All three designs asse ss sound localization ab ilities, however the latter two have enhanced real-life scenario applications by addressing the subjects ability to determine sound source directionality as sounds originate from different angles surrounding the body. Given that the Florida manatees visual and tactile sensory systems are better adapted for use with tasks in close proximity to their bodies, it seems likely that their auditory sensory system has developed to function with both nea r-field and far-field scenarios and that the ability to determine sound source directionality would be of great importance. This area of research however, has not been widely investigated and is relatively new. Heffner and Heffner (1992a) generated a regression equation that described the relationship between interaural time delays, the distance sound travels from one ear to the other divided by the speed of sound, and the upp er frequency hearing limits for a variety of species (Figure 2.1). Animals with narrower heads had smaller in teraural time delays and typically needed higher frequency sensitivit y to be able to local ize sounds. Ketten et al. (1992) calculated the manatee intermeatal distance as 278 mm with a maximum acoustic travel time of 258 sec, and the intercochlear distance as 82 mm with a maximum acoustic travel time of 58 sec. When these time delays were plotted on 19

PAGE 33

Heffner and Heffners regression line, it appeared that manatees would need to be able to hear in the 50 kHz frequency range to be able to localize sounds (Figure 2.1). ( Ketten et al., 1992 ) Figure 2.1. Interaural distances vs. maximum frequency perceived at 60 dB SPL (from Ketten et al., 1992). Behavioral audiogram sensitivity data are plotted by eac h species interaural time distances. The gophers and manatees (ic=intercochlear; im=intermeatal) maximum perceived frequency varied significantly from the regression. The hearing range of the manatee has b een assessed through the development of an audiogram and by utilizing auditory evoke d potential techniques. Gerstein et al. (1999) obtained a behavioral audiogram fo r two manatees, which showed hearing thresholds that ranged from 0.5 kHz for one subject and 0.4 kHz for the other. The frequency range of best hearing was between 10 kHz and maximum sensitivity was ~50 dB re: 1 Pa at 16 and 18 kHz, decreasing by ~20 dB per octave from 0.8 to 0.4 kHz and 40 dB per octave above 26 kHz. Auditory evoked potential measurements have been obtained in several studies. Bullock et al. (1980; 1982) and Popov and Supin (1990) found that the highest frequency detection reach ed 35 kHz when tested in air and Klishen et al. (1990) found it reached 60 kHz when tested in water. More recently, Mann et al. 20

PAGE 34

(2005) found that detection reached 40 kHz when tested in water, results similar to those found by Bullock (1980; 1982), a nd Popov and Supin (1990). Given that the upper end of the manatee s hearing range lies between 40 -60 kHz, below or bordering the 50 kHz frequency ra nge estimate needed for localization, the prediction was made that manatees may not possess effective sound lo calization abilities (Ketten et al., 1992). This pr ediction however, was not suppo rted by the results of two separate four-choice (45o, 90o, 270o, 315o) sound localization studies (Gerstein, 1999; Colbert, 2005). Gerstein (1999) tested th e ability of one manatee, Stormy, to localize 0.5, 1.6, 3, 6, or 12 kHz tonal test signals from four speak ers located 1 m away from the subject at a depth of 1.5 m below the surface. All signa ls were presented at 125 dB re: 1 uPa, approximately 30 dB above white noise that was also projected through the speakers. The signals were pulsed for either 200 or 500 ms paused for 400 ms, and then repeated, thereby creating a 400 ms signal which sp anned an 800 ms duration and a 1,000 ms signal which spanned a 1,400 ms duration. Each condition was composed of 80 trials. Results indicated that conditional accur acy was well above the 25% chance level (Table 2.1). Overall accuracy for frequency ranged from 58-78% with the 400 ms signals and 56-88% with the 1,000 ms signals. Over all accuracy for location ranged from 6268% with the 400 ms signals and 68-74% with the 1,000 ms signals. Subject performance decreased as frequency and durat ion decreased. Given the subjects reduced level of performance with the low frequency stimuli, Gerstein suggested that manatees may have difficulty localizing low frequency boat engine noise. 21

PAGE 35

Table 2.1. Overall performance within and betw een location, duration and frequency conditions of Gersteins (1999) four-choice localization experiment. Results are based on 80 trials per condition with chance at 25%. 400 ms Signal (over an 800 ms duration) Location 0.5 kHz 1.6 kHz 3 kHz 6 kHz 12 kHz Overall Accuracy by Location 45o 60% 65% 65% 70% 80% 68% 90o 60% 60% 60% 60% 75% 63% 270o 55% 60% 60% 65% 70% 62% 315o 55% 60% 60% 65% 85% 65% Overall Accuracy by Frequency 58% 61% 61% 65% 78% 1,000 ms Signal (over a 1,400 ms duration) 45o 55% 65% 80% 80% 90% 74% 90o 55% 65% 65% 70% 85% 68% 270o 55% 60% 70% 70% 85% 68% 315o 60% 60% 75% 85% 90% 72% Overall Accuracy by Frequency 56% 63% 73% 76% 88% Colbert (2005) expanded upon the previous four-choice sound localization task (Gerstein, 1999) by testing the abilities of tw o manatees, Hugh and Buffett, to localize sounds that were systematically varied acr oss dimensions of ba ndwidth and duration. Two tonal signals were used, a 4 kHz tone that was midway between the 2.5.9 kHz fundamental frequency range of typical mana tee vocalizations (Now acek et al., 2003) and a 16 kHz tone that was in the 10 kHz range of manatee best hear ing (Gerstein et al., 1999). Broadband stimuli were also intr oduced which spanned a wide range of frequencies (0.2-20 kHz) as well as those restricted to high frequencies that had wavelengths that were shorte r than their interaural time distances (6-20 kHz) and low frequencies with wavelengths that were longer than their interaural time distances (0.2 kHz). Duration was manipulated within th e broadband conditions and included signal 22

PAGE 36

lengths of 200 ms that prohibited head movement as well as 500, 1,000, and 3,000 ms. All stimuli were tested at 100 dB re: 1 uPa ( 1.5 dB) from four speakers located 1.05 m away from the subject at a depth of 0.75 m below the surface (Figure 2.2). Although white noise was not introduced, exhibit back ground noise was continuous and typically below 500 Hz, indicating the possibility of masking at lower frequencies. Figure 2.2. Testing configuration for Colberts (2005) four-choice localization experiment. Test speakers (yellow circles) were located 105cm from the subject and .75m below the surface. The blue octagon represents the Test Trainers position, the green squa re represents the Data Recorders position and the orange triangle represents the St ationing Trainers position. Each of the 14 conditions was composed of 72 trials. Both subjects performed well above the 25% chance level for all of the broadband frequency conditions (Table 2.2). Hugh showed a drop in percentage correct as the broa dband signal duration decreased, but this result was not observed with Buffett. Both animals also performed above chance levels with the tonal signals, but with lower accu racy than with the broadband signals. 23

PAGE 37

Table 2.2. Overall performance within and between the duration and frequency conditions of Colberts (2005) four-choice localization experiment. Results ar e based on 72 trials per condition with chance at 25%. Frequency (kHz) Duration : 0.2 20 6 20 0.2 2 Mean 4 16 Hugh 200 ms 64% 51% 58% 58% 500 ms 71% 63% 57% 64% 1000 ms 74% 71% 65% 70% 3000 ms 93% 86% 81% 87% 49% 32% Mean 76% 68% 65% Buffett 200 ms 93% 89% 85% 89% 500 ms 85% 92% 86% 88% 1000 ms 93% 79% 92% 88% 3000 ms 88% 82% 92% 87% 44% 33% Mean 90% 86% 89% The overall broadband error rate, derived from the complete data set (excluding tonal results) collapsed across all cond itions, was only 11% for Buffett and 22% for Hugh. Frequency selection distributions (per cent of location selec tions by frequency, collapsed across duration) reve aled that although differences in performance accuracy were found between subjects within the br oadband signal conditions, errors were generally consistent, with most equally distribu ted to the locations adjacent to the correct location, however error distribut ion for the tonal signal condi tions were almost equally scattered among the four locations (Figure 2.3) Similar results were found for duration selection distributions (per cent of location selections by duration, collapsed across frequency) and those calculated for each of the individual broadba nd conditions (percent of location selections within th e 12 individual broa dband conditions). 24

PAGE 38

Figure 2.3. Percent correct and distribution of sound localization errors by frequency collapsed across duration in Colberts (2005) 4-choi ce sound localization study. Tona l conditions are presented in the bottom row. Correct speaker location is notated by double parentheses. Buffe tts results are presented above the grid lines in magent a and Hughs below in teal. The results from Colberts (2005) sound localization study suggested that although manatees could localize tonal signal s, they were better able to localize broadband noises as is typical with many species (Stevens & Newman 1936; Marler, 1955; Casseday & Neff, 1973), which likely acco unted for higher accuracy as compared 6-15 kHz All Durations 10050050100 ((45)) 315 270 90 10050050100 45 315 270 ((90)) 10050050100 45 315 ((270)) 90 10050050100 45 ((315)) 270 90 Percent Percent 0.2-20 kHz All Durations 10050050100 ((45)) 315 270 90 10050050100 45 315 270 ((90)) 10050050100 45 315 ((270)) 90 10050050100 45 ((315)) 270 90 Percent Percent 0.2-20 kHz 6-20 kHz 0.2-2 kHz All Durations 10050050100 ((45)) 315 270 90 10050050100 45 315 270 ((90)) 10050050100 45 315 ((270)) 90 10050050100 45 ((315)) 270 90 Percent Percent 100500050100 ((45)) 315 270 90 100500050100 45 315 270 ((90)) 100500050100 45 315 ((270)) 90 100500050100 45 ((315)) 270 90 Percent Percent 100500050100 ((45)) 315 270 90 100500050100 45 315 270 ((90)) 100500050100 45 315 ((270)) 90 100500050100 45 ((315)) 270 90 Percent Percent 16 kHz 4 kHz 0.2-20 kHz Hugh Buffett PercentSelected PercentSelected PercentSelected Broadband Si g nals Tonal Si g nals 25

PAGE 39

to those in Gersteins (1999) study. While results from both sound localization studies indicated that manatees were able to localize test signals that originated from a distance of ~1 m to the front 180o of the azimuth plane, questions remain regarding their ability to localize sounds within all 360o. The objective of this study wa s to investigate the manatees ability to localize test signals that were systematically varied across dimensions of bandwidth, duration and level as they originated from 45o angles within all 360o of the azimuth plane at a distance ~3 times greater than previously tested. 26

PAGE 40

Hypotheses Five hypotheses were made. The first pos ited that subjects would be able to localize all of the broadband test signals above the 12.5% chance level at all eight locations within the 360o. Prior manatee localization st udies (Gerstein, 1999; Colbert, 2005) demonstrated that subjects were able to determine the origin of sound sources to the front 180o, but had difficulty with tonal signals A field study which investigated manatee responses to controlled boater appro aches suggested that ma natees angled away from, increased swimming speed, and oriented towards deeper channel waters when boats with broadband engine noise approached from all directions (Nowacek et al., 2004). The second hypothesis declared that subj ects would have gr eater localization accuracy with the 0.2-24 kHz test signal at the 3,000 ms duration versus the 200 ms duration. Colberts (2005) four choice manatee sound localization study found that subject accuracy decreased as duration d ecreased. The 0.2-24 and 4 kHz signals presented in the current investigation serves as a means to determin e manatee localization abilities without their ability to orient towa rds the sound source during its presentation. The third hypothesis stated th at subjects would have grea ter localization accuracy to the anterior 180o than to the posterior 180o. Previous studies have suggested that the ability to localize a sound source may be influenced by multimodal sensory systems and be a function of visual orientation res ponses (Heffner, 1997). Reflexive visual 27

PAGE 41

orientation towards startling s ounds have been found in a wide variety of species at birth (or when their auditory systems become func tional) including humans, gulls, ducks, cats, rats and guinea pigs (Brown, 1994). The fourth hypothesis asserted that subject errors would have a higher distribution to the correct locations near est neighbors rather than to other locations. Colberts (2005) previous manatee localization study de monstrated that error distribution was highest amongst the two speak ers neighboring the test speak er than the other speakers when broadband frequency signals were tested. The final hypothesis contended that s ubjects would make more differentiation errors between speakers located at 0 o and 180 o than any other contralateral pairs. Middlebrooks and Green (1991) demonstrated that front to back sound localization confusions were typical with human subjects an d attributed these results to the fact that stimulus locations lie in mirror symmetry with respect to the subjects ears which eliminate interaural time of arrival, phase and level cues. 28

PAGE 42

Materials and Methods Subjects The subjects of this study were tw o captive-born male Florida manatees ( Trichechus manatus latirostris ) that reside at Mote Marine Laboratory and Aquarium in Sarasota, Florida. All procedures used were permitted through the United States Fish and Wildlife Service (Permit # MA837923-6) and approved by the Institutional Animal Care and Use Committee of Mote Marine Laboratory and Aquarium. At the inception of this study Hugh was 23 years of age, weighed 547 kg and was 310 cm in length, while Buffett was 20 years of age, weighed 773 kg and was 334 cm in length. They were housed in a 265,000 liter exhibit that was composed of th ree inter-connected sections: a 3.6 x 4.5 x 1.5 m Medical Pool, a 4.3 x 4.9 x 1.5 m Shelf Area, and a 9.1 x 9.1 x 3 m Exhibit Area (Figure 2.4). Both animals had acquired an extensive training history over the previous seven years and participated in an auditory evoked poten tial study (Mann et al., 2005) and a four-choice sound local ization study (Colbert, 2005) making them excellent candidates for this project. In addition, they had been behaviorally conditioned for husbandry procedures (Colbert et al., 2001), a serum and urine creatinine study (Manire et al., 2003), a visual acuity study (Bauer et al., 2003), a l ung capacity study (Kirkpatrick et al., 2002), and a vibrissae tactile sensitivity study (Bauer et al., 2005). 29

PAGE 43

Figure 2.4: Diagram of the 265,000 L manatee exhibit composed of a Medical Pool, Shelf Area, and Exhibit Area. The lines in the Medical Pool represent a distance scale, used in a previous study that was painted on the floor of the exhibit. The oval masses in the Exhibit Area represent outcroppings in the bottom terrain (built of cement) to co nceal the two floor-level filtration dr ains (gratings). The rectangles represent a tree log and stump (built of cement). Subject Training The majority of animal training procedur es utilized in Colberts (2005) fourchoice sound localization st udy were maintained in this study, although some modifications were necessary (see Colber t, 2005 for the specific animal training procedures). The testing set-up was moved from the Shelf Area to the Exhibit Area where the stationing bar and test speakers we re lowered from a depth of 0.75 m to 1.5 m, half way between the waters surface and the exhibit bottom. Eight underwater speakers (Aquasonic AC 339) were positioned 45o apart at 0o, 45o, 90o, 135o, 180o, 225o, 270o and 315o. The distance between the stationing appa ratus and the test speakers was increased from 1.05 m to 3.05 m. Because the water in the deep area had a strong counter-current circulation originating at ~160o, the subjects needed to face east rather than south to reduce drag (Figure 2.5). 9.1 x 9.1 x 3 m Exhibit Area 4.3 x 4.9 x 1.5 m Shelf Area 4.5 x 3.6 x 1.5 m Medical Pool 30

PAGE 44

Deep Exhibit Area 3.05 m 270o 315o 225o 180o 45 o90o 135o 0o Shelf Area Medical Pool 1.05 m 45o270o 90 o315o Figure 2.5. Training setup comparison for the Colberts four-choice and eight-choice sound localization experiments. The setup was moved from the Shelf Area (outlined by red dashed line) to the Exhibit Area where the subjects faced east rather than south, the st ationing bar and test speakers were lowered from a depth of 0.75 m to 1.5, and the distance between the two was increased from 1.05 m to 3.05 m. The procedures utilized in Colbert s (2005) four-choice s ound localization study required that the subjects be trained to respond to a unique station sign al that was played from a speaker located on a stationing apparatu s. The call-to-station signal ranged from 10 to 20 kHz and played for a 2000 ms durati on, however Buffetts repeated at a slower rate of 1.5/sec while Hughs repeated at a faster rate of 5/sec. In response to their stationing signal, each subject was trained to position the crease on the top of its rostrum (approximately 10 cm posterior to the nostrils) up against a stationing bar located at the bottom of the stationing appara tus. The manatee remained stationed until a test signal was played from one of the four underwater speakers that were suspended from poles that pivoted, whereupon he swam to and pushed the speaker from which the sound originated. If correct, a secondary reinfor cer signal was emitted from th e test speaker and the subject returned to the stationing de vice to be fed primary food reinforcement. The secondary reinforcement signals were programmed in RPvds and matched the unique whistles used 31

PAGE 45

to bridge each animal (Appendix A). Buffetts reinforcement signal ranged from 14 to 120 kHz with a peak at 53 kHz, while Hughs had more of a warble to it and ranged from 12 to 110 kHz with a peak at 27 kHz (Figure 2.6). If incorrect, the stationing signal was played from the stationing apparatus speaker and the subject re-positioned correctly with no primary or secondary reinforcement give n, and waited a minimum of 60 seconds for the initiation of the next trial. Figure 2.6. Power spectra (top) and spectrograms (b ottom) of the secondary reinforcement signals. Buffetts ranged from 14 to 120 kHz with a peak at 53 kHz, while Hughs had more of a warble to it and ranged from 12 to 110 kHz with a peak at 27 kHz. Although these same procedures were used with the eight-choice sound localization study, several behaviors needed to be re-shaped to meet the change in stationing direction (from south to east), increased stationing depth (from 0.75 m to 1.5 m), and extended test speaker distance (from 1.05 m to 3.05 m) criteria. All new behaviors were trained using standard positive classical and operant conditioning techniques. Each animals unique secondary reinforcement whistle was used to bridge correct behaviors as they occurred and primar y reinforcers included bi te size pieces of 32

PAGE 46

apples, beets and baby peeled carrots. Zupr eem monkey biscuits, one of the manatees preferred foods, were used to reward especially desired behaviors during shaping procedures. In addition, verbal and tactile secondary reinforcers were used. All new or modified behaviors were shaped by reinfor cement of successive approximations (Pepper & Defran, 1975). Undesirable behaviors were ignored and time-outs, (Pepper & Defran, 1975; Domjan, 1998) or the removal of the oppor tunity to receive reinforcement, were used if a string of undesirable behaviors occurred. Both animals had previously been traine d to station and follow their own personal target, and in the early stages of shaping the stationing behavior when facing east, the trainer used the subjects targ et to guide him to the shorte r, four-choice sound localization stationing bar which was adapted to fit ove r a platform suspended across the Exhibit Area. Shaping of the correct position was f acilitated by the trainers reaching into the water to help maneuver the manatee into the correct position. When this was accomplished, the 23 cm wide stationing apparatus, constructed from 2.54 cm diameter polyvinyl chloride (PVC) pipe, was modified to reach 1.5 m below the surface of the water (Figure 2.7). Shaping of the sta tioning behavior at the 1.5 m depth was accomplished by lowering the stationing apparatus in gradual steps. 33

PAGE 47

Figure 2.7. Stationing apparatus used in the eight-choice sound localization study. The black circle represents the speaker that played the stationing tones. The subject pressed the crease of his rostrum up against the gray stationing bar on the bottom. Water Line 1.5 m 23cm Eight underwater speakers (Aquaso nic AC 339) were positioned 45o apart at 0o, 45o, 90o, 135o, 180o, 225o, 270o and 315o (with the subject facing 0o) (Figure 2.8). Each speaker was suspended from a 1.88 cm diameter PVC rod at a depth of 1.5 m. The rods were bolted to aluminum beams that radiat ed out from two suspension supports spanning the Exhibit Area, and were desi gned to pivot so that the spea ker at the bottom of the rod could be pushed backwards while the top of th e rod tilted forward in a pendulum motion. A 0.2-24 kHz, 3000 ms broadband signal was used to train the subjects to swim to the speakers at the increased distance, and to those introduced behind them. Distance 34

PAGE 48

increases were initiated by introducing only the 45o, 90o, 270o and 315o test speakers at a distance of 2 m and subject responses were a ssisted by the trainers use of the subjects target to guide him towards the speaker if n eeded. When subject responses were reliable, the distance was increased to 3.05 m and shap ing continued until their reactions were again consistent. Test speakers 0o, 135o, 180o and 225o were then introduced at the 3.05 m distance and subjects were guided by their ta rget towards the correct test speaker until each reliably oriented towards the 90o region the signal originated from. Figure 2.8. Testing setup for the eight-choice sound localization experiment. Subjects stationed facing 0o and test speakers were suspended from pivoting rods at 45o, 90o, 135o, 180o, 225o, 270o, and 315o. The blue octagon represents the Test Trainers location, the gr een square represents the Data Recorders location, and the orange triangle represents the Stationing Trainers location. Deep Exhibit Area 3. 05 m 270o315o225o180o 45 o 9 0 o135o 0o Shelf A rea Medical Pool Experimental Design An eight alternative forced-choice discrimi nation paradigm was used to test the sound localization abilities of two Florida manatees, Trichechus manatus latirostris Eight underwater test speakers (Aquasonic AC 339) were positioned in a 6.10 m diameter circle surrounding a stationi ng/listening apparatus at 0o, 45o, 90o, 135o, 180o, 225o, 270o and 315o and a depth of 1.5 m. (Figure 2.8). Each subject was trained to position the top of its rostrum, approximately 10 cm posterior to the nostrils, up ag ainst a stationing bar 35

PAGE 49

positioned at mid-water depth (1.5 m) in response to his stationing signal. The subject remained stationed facing 0o until a test signal was played from one of the eight test speakers. Upon hearing the test signal, th e subject would swim to and push the speaker from which he believed the sound originated. If correct, a secondary reinforcer signal (1.4 12 kHz with a peak at 5.3 kHz for Buff ett, 1.2 11 kHz with a peak at 2.7 kHz for Hugh) was emitted from the test speaker and th e subject returned to the stationing device to be fed a primary reinforcement of food (a pples, beets and carrots). If incorrect, the stationing tone was played fr om the stationing apparatus speaker and the subject restationed correctly with no reinforcement given, to await a minimum of 30 seconds before the initiation of the next trial. All training and testing sessions were run between 0700 and 1000 h five days per week before the Aquarium was open to the public. The manatees daily ration of food (72 heads of romaine lettuce and 12 bunches of kale) was fed to the animals from 1200 to 1400 h and was usually consumed by 1700 h, leaving a 14 to 16 hour overnight fast before training was initiated the following morning. Three people were required to run the expe riment: a Test Trainer, Data Recorder, and Station Trainer (Figure 2.8). The Test Trainer, blind to the test stimulus locations, wore noise-masking headphones and was positioned facing 180o on a platform suspended across the Exhibit Area. The Test Trainer en sured that a minimum 30 second inter-trial interval was met, the subject was positioned correctly, initiated each trial by verbally stating tone to the Data Recorder, info rmed the Data Recorder which location the subject selected, determined if the subject was correct by looking at the Data-recorder when he/she came into view for the appropria te head nod or shake, provided the subject 36

PAGE 50

primary reinforcement if he was correct, and requested the subject to station by stating station to the Data Recorder. The Data Recorder was positioned behind a laptop computer, out of sight of the Test Trainer and subject, and set up the expe rimental conditions needed for each session on the computer using a graphical user interf ace that was programmed in Visual C (see Appendix A for set up protocols). The Data Reco rder initiated trials when instructed to do so by the Test Trainer, informed the Test Trainer and subject if the location selection was correct by leaning out from behind the comp uter to provide a head nod to the Test Trainer and playing the subjects secondary reinforcer signal, or incorrect by providing a head shake and playing the station signal, re corded all data on a tank-side session sheet (Appendix B), and ran the video equipment. The Station Trainer was positioned at the no rtheast end of the Medical Pool out of the test subjects line of si ght and was responsible for holding the non-test animal at station throughout the subjects session. The Station Trai ner was unaware of the correct locations and unable to see th e subject during testing. Acoustic Stimuli A total of six experimental conditions were tested (Table 2.3). A speaker frequency response normalization procedure (see experimental controls section) generated test signals which were presented at the same spectrum level, meaning that signals with broader frequency spectra had louder root mean square (rms) amplitudes (Figure 2.9). The average spectrum level (dB re 1 Pa/sqrt (Hz)) of each stimulus was defined by condition (Table 2.3). 37

PAGE 51

Table 2.3. Frequency, duration and level conditions of the eight-choice manatee localization experiment. Frequency (kHz) Duration (ms) Averaged Level (dB re 1 Pa/sqrt (Hz)) 4 200 101 0.2-24 200 95 0.2-24 3000 95 18-24 3000 80 0.2-1.5 3000 110 0.2-24 3000 95 then decreasing to < 75% Figure 2.9. Sound calibration from the eight test spea kers that were normalized with a 500-tap FIR filter (top) and spectrum level of the background exhibit noise (bottom) in the eight choice localization experiment. Sound from each speaker was normalized to approximately follow the shape of the manatee audiogram, with decreasing sound levels at higher frequencies. Each curve shows the recording from one of the eight speakers. The 0.2-24 kHz stimulus was tested at both 3000 ms and 200 ms durations and spanned a wider range of frequencies than ha d been previously tested (0.2-20 kHz). The 18-24 kHz stimulus was tested at a 3,000 ms duration and was composed of a more extreme higher frequency range th an previously tested (6-20 kHz) with wavelengths that were shorter than a manatees interaural distance. The 0.2-1.5 kHz stimulus was presented at a 3,000 ms duration and was compri sed of a slightly smaller range of low 38

PAGE 52

frequencies than previously tested (0.2-2 kHz) with wavelengths that were greater than a manatees interaural distance. The 4 kHz tonal signal was midway between the 2.5.9 kHz fundamental frequency range of typical manatee vocal izations (Nowacek et al., 2003) and was presented at 200 ms, a duration shorter than ha d previously been tested (3,000 ms) which prohibited subject head movement adjustments while it was presented. The 4 and 16 kHz, tonal signals used in Colberts (2005) four-choice localization study were only tested at a 3,000 ms duration because the s ubjects exhibited strong signs of behavioral frustration at lower durations. To prevent th is frustration from occurring with the shorter 200 ms 4 kHz signal at the more distant sp eaker locations, four tonal probes were included in 16-trial blocks of 0.2-24 kHz, 3,000 ms signals until 80 trials were completed (Appendix B). In the sixth condition, the level of the 0.2-24 kHz, 3,000 ms, 95 dB test signal was attenuated in 3 dB increments per block until each subjects overall percent correct within a 16-trial block fell below 75%. A total of 80 trials were collected at the level in which the subject fell below 75% accuracy to ensure consistency. Signal Generation & Programming All signals including each subjects stat ion and secondary reinforcement signals and the test stimuli were programmed in RPvds language (Appendix A), digitally generated by a Tucker-Davis Technologies real-time processor (RP2.1), and attenuated with a programmable attenuator (PA5) to contro l level. Signals were amplified with a Hafler power amplifier and switched to the eight test speakers through a power multiplexer (PM2R). A separate digital to analog channel was used to generate the 39

PAGE 53

stationing signal from the speaker located on th e stationing apparatus in the center of the array. MATLAB programming was used to generate blocks of sixteen trials that were counterbalanced between the ei ght speaker locations in a quasi-random order, meaning that the test signal location wa s randomized, but had a criterion of no more than two trials in a row from the same location. A Dell laptop computer (model Latitude D505) with Windows XP was used to run the signal generation equipment, set up the testing conditions and to automatically download the parameters of each trial into an Excel file (Appendix B). Trials were initiated and completed through an electronic control box which was connected to the RP2 unit, and then into the laptop computer (Figure 2.9). The control box had four buttons with corresponding colored LED light s built into it. The station signal button was used to call the subject to station and the actual speaker switching occurred while the station signal was played. The test signal button was used to play each conditions test signal once per trial. The corr ect button was used to play th e subjects unique secondary reinforcement signal for corre ct location selections. The wrong button was used to digitally record incorrect location selectio ns and was immediately followed by playing the station signal. The four LED lights provided visual veri fication that their corresponding signals were play ed and that the trial was dow nloaded into the Excel file. 40

PAGE 54

Figure 2.10. Electronic button box used to run the sessions and automatically download each trial into a digital excel file. Data Recording Data from each session were recorded in three ways. Automated digital computer reports were uploaded into an Excel file as each trial occurred in real-time on the laptop computer. The Data Recorder documented session information by completing a tank-side data sheet (Appendix C). This information was then manually entered into a Microsoft Access database created on a Dell desktop computer (model Dimension 8300) after the completion of each session. This database wa s designed specifically for this experiment and had a user-friendly data entry screen (Figure 2.10). All data entered into the database were double-checked for accuracy by a second trainer after th ey were entered. Finally, each test block was video recorde d. A Sony variable zoom, high resolution, outdoor weather proof, color dome camera (m odel SCW-CD358DVP) was attached to the trainers platform directly over the subjects head a nd connected to a Sony digital video camera (model DCR-TRV50). Pre-printed data sheets which identified the date, subject, test frequency, and s ound duration were video recorded prior to the initiation of each block to stipulate each blocks parameters as they occurred. Station S i g nal Test S i g nal Correct/ Bridge Wrong Station Signal LED Correct LED Wrong LED Test Signal LED 41

PAGE 55

Figure 2.11. Data entry screen used to enter session information into the Access database. Experimental Controls Several experimental controls were put into place to ensure that cues which might possibly arise from trainers, test signals, speakers and their locatio ns were avoided, and that the subjects were motivated for each te sting session. All personnel were positioned out of the test subjects line of sight except for the Test Trainer. The Test Trainer was required to wear sound-dampening headphones to avoid the possibility of hearing the test signals and was ignorant of th e test signals location. The Data Recorder was the only individual who knew which location the test signal would originate from and only obtained this knowledge at the initiation of each trial. Th e Data Recorder was seated approximately 6 m to the back of the Test Trainer behind the laptop computer screen and was not visible until after the subject had made his location selection at the end of each trial. At this point the Test Trainer w ould look backwards towards the Data Recorder who would move into view to indicate if the subjects choice was correct or incorrect. 42

PAGE 56

All test signals had a 100 ms rise-fall time to eliminate transients and levels were randomly presented +/1.5 dB over the nominal ac oustic pressure to eliminate level cues. A speaker frequency response normalization pr ocedure was developed to eliminate the possibility that small differences in the speak ers and their locations in the exhibit would produce localization cues that could be detected by the mana tees (Figure 2.9). This was accomplished by measuring each speakers fr equency response from the stationing apparatus via a hydrophone (HTI 96 min; sensitivity -164 dBV/ Pa from 0.2 Hz to 37 kHz) and then developing a 500-tap FIR filter for each speaker to produce normalized responses over the frequency bands. Note th at the frequency res ponse was not flat, but louder at lower sound frequencies, similar to the spectra produced by boats. The signal also tracked the manatee audiogram which wa s more sensitive at higher frequencies (Gerstein et al., 1999). No effort was made to test the subjec ts in an anechoic setting, in fact, the exhibit background noise (pumps for filtration) was continuous throughout testing. The spectrum level of the exhibit noise was ~30-50 dB re 1 Pa lower than the normalized speaker outputs (Figure 2.9). To control for motivational effects, each animals session was started with eight warm-up trials, one from each location in a randomized order, and ended with four cool-down trials, the locations of which were randomly generated via the computer program. The signal stimulus used for these trials was the same 3000 ms, 0.2 kHz, broadband noise used throughout training. In addition, two criteria were defined as reasons to drop a test block. The first stipul ated that a minimum performance accuracy of 75% was required on the warm-up and cool -down trials. The second defined a maximum allowance of any combination of th ree interruptions from the non-test manatee 43

PAGE 57

44 and/or leaves or attempted leaves from the test subject per block. If a block was dropped, the experimental condition was re peated in the next session.

PAGE 58

Results Training was initiated on July 7, 2006 and completed for Hugh on September 1, 2006 and for Buffett on August 18, 2006. Both manatees learned the new requirements of the task easily. Testing with the 0.24-24 kHz, 3,000 ms signal took place from September 4 to October 20, 2006 for Hugh and from August 21 to October 12, 2006 for Buffett. Testing that attenuated the 0.24-24 kHz, 3,000 ms signal took place from October 23 to November 20, 2006 for Hugh a nd from October 13 to December 15, 2006 for Buffett. Testing with the 18-24 kHz, 3,000 ms signal took place from November 22 to December 5, 2006 for Hugh and from December 20, 2006 to January 2, 2007 for Buffett. Testing with the 0.2-1.5 kHz, 3,000 ms signal took place from December 6, 2006 to January 16, 2007 for Hugh and from Janua ry 3 to January 18, 2007 for Buffett. Testing with the 0.2-24 kHz, 200 ms signal t ook place from February 6 to February 20, 2007 for Hugh and from January 31 to February 20, 2007 for Buffett. Testing with the 4 kHz, 200 ms signal took place from February 21 to April 19, 2007 for Hugh and from February 21 to April 2, 2007 for Buffett. A total of 27 blocks were dropped from both manatees as they met the pre-defined drop criteria. Buffett dropped 17 and Hugh dropped 10. Three data analyses were conducted for each subject including one that examined the possibility of speaker artifact or lo cation cues, one that determined overall performance accuracy, and one that investigated selecti on distribution. The frequency response normalization procedure was integrat ed during the training of the eight choice 45

PAGE 59

sound localization study as a safe guard to eliminate the possibility that small differences arising from the speakers themselves and/or their positions in the exhibit would be detected and used to facilitat e the subjects response (Figure 2.11). Determination of accuracy by speaker location before and after the speaker frequency normalization calibration was done showed no large or c onsistent differences in performance, suggesting that the manatees di d not use other cues for sound localization (Figure 2.12). Subject Performance Before & After Speaker Normalization0 10 20 30 40 50 60 70 80 90 100 0o45o90o135o180o225o270o315o Speaker LocationPercent Correct Hugh Before Hugh After Buffett Before Buffett After Figure 2.12. Subject accuracy before and after speaker normalization calibration. Note that the speaker at 180 degrees is biased by multiple presentations during the before calibration. Overall performance accuracy was determined and described in Table 2.4. Percentage correct was calculated for each subject based on 15 trials per speaker for the 0.2-24 kHz, 200 ms condition and 10 trials per sp eaker for all other conditions. The level of the 0.2-24 kHz. 3,000 ms, 95 dB signal was decreased in 3 dB increments if the subject achieved 75 % correct or greater for tw o consecutive blocks. Five blocks were completed when accuracy fell below 75% (86 dB ), and then five additional blocks were 46

PAGE 60

completed at a level that was 6 dB lower th an this point to ex amine how performance changed in these increments. Table 2.4. Results for the conditions of the eight-choice localization experiment with chance level at 12.5%. Percentages are based on 15 trials/speak er for the 0.2-24 kHz, 200 ms condition and 10 trials/speaker for all other conditions. Trials which measured accuracy as level was decreased are shown in italics. Frequency & Averaged Level (dB re 1 Pa/sqrt (Hz)) 4 kHz 101 dB 0.2-24 95 dB 18-24 kHz 80 dB 0.2-1.5 kHz 110 dB Duration Hugh 200 ms 14% 55% 3000 ms 69% 40% 46% 3000 ms; 86 dB 72% 3000 ms; 80 dB 48% Buffett 200 ms 20% 65% 3000 ms 79% 60% 64% 3000 ms; 86 dB 77% 3000 ms; 80 dB 56% Both subjects performed well above the 12.5% chance level for all of the broadband frequencies tested. When the 4 kHz tonal signal was tested however, performance decreased dramatically for bot h subjects with Hughs accuracy at only 14% Buffetts at 20%. When level was decrease d with the 0.2-24 kHz signal, both were able to localize the signal over a fairly large sound level range, however, Hughs performance deteriorated more than Buffetts. Selection distribution was investigated fo r each of the conditions tested. The 0.224 kHz, 3,000 ms signal was composed of the wi dest range of freque ncies presented at the longest duration tested, making it the eas iest discernible signal. The 3,000 ms duration allowed subjects to move their heads to physically and visually orient towards speaker locations to the front 180o during signal presentation, but did not provide enough time for orientation towards speakers behind them. Results showed that the few errors 47

PAGE 61

made were primarily located at the near est neighbor of the te st speaker for both subjects but became somewhat more disp ersed when they originated from 135o and 180o for Buffett and more widely di spersed when they came from 180o and 225o. Hugh never selected the 180o location and made back to front c onfusions on 20% of the trials when the signal originated at 180o (Figure 2.13). Figure 2.13. Selection distribution with the 0.2-24 kHz, 3,000 ms, 95 dB re 1 Pa test signal. The percent correct is notated at the locations de marked by the yellow circles. Hughs results are always presented to the right of the graph lines in teal a nd Buffetts are to the left in maroon. The exterior circle of the grid represents 100% accuracy, the middle 50% and the inner 0%. Hugh Buffett 45 315 270 90 135 225 ((0)) 180 0.2-24 kHz, 3,000 ms, 95 dB ((45)) 315 270 90 135 225 0 180 45 315 270 ((90)) 135 225 0 180 45 315 270 90 ((135)) 225 0 180 45 315 270 90 135 225 0 ((180)) 45 315 ((270)) 90 135 225 0 180 45 ((315)) 270 90 135 225 0 180 45 315 270 90 135 ((225)) 0 180 48

PAGE 62

The 0.2-24 kHz, 200 ms signal was also composed of the widest range of frequencies, however the shorter duration di d not provide enough time for subjects to physically or visually orient towards any of th e speaker locations. Errors tended to be to the nearest neighbor of the test speaker for both subjects but became somewhat more dispersed when signals came from 135o, 180o and 270o for Buffett and more widely dispersed when they came from 180o for Hugh. Hugh never selected the 180o location and made back to front confusions on 27% of the trials when the signal originated at 180o. Hugh also made contralateral confusi ons on 7% of the trials when the signal originated from 315o (Figure 2.14). 49

PAGE 63

Figure 2.14. Selection distribution with the 0.2-24 kHz, 200 ms, 95 dB re 1 Pa test signal. The percent correct is notated at the locations de marked by the yellow circles. Hughs results are always presented to the right of the graph lines in teal a nd Buffetts are to the left in maroon. The exterior circle of the grid represents 100% accuracy, the middle 50% and the inner 0%. The 18-24 kHz, 3,000 ms signal was com posed of higher frequencies with wavelengths that were shorter th an the manatees intermeatal or intercochlear distances. Errors tended to be to the n earest neighbor of the test speak er with both subjects but became more dispersed when signals came from 135o and 270o for Buffett and more widely dispersed when they came from 135o, 180o, 225o and 270o for Hugh. Buffett made contralateral confusions on 10% of the trials when the signal originated from 90o 0.2-24 kHz, 200 ms, 95 dB Hugh Buffett 45 315 270 90 135 225 ((0)) 180 ((45)) 315 270 90 135 225 0 180 45 315 270 ((90)) 135 225 0 180 45 315 270 90 ((135)) 225 0 180 45 315 270 90 135 225 0 ((180)) 45 315 270 90 135 ((225)) 0 180 45 315 ((270)) 90 135 225 0 180 45 ((315)) 270 90 135 225 0 180 50

PAGE 64

and 180o. Hugh never selected the 180o location and made back to front confusions on 10% of the trials when the signal originated at 180o. Hugh also made contralateral confusions on 10 % of the trials wh en the signal originated from 225o and 270o (Figure 2.15). Figure 2.15. Selection distribution with the 18-24 kHz, 3,000 ms, 80 dB re 1 Pa test signal. The percent correct is notated at the locations de marked by the yellow circles. Hughs results are always presented to the right of the graph lines in teal a nd Buffetts are to the left in maroon. The exterior circle of the grid represents 100% accuracy, the middle 50% and the inner 0%. The 0.2-1.5 kHz, 3,000 ms signal was composed of lower frequencies with wavelengths that were longer th an the manatees intermeatal or intercochlear distances. Hugh Buffett 18-24 kHz, 3,000 ms, 80 dB 45 315 270 90 135 225 ((0)) 180 ((45)) 315 270 90 135 225 0 180 45 315 270 ((90)) 135 225 0 180 45 315 270 90 ((135)) 225 0 180 45 315 270 90 135 225 0 ((180)) 45 315 270 90 135 ((225)) 0 180 45 315 ((270)) 90 135 225 0 180 45 ((315)) 270 90 135 225 0 180 51

PAGE 65

Errors tended to be to the nearest neighbor for all speaker locations but became more dispersed when signals originated from 180o, 225o and 270o for both Hugh and Buffett. Hugh never selected the 180o location and made back to fr ont confusions on 10% of the trials when the signal originated at 180o (Figure 2.16). Figure 2.16. Selection distribution with the 0.2-1.5 kHz, 3,000 ms, 110 dB re 1 Pa test signal. The percent correct is notated at the locations de marked by the yellow circles. Hughs results are always presented to the right of the graph lines in teal a nd Buffetts are to the left in maroon. The exterior circle of the grid represents 100% accuracy, the middle 50% and the inner 0%. The 4 kHz, 200 ms tonal signal had a wavelength longer than the manatees intermeatal or intercochlear di stances and was presented at a duration that did not provide Hugh Buffett 0.2-1.5 kHz, 3,000 ms, 110 dB 45 315 270 90 135 225 ((0)) 180 ((45)) 315 270 90 135 225 0 180 45 315 270 90 ((135)) 225 0 180 45 315 270 ((90)) 135 225 0 180 45 315 270 90 135 225 0 ((180)) 45 315 270 90 135 ((225)) 0 180 45 315 ((270)) 90 135 225 0 180 45 ((315)) 270 90 135 225 0 180 52

PAGE 66

enough time for subjects to physically or visu ally orient towards any of the speaker locations. Errors were scattered among th e locations with no obvious patterns observed for both subjects. Contrala teral confusions were not considered due to the high variability of speaker locati on selections (Figure 2.17). Figure 2.17. Selection distribution with the 4 kHz, 200 ms, 101 dB re 1 Pa test signal. The percent correct is notated at the locations demarked by the yellow circle s. Hughs results are alwa ys presented to the right of the graph lines in teal and Buffetts are to the left in maroon. The exterior circ le of the grid represents 100% accuracy, the middle 50% and the inner 0%. Hugh Buffett 4 kHz, 200 ms, 101 dB 45 315 270 90 135 225 ((0)) 180 ((45)) 315 270 90 135 225 0 180 45 315 270 ((90)) 135 225 0 180 45 315 270 90 ((135)) 225 0 180 45 315 270 90 135 225 0 ((180)) 45 315 270 90 135 ((225)) 0 180 45 315 ((270)) 90 135 225 0 180 45 ((315)) 270 90 135 225 0 180 53

PAGE 67

The data were also evaluated by determin ing the number correct for each of the broadband stimuli and separating these results into three regions: front, back and side. These numbers were averaged by the total numbe r of trials presented at each location for each stimulus (15 trials for the 0.2-24 kHz, 200 ms stimulus; 10 trials for the remaining stimuli) (Table 2.5). Buffetts accuracy ranged between 83% 97% when the stimuli were presented in front (0o, 45o, 315o), 40% 67% when presented in back (135o, 180o, 225o), and 60% 75% when presented to the sides (90o and 270o). Hughs accuracy ranged between 62% 90% when the stimuli were presented in front, 13% 51% when presented in back, and 45% 80% when presented to the sides. Table 2.5. Average percents correct by front, back an d side regions. The numbers of correct trials were averaged by the total number of tr ials presented at each location for each stimulus. Averages were based upon 15 trials per location with the 0.2-24 kHz, 200 ms stimulus and 10 trials per location for the remaining stimuli. Hugh Front Back Side 0o 45o 315o 135o 180o 225o 90o 270o 0.2-24 kHz, 200 ms 10 14 4 62% 12 0 11 51% 13 11 80% 0.2-24 kHz, 3000 ms 10 10 7 90% 9 0 6 50% 8 5 65% 18-24 kHz, 3000 ms 7 9 4 67% 4 0 0 13% 6 2 40% 0.2-1.5 kHz, 3000 ms 6 8 6 67% 7 0 1 27% 6 3 45% Buffett Front Back Side 0o 45o 315o 135o 180o 225o 90o 270o 0.2-24 kHz, 200 ms 12 12 15 87% 6 7 5 40% 11 10 70% 0.2-24 kHz, 3000 ms 10 10 9 97% 7 6 7 67% 9 5 75% 18-24 kHz, 3000 ms 10 9 6 83% 1 7 5 43% 6 4 50% 0.2-1.5 kHz, 3000 ms 10 10 5 83% 2 5 7 47% 8 4 60% 54

PAGE 68

Discussion The results of this study provide info rmation about the manatees ability to localize specific broad-band a nd tonal signals of specific durations and levels in a controlled environment. Numerous controls were put in place to avoid test signal distortions and/or the projection and recogni tion of speaker artifac t or location cues. These included the incorporation of a 100 ms rise-fall time within signals to eliminate transients, the addition of a +/1.5 dB randomi zation of signal levels between trials to eliminate level cues, switching the test sign al location during th e presentation of the stationing tone to avoid tran sients, and the frequency response normalization procedure done between speakers. Analysis of the te st signals showed no obvious temporal or harmonic distortions and performance prio r to and after frequency normalization calibration procedures showed no large or c onsistent differences, suggesting that the subjects were localizing the actual test signals and not artifact or spatial cues. The subjects of this study were readily able to adapt behaviors learned in a prior four-choice sound localization study (Colbert, 20 05) to meet the change in stationing direction (from south to eas t), increased stationing de pth (from 0.75 m to 1.5 m), and extended test speaker distance (from 1.05 m to 3.05 m) criteria for the eight choice paradigm. Reshaping of these behaviors took approximately six weeks to complete. Testing was completed in approximately ei ght months. Results indicated that the subjects were able to localize all of the te st signals specified within the conditions and, similarly to the two prior four choice sound localization studies (Gerstein, 1999; Colbert, 55

PAGE 69

2005), do not support the anatomi cal hypothesis that suggested manatees may be poor at sound localization (Ketten, 1992). The first hypothesis posited that subjects would be able to localize all of the broadband test signals above the 12.5% chance level at all eight locations within the 360o. Results indicated that Buffett was capable of localizing all of the broad-band signals, even the 0.2-24 kHz, 200 ms shorter signal, when they originated from all angles including behind him. Hugh demonstrated th at he was able to localize all of same signals when they originated from all locations except 180o. Hugh never selected the 180o speaker when the broadband signals were test ed but seemed to instead default to the speakers located at 135o or 270o most of the time. Both subjects had difficulty localizing the 4 kHz, 200 ms tonal signal at all locations and their speaker selections were distributed randomly. Hugh did select the 180o speaker in this condition, however these selections also appeared to be random. This hypothesis was supported by the subjects performance with the broadband stimuli except for Hughs performance at the 180o location. Interestingly, both subjects were ab le to localize the 0.2-24 kHz test signal over a fairly large sound level range although H ughs accuracy decreased more rapidly than Buffetts. Although psychoacoustic studies often us e tonal sound stimu li in a controlled setting, studies with many species have demons trated that broadband signals are easier to localize than tonal signals (Stevens & Ne wman 1936; Marler, 1955; Casseday & Neff, 1973). The manatees natural environment contains a multitude of complex sounds that are primarily broadband and have rapid amp litude, frequency and bandwidth fluctuations on an ongoing basis. Recreational boat engine noise is characterized as broadband with 56

PAGE 70

a typical dominant frequency range of 0.01 kHz although it can re ach over 20 kHz with the 1/3-octave source levels at 1 m for sm all motorboats estimated at 120-160 dB re 1 Pa (Miksis-Olds, 2006; Gerstein, 1999; Richar dson et al., 1995). The subjects ability to localize broadband test signals ranging fr om a 80 110 dB re 1 Pa spectrum level suggests that they are able to localize typi cal recreational boat e ngine noises. Manatee vocalizations, categorized as chirps, squeaks and squeals, are characteristically short tonal complexe s which contain several harmonics. The fundamental frequencies of manatee vocal izations range from 2.5.9 kHz, but can extend up to 15 kHz (Nowacek et al., 2003). Although Buffetts 20% accuracy with the 4 kHz, 200 ms test signal was above the 12.5% chance level, Hughs accuracy was only 14%. The decreased accuracy with the 4 kHz tonal signal might imply that localization of manatee tonal vocalizations would be difficult, however the harmonics of different frequencies contained within these vocalizati ons likely provide additi onal cues to aid in this capacity. Some vocalizations transi tion from a tonal harmonic complex to more strongly modulated calls cove ring a greater frequency range and are often produced by calves, facilitating locali zation (Nowacek, et al., 2003; Mann et al., 2005; OShea & Poche, 2006). The second hypothesis declared that s ubjects would have greater localization accuracy with the 3,000 ms, 0.2-24 kHz test si gnal than the 200 ms, 0.2-24 kHz test signal. This hypothesis was supported by both subjects performance. Hugh had 69% accuracy with the 3,000 ms duration and 55% accuracy with the 200 ms duration. Buffett had 79% accuracy with the 3,000 ms duration and 65% accuracy with the 200 ms duration. The performance differences found between Hugh and Buffett are 57

PAGE 71

characteristic of the results found in previ ous sensory studies conducted with the same subjects (Bauer et al., 2003; 2005; Mann et al., 2005) and are assumed to represent normal variation (Ridgway & Ca rder, 1997; Brill et al., 2001). Videotape analysis of the test trials dem onstrated that the subjects remained at the stationing bar for a minimum of 270 ms before they began to move in response to the presentation of the test signa l. The 200 ms test signals, presented at 0.2-24 and 4 kHz, impeded head and/or body movements, requiri ng the subjects to navigate all 3.05 m to the test speakers without the presence of the test signal. Underw ater studies conducted with human divers have demonstrated that s ounds are easier to localize if the head is allowed to move during the sound presen tation (Wells & Ross, 1980) due to the accentuation of interaural cue differences (Thurlow et al., 1967; Richardson et al., 1995; Yost & Dye 1997). The longer 3,000 ms duratio n with the 0.2-24 kHz signal allowed the manatees to utilize interaural cues while tr aversing the ~ >2 m distance towards the sound source and likely accounts for the increased accuracy as compared to the 200 ms duration. The third hypothesis stated that subjects would have greater lo calization accuracy to the anterior 180o than to the posterior 180o. The data were separated into three regions (front, back and side) and the number corr ect for each of the broadband stimuli were found for each location (Table 2.5). This hypothesis was supported as results indicated that both subjects had gr eater localization accura cy to the anterior 180o than to the posterior 180o with the exception of Hughs higher side performance with the 0.2-24 kHz, 200 ms stimulus. 58

PAGE 72

Previous studies have sugge sted that the ability to localize a sound source may be influenced by multimodal sensory systems and be a function of visual orientation responses (Brown, 1994; Heffner, 1997). The s ubjects of this study were unable to see the test speakers located behind them while at station and even si gnals tested at the maximum 3,000 ms duration condition did not allow enough time for them to turn and see the speakers before the signal ceased. The increased number and dispersal of errors found when test signals originated from the 135o, 180o and 225o locations suggest that manatees utilize visual orienting resp onses to assist w ith localization. The fourth hypothesis asserted that subject errors would have a higher distribution to the correct locations neares t neighbors rather than to ot her locations as was found in a prior four choice localization study (Colbert, 2005). Results from this experiment show that errors are typically distri buted to the correct locations nearest neighbors for the front 180o for both subjects, however increased se lection confusion was found at the 135o location for Buffett and at the 180o and 225o locations for Hugh. The data derived from Buffett suggests that he is able to localize the region if not the source from which the signal originated, which suppor ts this hypothesis, but H ughs performance at the 180o and 225o locations suggest that he has difficulty determining sound source directionality directly behind and to the left posterior regions of his body. The final hypothesis contended that s ubjects would make more differentiation errors between speakers located at 0 o and 180 o than any other contralateral pairs. The data were separated into four contralateral pairs (0o & 180o; 45o & 225o; 90o & 270o; and 135o & 315o) and the number of confusion instances that occurred for each were calculated for each of the broadband stimuli co nditions (Table 2.6). Of the 360 trials run 59

PAGE 73

for each subject, only eleven contralateral confusions were made by Hugh and three by Buffett. This hypothesis was supported by Hughs performance as eight of these confusions were made between the 0o and 180o pair (all were made when test signals originated from 180o), while the remaining three were found spread between different pairs. This hypothesis was not supporte d by Buffetts performance however, who had only one confusion between the 0o and 180o pair, but two confusions between the 90o and 270o pair. Table 2.6. Number of confusions made between contralateral pairs for each subject. Numbers were derived from a possible 45 trials per pair when tested with the 0.2-24 kHz, 200 ms, 0.2-24 kHz, 3000 ms, 18-24 kHz, 3000 ms, and 0.2-1.5, 3000 ms stimuli. Hugh Buffett Signal @ 0; Selected 180 Signal @ 45; Selected 225 Signal @ 90; Selected 270 2 Signal @ 135; Selected 315 Signal @ 180; Selected 0 8 1 Signal @ 225; Selected 45 1 Signal @ 270; Selected 90 1 Signal @ 315; Selected 135 1 Middlebrooks and Green (1991) demons trated that front and back sound localization confusions were t ypical with human subjects and attributed these results to the fact that stimulus locations lie in mirro r symmetry with respect to the subjects ears which eliminate interaural time of arrival, phase and intensity cues. Out of the possible 90 trials that front to back confusions w ould have the opportunity to occur, results demonstrated Buffett had only one and Hugh ha d just eight. These results suggest that the manatees may have been able to use some t ype of interaural cue(s) to assist with these discriminations even though the stimuli were presented symmetrically. 60

PAGE 74

Almost all localization studies conducted with terrestrial and marine mammals utilize minimal audible angle techniques in which subjects identify a just detectable change of a sound source from a particular reference point (Mills, 1958). The design used in this study instead re quired subjects to locate sound sources relative to their own location. This testing paradigm, similar to t hose used recently with a harbor seal (Bodson et al., 2006) and a harbor porpoise (Kastelein et al., 2007), has more realistic applications which address the subjects ab ility to determine the direc tionality of sounds as they originate from different angl es surrounding their bodies. Experiments in controlled settings provide valuable information about the specific conditions tested. Results from this study demonstrate that the s ubjects could localize short and solitary test signals within the frequency ranges of r ecreational boat engines and conspecifics in all 360o of the azimuth plane at distan ces of at least 3 meters. Attenuation of the level showed that both subj ects were able to localize the test signal well above the 12.5% chance level (Hugh, 48%; Buffett, 56%) at 80 dB re 1 Pa, a relatively quiet level. Understanding how the manatees sensory systems assimilate information and react to environmental stimuli is an importa nt factor that should be considered in conservation management strate gies that are incorporated into the Florida Manatee Recovery Plan (US Fish and Wildlife Service, 2001). Implications derived from this controlled study suggest that manatees would be better able to localize sounds in their natural environment considering most stimuli ar e repetitive and/or of longer duration than the test signals. Natural sounds provide increased opportunities to alter head or body orientation to better utilize interaural cue differences. This study provides strong 61

PAGE 75

evidence that manatees are capable of localizing the sounds produced by boats and conspecifics. Their ability to successf ully localize sounds at the 110 dB re 1 Pa level as well as when it was attenuated implies that manatees are able to lo calize the loud sounds of nearby stimuli as well as the quieter sounds of distant stimuli. Norris (1967) suggested that marine mammals may even localize s ounds derived from abiotic sources (shore waves) to assist with navigation. Manatees may also utilize a uditory landmarks to facilitate their biannual migrations. Several areas of study should be consider ed for future investigations that would enhance our knowledge about the ability a nd means by which manatees are able to localize sounds. Localization tasks with manatees to date have only investigated their abilities within the azimuth plane. Field tests that measured manatee responses to controlled boater approaches found that manat ees increased swim speed and oriented to deeper channel waters as boats approached (Nowacek et al., 2004). Localization ability assessments in the remaining dimensions may fi nd that interaural cues in the vertical plane hold equal or more salience than those in the azimuth plane, partially explaining why these animals increase their depth in response to surface threats. Manatee localization investigations to date demonstrate their ability to determine sound source directionality in all 360o, including a capacity to interpret sounds originating directly to the fr ont and back of them. The means by which they accomplish these tasks however, remain unclear. Most terrestrial mammals utilize some combination of interaural time, level, and phase differe nce cues to localize sounds, however several species have reduced or lost the ability to us e one or even all of them (Heffner & Heffner, 1992a). Head related transfer function measurements for si gnals presented in different 62

PAGE 76

locations on the azimuth plane may provide clues as to how interaural level and frequency differences might be used to facilitate sound localization. 63

PAGE 77

References Cited Ahnelt, P.K. & Bauer, G.B. (2000). Unpublished data reported in Ahnelt, P. K. & Kolb, H. (2000). The mammalian photore ceptor mosaic-adaptive design. Progress in Retinal and Eye Research 19, 711-777. Ahnelt, P.K. & Kolb, H. (2000). The mamm alian photoreceptor mosaic-adaptive design. Progress in Retinal and Eye Research 19, 711-777. Anderson, S. (1970). Directional hearing in th e harbor porpoise, Phocoena phocoena. In G. Pilleri (Ed.), Investigations on Cetacea, Vol II. Berne, Benteli Ag. Au, W. (1993). The Sonar of Dolphins NY: Springer-Verlag. Babushina, E.S. & Poliakov, M.A. (2004). The underwater and airbor ne sound horizontal localization by the northern fur seal. Biophysics 49, 723. Bachteler, D. & Dehnhardt, G. (1999). Ac tive touch performance in the Antillean manatee: Evidence for a functional differen tiation of the facial tactile hairs. Zoology, 102, 61-69. Barrett-Lennard, L.G., Ford, J.K.B. & Heise, K. A. (1996). The mixed blessing of echolocation: Differences in sonar use by fish-eating and mammal eating killer whales. Animal Behavior 51, 553. Bartsh, S.S., Johnston, S.D. & Siniff, D.B. (1992). Territorial be havior and breeding frequency of male Weddell seals, Leptonychotes weddelli in relation to age, size, and concentrations of serum testosterone and cortisol. Canadian Journal of Zoology 70, 680. Bauer, G.B., Colbert, D.E., Gaspard, J.C., Littlefield, B. & Fellner, W. (2003). Underwater visual acuity of Florida manatees ( Trichechus manatus latirostris.). International Journal of Comparative Psychology 16, 130-142. Bauer, G.B., Gaspard III, J.C., Colbert, D.E., Leach, J.B. & Reep, R. (2005, March). Tactile Discrimination of Textures by Florida Manatees, Trichechus manatus latirostris. Paper presented at the 12th annua l International Conference on Comparative Cognition, Melbourne, Florida. 64

PAGE 78

Bjrgesaeter, A., Ugland, K.I. & Bjrge, A. (2004). Geographic variation and acoustic structure of the underwater vocalization of harbor seal, Phoca vitulina, in Norway, Sweden and Scotland. Journal of the Acoustical Society of America 116, 2459. Bodson. A., Miersch, L., Mauck, B. & Dehnhard t, G. (2006). Underwater auditory localization by a swim ming harbor seal ( Phoca vitulina ). Journal of the Acoustical Society of America 120 (3) 1550-1557. Brill, R.L., Moore, W.B. & Dankiewicz, L.A. (2001). Assessment of dolphin ( Tursiops truncatus ) auditory sensitivity and hearing loss using jawphones. Journal of the Acoustical Society of America 109 (4): 1717-1722 Brown, C.H. (1994). Sound Localization. In R. R. Fay and A. N. Popper (Eds.) Comparative Hearing: Mammals New York, Springer-Verlag, pp. 57-97. Brown, C.H., Beecher, M.D., Moody, D.B. & St ebbins, W.C. (1980). Localization of noise bands by old world monkeys. Journal of the Acoustical Society of America 68, 127-132. Brown, C.H. & May, B.J. (1990). Sound localiza tion and binaural processes. In M. A. Berkley and W. C. Stebbins (Eds.) Comparative Perception Vol I, New York, John Wiley and Sons, pp. 247-284. Buckstaff, K.C. (2004). Effects of watercraft noise on the acoustic behavior of bottlenose dolphins, Tursiops truncatus in Sarasota Bay Florida. Marine Mammal Science 20, 709-725. Bullock, T.H., Domning, D.P. & Best, R.C. (1980). Evoked potentials demonstrate hearing in a manatee ( Trichechus inunguis ). Journal of Mammalogy 61 (1), 130133. Bullock, T.H., OShea, T.J. & McClune, M.C. (1982). Auditory evoked potentials in the West Indian manatee (Sirenia: Trichechus manatus). Journal of Comparative Physiology, 148, 547-554. Caldwell, M.C. & Caldwell, D.K. (1965). Individualized whistle contours in bottlenosed dolphins ( Tursiops truncatus ) Nature, 207, 434 435 Casseday, J.H. & Neff, W.D. (1973). Localization of pure tones. Journal of Acoustical Society of America 54, 365-372. Cleator, H.J. & Stirling, I. (1990). Wint er distribution of bearded seals, Erignathus barbatus, in the Penny Strait Area, Northw est Territories, as determined by underwater vocalizations. Canadian Journal of Fish and Aquatic Science 47, 1071. 65

PAGE 79

Cohen, J.L., Tucker, G.S. & Odell, D.K. ( 1982). The photoreceptors of the West Indian manatee. Journal of Morphology, 173, 197-202. Colbert, D.E., Fellner, W., Bauer, G. B., Mani re, C. & Rhinehart, H. (2001). Husbandry and research training of two Florida manatees, Trichechus manatus latirostris. Aquatic Mammals 27 (1), 16-23. Colbert, D. (2005). Sound Localization Abilities of Two Florida Manatees, Trichechus manatus latirostris Unpublished Masters Thesis, University of South Florida, Tampa, FL. Deecke, V.B., Slater, P.J.B. & Ford, J.K.B. (2002).Selective habitu ation shapes acoustic predator recognition in harbour seals. Nature, 420, 171. Domjan, M. (1998). The principles of l earning and behavior. Pacific Grove, CA: Brooks-Cole Publishing Company. Don, M. & Starr, A. (1972). Lateralizat ion performance of the squirrel monkey ( Samiri sciureus ) to binaural click signals. Journal of Neurophysiology 35, 493-500. Florida Fish & Wildlife Res earch Institute (2006). Annual mortality rates http://research.myfwc.com/featur es/view_article.asp?id=12084 Gannon, D.P., Barros, N.B., Nowacek, D.P., R ead, A.J., Waples, D.M. & Wells, R.S. (2005). Prey detection by bottlenose dolphins, Tursiops truncatus : an experimental test of the passive listening hypothesis. Animal Behavior, 69, 709 720. Gentry, R.L. (1967). Underwater auditory localization in the California seal lion ( Zalophus californianus ), Journal of Auditory Research 7, 187-193 Gerstein, E. (1999). Psychoacoustic Evaluations of the West Indian manatee (Trichechus manatus latirostris) Unpublished Doctoral Di ssertation, Florida Atlantic University, Boca Raton, FL. Gerstein, E., Gerstein, L., Forsythe, S. & Bl ue, J. (1999). The unde rwater audiogram of the West Indian manatee (Trichechus manatus ). Journal of the Acoustical Society of America 105, 3575-3583. Gerstein, E.R. (2002). Manatees, bioacoustics and boats. American Scientist 90, 154163. Griebel, U. & Schmid, A. (1996) Color vision in the manatee ( Trichechus manatus). Vision Research, 36, 2747-2757. 66

PAGE 80

Griebel, U. & Schmid, A. ( 1997). Brightness discrimination ab ility in the West Indian manatee ( Trichechus manatus). Journal of Experimental Biology, 200, 15871592. Hanggi, E.B. & Schusterman, R.J. (1994). Unde rwater acoustic displa ys and individual variation in male harbour seals, Phoca vitulina. Animal Behavior 48, 1275. Hayes, S.A., Kumar, A., Daniel, P.C., Melli nger, D.K., Harvey, J.T., Southall, B.L. & LeBoeuf, B.J. (2004). Evaluating the function of the male harbour seal, Phoca vitulina roar through playback experiments. Animal Behavior, 67, 1133. Heffner, R.S. (1997). Comparative study of sound localization and its anatomical correlates in mammals. Acta Otolaryngologica 532, 46-53. Heffner, R.S. & Heffner, E.H. (1982). Hearing in the elephant ( Elephas maximus): Absolute sensitivity, frequency disc rimination, and sound localization. Journal of Comparative and Physiological Psychology 96 (6), 926-944. Heffner, R.S. & Heffner, E.H. (1984). Sound localization in large mammals: Localization of complex sounds by horses. Behavioral Neuroscience 98 (3), 541555. Heffner, R.S. & Heffner, E.H. (1985). S ound localization in wild Norway rats ( Rattus Norvegicus). Hearing Research 19, 151-155. Heffner, R.S. & Heffner, E.H. (1988a). Sound localization and use of binaural cues by the gerbil ( Meriones unguiculatus ). Behavioral Neuroscience 102 (3), 422-428. Heffner, R.S. & Heffner, E.H. (1988b). Sound lo calization acuity in the cat: Effect of azimuth, signal duration and test procedure. Hearing Research 36, 221-232. Heffner, R.S. & Heffner, E.H. (1988c). S ound localization in a predatory rodent, the Northern Grasshopper mouse ( Onychomys leucogaster ). Journal of Comparative Psychology, 102 (1), 66-71. Heffner, R.S. & Heffner, E.H. (1989). Sound lo calization, use of binaural cues and the superior olivary complex in pigs. Brain Behavior Evolution 33, 248-258. Heffner, R.S. & Heffner, E.H. (1992a). Evolut ion of sound localizati on in mammals. In, D. Webster, R. Fay and A. Popper (Eds.) The Biology of Hearing SpringerVerlag, pp. 691-715. Heffner, R.S. & Heffner, E.H. (1992b). H earing in large mammals: Sound localization acuity in cattle (Bos taurus) and goats (Capra hircus ). Journal of Comparative Psychology, 106 (2), 107-113. 67

PAGE 81

Heffner, R.S. & Masterson, R.B. (1990). Sound localization in mammals: Brain stem mechanisms. In M. A. Berkley, & W. C. Stebbins (Eds.) Comparative Perception New York, John Wiley and Sons, pp. 285-314. Heffner, R.S. (1997). Comparative study of sound localization and its anatomical correlates in mammals. Acta Otolaryngologica 532, 46-53. Holt, M.M., Schusterman, R.J., Sout hall, B.L. & Kastak, D. (2004). Localization of aerial broadband noise by pinnipeds Journal of the Acoustical Society of America, 115, 2339-2345. Houben, D. & Gourevitch, G. (1979). A uditory lateralization in monkeys: An examination of two cues serving directional hearing. Journal of Acoustical Society of America, 66, 1057-1063. Isley, T.E. & Gysel, L.W. (1975). S ound source localizatio n by the red fox. Journal of Mammalogy 56, 397-404. Kastelein, R.A., de Haan, D. & Verboom, W.C. (2007). The influence of signal parameters on the sound source localizat ion ability of a harbor porpoise ( Phocena phocena ). Journal of the Acoustical Society of America 122 (2), 1238-1248. Ketten, D.R., Odell, D.K. & Domning, D.P. ( 1992). Structure, function and adaptation of the manatee ear. In J. A. Thomas, R. A. Kastelein, & A. Y. Supin (Eds.) Marine Mammal Sensory Systems (pp.77-79). NY: Plenum Press. Kirkpatrick, B., Colbert, D.E., Dalpra, D., Ne wton, E.A.C., Gaspard, J., Littlefield B. & Manire, C.A. (2002). Florida Red Tides, Manatee Brevetoxicosis, and Lung Models. In K. A. Steidinger, J. H. La ndsberg, C. R. Tomas, and G. A. Vargo (Eds.) Xth International Conference on Harmful Algae Florida Fish and Wildlife Conservation Commission and Intergove rnmental Oceanographic Commission of UNESCO. Klishen, V.O., Diaz, R.P., Popov, V.V. & Supin, A.Y. (1990). Some characteristics of hearing of the Brazilian manatee, Trichechus inunguis. Aquatic Mammals 16 (3), 139-144. Manire, C.A., Walsh, C.J., Rhinehart, H.L., Colbert, D.E., Noyes, D.R. & Luer, C.A. (2003). Alterations in blood and urine parameters in two Florida manatees, Trichechus manatus latirostris from simulated conditi ons of release following rehabilitation. Zoo Biology 22, 103-120. Mann, D., Colbert, D.E., Gaspard, J.C. III, Casper, B., Cook, M.L.H., Reep, R.L. & Bauer, G.B. (2005). Temporal resolution of the Florida manatee ( Trichechus manatus latirostris ) auditory system. Journal of Comparative Physiology 191, 903-908. 68

PAGE 82

Marler, P. (1955). Some charac teristics of animals calls. Nature, 176, 6, 6-8. Mass, A.M., Odell, D.K., Ketten, D.R. & Supin, A.Y. (1997). Ganglion layer topography and retinal resolution of the Caribbean manatee Trichechus manatus latirostris Doklady Biological Sciences, 355, 392-394. Masterson, B., Heffner, H. & Ravizza, R. (1969). The evolution of human hearing. Journal of the Acoustical Society of America 45, 966-985. Masterson, B., Thompson, G.C., Bechtol d, J.K. & RoBards, M.J. (1975). Neuroanatomical basis of binaural phase-difference analysis for sound localization: A comparative study. Journal of Comparative Physiological Psychology, 89, 379-386. Medwin, H. & Clay, C.S. (1998). Fundamentals of Ac oustical Oceanography NY: Academic Press. Middlebrooks, J.C. & Green, D.M. (1991). Sound localization by human listeners. Annual Review of Psychology, 42, 135-159. Miksis-Olds, J.L. (2006). Manatee Respons e to Environmental Noise. Unpublished Doctoral Dissertation. University of Rhode Island, Narragansett, RI. Mills, A.W. (1958). On the minimum audible angle. Journal of the Acoustical Society of America 30, 127-246. Mills, A.W. (1972). Auditory loca lization. In J. V. Tobias (Ed.) Foundations of Modern Auditory Theory : Vol. II, (pp. 3030-348). NY: Academic Press. Moore, P.W.B. (1974). Underwater localiza tion of click and pulse d pure-tone signals by the California sea lion ( Zalophus californianus ). Journal of the Acoustical Society of America 57 (2), 406-410. Moore, P.W.B. & Au, W.L. (1975). Underwat er localization of pul sed pure tones by the California sea lion ( Zalophus californianus). Journal of the Acoustical Society of America, 58 (2), 721-727. Moore, P.W.B. & Brill, R.L. (2001). Binaural hearing in dolphins. The Journal of the Acoustical Society of America 109 (5), 2330-2331. Moore, P.W.B. and Pawloski, D.A. (1993). Interaural time discrimination in the bottlenose dolphin. The Journal of the Acoustical Society of America, 94 (3), 1829-1830. 69

PAGE 83

Norris, K.S. (1967). Some observations on th e migration and orientation of marine mammals. In: Animal Orientation and Navigation Proceedings of the 27th Annual Biology Colloquium, Oregon State., Corvallis, OR, pp. 101. Nowacek, D.P., Casper, B.M., Wells, R.W ., Nowacek, S.M. & Mann, D.A. (2003). Intraspecific and geographic variat ion of West Indian manatee ( Trichechus manatus spp.) vocalizations. Journal of the Acoustical Society of America 114 (1), 66-69. Nowacek, S.M., Wells, R.S., Owen, E.C.G., Speakman,T.R., Flam, R.O. & Nowacek, D.P. (2004). Florida manatees, Trichechus manatus latirostris respond to approaching vessels. Biological Conservation 119, 517-523. Odell, D.K. & Reynolds, J.E. III. (1979). Observations on manatee mortality in South Florida. Journal of Wildlife Management 28 (3), 572-577. OShea T.J. & Poche L.B. (2006). Aspect s of underwater sound communication in Florida manatees ( Trichechus manatus latirostris ). Journal of Mammalogy, 87 (6): 1061-1071. Pepper, R.L. & Defran, R.H. (1975). Dolphin Trainers Handbook, Part 1: Basic Training (50p). Naval Unde rsea Center, San Diego, CA. Piggins, D.J., Muntz, W.R.A. & Best, R.C. (1983). Physical and mo rphological aspects of the eye of the manatee, Trichechus inunguis Marine Behaviour and Physiology 9, 111-130. Popov, V.V. & Supin, A.Y. (1990). Electrophys iological studies on hearing in some cetaceans and a manatee. In J.A. Thomas, & R. A. Kastelein (Eds.), Sensory Abilities of Cetaceans: Laboratory and Field Evidence NY: Plenum Press. Richardson, W., Greene, C., Malme, C. & Thompson, D. (1995). Marine Mammals and Noise. San Diego, CA: Academic Press. Ridgway S.H. & Carder D.A. (1997). Hearing deficits measured in some Tursiops truncatus and discovery of a deaf/mute dolphin. Journal of the Acoustical Society of America 101: 590-594 Reep R.L. Marshall, C.D. & Stoll, M.L. (2002). Tactile hairs on the postcranial body in Florida manatees: a mammalian lateral line? Brain Behavior and Evolution 59, 141-154. Renaud, D.L. & Popper, A.N. (1975). S ound localization by the bottlenose porpoise Tursiops truncatus Journal of Experimental Biology 63, 569-585. Reynolds, J.E. III. (1979). The semisocial manatee. Natural History 88 (2), 44-53. 70

PAGE 84

Reynolds, J.E. & Odell, D.K. (1991). Manatees and Dugongs (192 pp.). NY: Checkmark Books. Reynolds, J.E. III, & Wilcox, J.R. (1986). Distribution and abundance of the West Indian manatee, Trichechus manatus, around selected Florida power plants following winter cold fronts: 1984-1985. Biological Conservation 38, 103-113. Richardson, W.J., Greene, C.R., Malme, C.I. & Thomson, D.H. (1995). Marine Mammals and Noise. Academic, San Diego. Rogers, T.L., Cato, D.H. & Bryden, M.M. (1996 ). Behavioral significance of underwater vocalizations of cap tive leopard seals, Hydrurga leptonyx. Marine Mammal Science 12 (3), 414. Ross, D. (1976). Mechanics of Underwater Noise. NY: Pergamon. Sayigh, L.S., Tyack, P.L., Wells, R.S. & Scott, M.D. (1990). Signature whistles of freeranging bottlenose dolphins Tursiops truncatus : stability and mother-offspring comparisons. Behavioral Ecology and Sociobiology, 26 (4), 247-260. Sayigh, L.S., Tyack, P.L., Wells, R.S., Scott, M.D. & Irvine, A.B. (1995). Sex difference in signature whistle production of free-ranging bottlenose dolphins, Tursiops truncatus. Behavioral Ec ology and Sociobiology, 36 (3), 171-177. Serrano, A. & Terhune, J.M. (2002). An ti-masking aspects of harp seal, Pagophilus groenlandicus, underwater vocalizations. Journal of the Acoustical Society of America 112, 3083. Stevens, S.S. & Newman, E.B. (1936). The localization of actual sources of sound. American Journal of Psychology. 48, 297-306. Smolker, R. & Pepper, J.W. (1999). Whistle Convergence among allied male bottlenose dolphins (Delphinidae, Tursiops sp.). Ethology, 105 (7), 595. Terhune, J.M. (1974). Directional hearing of a harbor seal in air and water. Journal of the Acoustical Society of America, 56 (6), 1862-1865. Thurlow, W.R., Mangels, J.W. & Runge, P.S. (1967). Head movements during sound localization. Journal of the Acoustical Society of America, 42, 489. Tyack, P.L. & Clark, C.W. (2000). Communica tion and acoustic behavior of dolphins and whales. In: Au, W.W.L. Popper, A.N. & Fay, R.R. (Eds.) Hearing by Whales and Dolphins Springer, New York, pp. 156. 71

PAGE 85

Urick, R.J. (1996). The Principles of Underwater Sound 3rd Edition (423 pp.). CA: Peninsula Publications. U.S. Fish and Wildlife Service (2001). Florida Manatee Recovery Plan, (Trichechus manatus latirostris), Third Revision. U.S. Fish and Wildlife Service, Atlanta, Georgia. 144 pp. Van Parijs, S.M., Hastie, G.D. & Thomps on, P.M. (1999). Geogra phic variation in temporal and spatial vocalization patterns of male harbour seals in the mating season. Animal Behavior 58, 1231. Van Parijs, S.M., Hastie, G.D. & Thompson, P.M. (2000 a). Individual and geographical variation in display behaviour of male harbour seals in Scotland. Animal Behavior, 59, 559. Van Parijs, S.M., Janik, V.M. & Thompson, P.M. (2000 b). Display area size, tenure length, and site fidelity in the aqua tically mating male harbour seal, Phoca vitulina. Canadian Journal of Zoology. 78, 2209. Van Parijs, S.M., Lyderson, C. & Kovacs, K.M. (2003). Vocalizations and movements suggest alternative mating tact ics in male bearded seals. Animal Behavior, 65, 273. Wakeford, O.S. & Robinson, D.E. (1974). Late ralization of tonal stimuli by the cat. Journal of the Acoustical Society of America 55, 649-652. Walls, G.L. (1963). The vertebrate eye and it s adaptive radiation NY: Hafner. Watkins, W.A. & Schevill, W.E. (1979). Dis tinctive characteristics of underwater calls of the harp seal, Phoca groenlandica, during the breeding season, Journal of the Acoustical Society of America 66, 983. Wells, M.J. & Ross, H. (1980). Distorti on and adaptation in underwater sound localization. Aviation, Space, and Environmental Medicine 51, 767. West, J.A., Sivak, J.G., Murphy, C.J. & K ovacs (1991). A comparative study of the anatomy of the iris and ciliary body in aquatic mammals. Canadian Journal of Zoology, 69, 2594-2607. Yost, W.A. (2000). Sound localizat ion and binaural hearing. In Fundamentals of Hearing: An Introduction. San Diego, CA: Academic Press (pp. 179-192). Yost, W.A. & Dye, R.H. (1997). Fund amentals of directional hearing, Seminars in Hearing, 18, 321. 72

PAGE 86

Chapter Three: Head/Body Related Transfer Functions of the Florida Manatee, Trichechus manatus latirostris Abstract Head and body related transfer function measurements were investigated for two Florida manatees ( Trichechus manatus latirostris) to determine how different frequencies of a test signal, presented in different locations on the azim uth plane, are filtered by the manatees head and torso. A previous invest igation demonstrated that manatees were capable of localizing sounds in all 360o of the azimuth plane and may be able to differentiate signals originating directly in front or behind them (C hapter 2). The means by which manatees determine sound source directionality however are unknown. To determine if different frequencies are filtered by the manatees head and torso, thereby providing level cues which may aid s ound localization, subjects were positioned in the center of a 360o array of speakers positioned 45o apart with one hydrophone suspended next to but not touching each exte rnal auditory meatus. The test stimulus presented was a 0.2-30 kHz, 3000 ms broadband noise burst. Head/body related transfer functions were determined by subtracting the averaged animal present FFTs (10 Hz frequency reso lution) from the aver aged animal absent FFTs (10 Hz frequency resolution). The magn itude of interaural level differences was 73

PAGE 87

then derived for all frequencies in addition to specific 0.2-1.5, 0.2-5 and 18-30 kHz bands of frequencies. Results indicated that intera ural level differences were found for all frequencies, starting below 1 kHz and extending up to 30 kHz, as a function of source location. Interaural level differences were of the gr eatest magnitude with frequencies above 18 kHz which have wavelengths shorter than th e manatees intercochlear distance. Test signals originating at 90o and 270o provided greater ILD cues than those originating from other locations, however ILDs we re greater when the signal originated behind the subject at 180o, 225o and 135o than in front of them at 0o, 45o and 315o. These results suggest that the manatees torso provided greater shadowing eff ects than the head, thereby increasing ILD cue salience to facilitate lo calization when sounds originate behind them. 74

PAGE 88

Introduction The Florida manatee ( Trichechus manatus latirostris ) is an endangered species, protected by both the Marine Mammal Protect ion Act (1972) and the Endangered Species Act (1973). It is the only marine mammal known to migrate considerable distances from fresh water habitats to salt water habitats in the summe r months, and the opposite in winter months. It is considered a semi-s ocial species, often grazi ng or traveling alone with conspecifics out of visual range, although females with calves will congregate together and males will mass around estrous females (Reynolds, 1979). The Florida manatee also lives in lives in a habitat where boats are found in high numbers with over 1,027,000 registered in the state of Florida in 2007 (Florida Department of Highway Safety and Motor Vehicles, 2007). The mean s by which manatees are able to find one another, navigate and avoid dange r in their vast habitat is uncl ear. Research has not been published regarding the manatees gustatory and olfactory sensory systems, however anatomical and behavioral studies have provide d insight into the manatees visual, tactile, and auditory sensory processes. The manatee visual sensory system appears to be built for sensitivity in dim light conditions with the ability to differentiate brightness differences (Griebel & Schmid, 1997) and blues from greens (Cohen et al., 1982; Griebel & Schmid, 1996; Ahnelt & Kolb, 2000; Ahnelt & Bauer, 2000), but acuity is poor (Walls, 1963; Piggins et al., 1983; West et al., 1991; Mass et al., 1997; Bauer et al., 2003) and not useful for fine details. The tactile sensitivity of the manatees facial vibrissae is excellent and comparable to that 75

PAGE 89

of the human index finger (Bachteler & De hnhardt, 1999; Bauer et al., 2005), and the vibrissae hairs dispersed acro ss the torso are also sensitiv e and may act similarly to a fishs lateral line (Reep et al., 2002). The ma natees hearing threshol d is quite wide and ranges between 0.4-40 kHz (Bullock et al., 1980, 1982; Popov & Supin, 1990; Gerstein et al., 1999; Mann et al., 2005) although one i nvestigation estimates it reaches 60 kHz (Klishen et al., 1990). An audiogram demonstr ates that the range of best hearing lies between 10 kHz with maximum sensitivity at ~50 dB re: 1 Pa with 16 and 18 kHz, decreasing by ~20 dB re: 1 Pa per octave from 0.8 to 0.4 kHz and 40 dB re: 1 Pa per octave above 26 kHz (Gerstein et al., 1999). These hearing capabilities indicate that manatees have the capacity to detect cons pecific vocalizations which typically range between 2.5.9 kHz (Nowacek, et al., 2003) and boat engine noise which typically range between 0.01 kHz (Gerstein, 2002; Richardson et al., 1995). These results suggest that the manatees senses of vision and touch are probably designed to function with tasks in close proximity to its body It seems likely that the manatees auditory system plays a crucia l role with functioning in both nearby and distant scenarios and that the ability to localize or determin e sound source directionality would be of great importance for tasks such as navigation, finding c onspecifics and boat avoidance. Sound localization is the aud itory systems ability to process the frequency, level and phase of a sound and associate it with th e spatial location of that sounds source (Yost, 2000). The ability to localize sounds is considered a primar y source of selective pressure in the evolution of mammalian hear ing (Masterson et al., 1969) and is vital for many species ability to find food and conspeci fics while avoiding predation. Previous 76

PAGE 90

studies have suggested that the ability to localize a sound source may be influenced by multimodal sensory systems and be a function of visual orientation responses (Brown, 1994; Heffner, 1997). Numerous species, incl uding humans, gulls, ducks, cats, rats and guinea pigs possess a reflexive visual orientation response towards startling sounds at birth or shortly thereafter wh en their auditory systems b ecome functional (Brown, 1994). Many species, including cats, mice, rats, chinch illas, guinea pigs and horses, also possess a Preyer reflex, which is a distinctive movement of the pinna towards a sudden sound to assist with localization (Francis, 1979; Eh ret, 1983). These multi-modal arrangements are extremely beneficial for determining the lo cation of an acoustic s timulus when it is of a long enough duration to do so. However, th e ability to localize sounds that are of shorter durations and cannot be tracked or s canned using head, eye, or pinna movements provides obvious additional advantages. In our three dimensional world, sounds can be localized from the vertical, horizontal (azimuth) and dist ance dimensions by extracting information from the sounds temporal, phase and level cues with each of our two ears ( Middlebrooks & Green, 1991; Hartman, 1999). Interaural tim e differences (ITD), also known as time of arrival cues, compare the sounds time of arrival at each ear (Figure 3.2). Because the speed of sound is relatively constant, variat ions in frequency do not have an effect on the perception of interaural time differences. Interaural level differences (ILD) are inte rpreted when the sound is one level when it reaches the closest ear but due to the shadowing effect of the pinna, head, or body, is a lower level when it reaches the farthest ear (Figure 3.2). The level difference is dependent on sound wave lengths. Higher frequencies have shorter wavelengths causing 77

PAGE 91

a greater sound shadow. Sounds are generally perceived to be closer to the ear they arrive earliest and with the greatest amplitude. Interaural phase differences (IPD) are interpreted when the sound that arrives in the first ear is in one period of the frequency but is out of phase when it hits the second ear (Figure 3.1). The phase difference is also dependent on sound wave lengths. Figure 3.1. Interaural time (ITD), phase (IPD), and level (ILD) cues used for sound localization. Several divisional planes have been identi fied around an organism which facilitate dimension and cue integration (Figure 3.2). Th e elevation plane, also called the vertical plane, runs vertically thr ough the body, dividing the left a nd right sides, and provides information about a sounds location as it is positioned anywhere in a circumference above or below the body. The azimuth plane also called the horizontal plane, runs laterally around the body and provides inform ation about a sounds location as it is positioned anywhere in a circumference from the left to the right. The medial plane runs Sound Onset IPD Time ITD Left Ear Right Ear ILD 78

PAGE 92

vertically through the body, di viding the anterior and poste rior portions, and provides information about a sounds location as it is positioned anywhere in a circumference from front to back (Figure 3.1). Monaural ear signal attributes (tho se derived from only one ear) provide information about anterior and poste rior areas of the median plane as well as the elevation angle and distan ce of the sound source location. Interaural signal attributes (those derived from both ears) provide inform ation about lateral displacement of the sound source location. The combination of mona ural and interaural ear signal attributes provides angular information about a sounds distance and lo cation in the azimuth and elevation planes. Figure 3.2. Azimuth, elevation and medial planes us ed to integrate the vertical, horizontal and distance dimensions with a sounds temporal, phase and level cues. Behavioral testing of sound localization ab ilities has typically been investigated by measuring the species minimum audibl e angle (MAA) (Brown, 1994; Brown & May, 79

PAGE 93

1990). This method determines the smallest detectable angular difference between two sound source locations positioned in front of the subject in the azimuth plane (Mills, 1958). Numerous in-air aud itory MAA studies have been conducted with terrestrial mammals including humans (Stevens & Newman, 1936; Mills, 1972), monkeys (Don & Star, 1972; Houben & Gourevitch, 1979; Brown et al., 1980), the domestic cat (Casseday & Neff, 1973; Wakeford & Robinson, 1974; Heffner & Heffner, 1988b), red fox (Isley & Gysel, 1975), hedgehog (Masterson et al., 1975), el ephant, horse, Norway rat, pig, gerbil, Northern grasshopper mouse, pocket gopher, goat and cattle (Heffner & Heffner, 1982; 1984; 1985; 1988a; 1988c; 1989; Heffner & Masterson, 1990; Heffner & Heffner, 1992b respectively). Although less common, MAA measurements have been assessed for marine mammals including pinnipeds (Gentry, 1967; Anderson, 1970; Moore, 1974; Terhune, 1974; Moore & Au, 1975; Babushina and Poliakov, 2004; Holt et al., 2004) and cetaceans (Renaud & Popper, 1975; Moore & Pawloski, 1993; Moore & Brill, 2001; Branstetter et al., 2003; Bran stetter, 2005; Branstetter & Mercado, 2006; Branstetter et al., 2007). More recently, some pinniped and manatee sound localiza tion investigations have required subjects to identify sound sources relative to different locations surrounding the subjects body. This has been done by presenting sign als in the frontal 180 or complete 360 of the horizontal plan e surrounding a stationary subject (Kastelein et al., 2007; Gerstein, 1999; Colbert 2005; Chapter 2) or by having the subject swim along a half circle diameter and orient to wards a sound source when presented (Bodson et al., 2006). All three designs a ssess sound localization abilities, however the latter two 80

PAGE 94

have more realistic applications by addressing the subjects ability to localize sounds that originate from different angl es surrounding their bodies. Although these testing paradigms provide va luable information about a species localization abilities, they do not address the means by which they are able to accomplish these tasks. The first investigation that examined how the shape of the human head affects sounds within the azimuth plane was conducted by Lord Rayleigh (Strutt, 1907). Lord Rayleigh modeled the head as a ri gid sphere and measured how sound waves propagated around it. His early results provided considerable information about interaural level and time differences. He found that ILDs were not linear with frequency. Frequencies with wavelengths greater than the diameter of the head (<1500 Hz) were not filtered or shadowed as much as frequencies with wavelengths smaller than the diameter of the head (>1500 Hz). These data suggested th at higher frequency components of a sound were more salient th an lower frequency components when the brain evaluated ILDs. He also found that lower frequency com ponents of a sound were more important for evaluating ITDs because independent comparisons of points within one phase of the sound wave could be interpreted by the brain. The spherical head model has been used by many researchers to explain how sounds from various locations within the azimuth dimension are filtered by the human head (Hartley & Fry, 1921; Kuhn, 1977, 1987; Brungart & Rabinowitz, 1999). It did not, however address how sounds originating from several different locations can produce identical ITDs and ILDs to create a cone of confusion (Figure 3.3), how sounds are filtered in the remaining dimensions or how sounds are filtered by the by the pinna or torso. 81

PAGE 95

a d c b Figure 3.3 Cone of confusion caused from sounds originating in different locations. Sounds from sources a & b as well as c & d have identical interaural time and level differences. Head-related transfer functions (HRTFs) have been comprehensively studied to understand the mechanisms of spatial hear ing (Blauert, 1997; Wightman & Kistler, 1997). HRTFs are determined by identifying differences between the sounds characteristics at its source and at the poi nt of the ears as a function of frequency (Blauert, 1997). HRTFs with terrestrial animals have commonly been conducted by playing bursts of broadband noise from different spatial locations surrounding a fixed head and measuring the sounds spectral ch aracteristics from small microphones that were implanted deep in the ear canal. HRTFs illustrate how sound waves are filt ered by the diffraction and reflection properties of the head, pinna, and torso before they reach the inner ear ( Searle et al., 1975; Middlebrooks et al., 1989) and how the conundrum of the cone of confusion is resolved. For species with pinnae, sounds may travel directly into the ear canal or be reflected off the pinna and travel into the ear canal fractions of a second later. Because sounds are typically composed of multiple fr equencies, many copies of the signal enter the ear at different times depending on their frequency. Some copies overlap and have 82

PAGE 96

matching phase signals that are enhanced, wh ile others have non-ma tching phase signals that are canceled out. Studies with humans have demonstrated that the pinna has substantial effects on HRTFs at higher freque ncies with wavelengths smaller than the pinna size (>3 Hz) but that these effects were minimized with lower frequencies (Mehrgardt & Mellert, 1977; Wightman & Ki stler, 1989). The pinna was also found to provide substantial information about the el evation of the sound source (Batteau, 1967; Wright et al., 1974). Similarly, the huma n torso influenced HRTFs, although not as significantly as the pinna, and primarily w ith lower frequencies (Kuhn & Gurnsey, 1983; Kuhn, 1987). Algazi et al. (2001; 2002) demonstr ated that the torso produced reflections and shadows that also provided elevation cues. Measurements of how interaural time phase and level difference cues are interpreted has been investigated with many terre strial species including humans ( Stevens & Newman, 1936; Mills, 1972; Middlebrooks & Green, 1991), rats (Heffner & Heffner, 1985), Northern grasshopper mous e (Heffner & Heffner, 1988c), gerbils ( Kelly & Potas, 1986; Heffner & Heffner, 1988a; Maki & Fu rukawa, 2005), guinea pigs (Carlile & Pettigrew, 1987), pocket gopher (Heffner & He ffner, 1992a), ferrets (Carlile, 1990), hedgehog (Masterson et al., 1975), Tammar wallabies (Coles & Guppy, 1986), monkeys (Don & Star, 1972; Houben & Gourevitch, 1979; Br own et al., 1980; Spezio et al., 2000), cats (Casseday & Neff, 1973; Wakeford & R obinson, 1974; Roth et al., 1980; Phillips et al., 1982; Irvine, 1987; Heffner & Heffner, 1988b; Musicant et al., 1990; Rice et al., 1992), fox (Isley & Gysel, 1975), elephant (H effner & Heffner, 1982), horse (Heffner & Heffner, 1984), pig (Heffner & Heffner, 1989) goat and cattle (Heffner & Heffner, 83

PAGE 97

1992b), bats (Jen & Chen, 1988; Obrist et al ., 1993; Fuzessery, 1996; Firzlaff & Schuller, 2003), and barn owls (Knudsen & Konishi, 1979 ; Moiseff, 1989; Keller et al., 1998). Results from these investigation indicate that most terrestrial mammals, including humans, gerbils, squirrel monkeys, Norway ra ts, macaques, red fox, and the domestic cat utilize some combination of all three intera ural cues. Some sp ecies, however, only use two interaural cues. For instance, the hedgehog and the Northern grasshopper mouse use only interaural time and level difference cues and the elephant, horse, pig, goat and cattle use only interaural time and phase differen ce cues. At least one species, the pocket gopher, is incapable of using any of the interaural cues and it has been suggested that this may be a result of this fossorial species adaptation to living in an underground environment where azimuth cues have li ttle meaning (Burda et al., 1990). While the ability to interpret interaural cues for localization may be difficult or impossible for some terrestrial mammals, the ability for marine mammals to use these cues for underwater localization is complicated by several factors. The speed of sound in water (1500 m/second) is approximately five times faster than in air (340m/second) (Urick, 1996) requiring marine mammal audito ry systems to process interaural time, phase and level differences much more rapi dly than those of terrestrial mammals. Although acoustic energy propaga tes more efficiently in water than light, thermal or electromagnetic energy (Au, 1993), higher fr equencies become more directional, reflecting off the surface and bottom and lo w frequencies may not propagate well in shallow waters (Medwin & Clay, 1998). It is apparent that sound characteristics differ between the source and inner ear due to attenuation from refraction, reflecti on, scattering and absorption caused by objects 84

PAGE 98

in the environment as well as by the shape of the pinna, head and torso of terrestrial mammals (Urick, 1996). Fully aquatic ma rine mammals however, have become streamlined for hydrodynamic efficiency and l ack the very important pinna terrestrial mammals and even semi-aquatic marine mammals such as the sea li on use to facilitate localization. Investigations with dolphins have shown that they are very competent at localization tasks (Renaud & Popper, 1975; Moore & Pawloski, 1993; Moore & Brill, 2001; Branstetter et al., 2003; Branstetter, 2005; Br anstetter & Mercado, 2006; Branstetter et al., 2007), but only one study has measured thei r interaural time and level difference thresholds. Moore et al. (1995) investigated the dolphins abilit y to utilize ITDs and ILDs by presenting binaural stimuli through jaw phones (hydrophones embedded in rubber suction cups) that were a ttached to the right and left lower jaws. Results found ITDs and ILDs were salient cues dolphins could dete ct and suggest that they likely use the same interaural differen tial sound cues as terres trial mammals. It has been shown that dolphins receive sonar ec hoes through complex fat channels in their lower jaw which may function as a pinna anal ogue (Brill, 1988; Ke tten et al., 1992; Mhl et al., 1999). The Florida manatee spends a significant amount of time grazing in shallow water where sounds tend to have more reflection off the surface of the water and bottom terrain making localization more challenging. The localization abilities of the manatee have only recently been investigated and results i ndicate that they are quite proficient at localizing sounds over a wide range of frequencies (Gerst ein, 1999; Colbert, 2005) and within all 360o (Chapter 2). These findings are somewhat perplexing given that the manatee lacks pinnae, but also possesses an extern al auditory meatus (e ar canal) that is of 85

PAGE 99

minute size, completely occluded with cellula r debris, and reaches a blind end that is separated from the tympanic memb rane (Chapla et al., 2007). The means by which manatees determin e sound source directionality have not been investigated as yet. It may be that the manatees elliptical and rotund body shape plays a more important role as a filter for ge nerating interaural level cues than its much smaller head. The objective of this study was to measure head/body related transfer functions (HBRTF) from tw o Florida manatees. 86

PAGE 100

Hypotheses Two hypotheses were made. The first posit ed that subjects would have greater sound shadows present when the test stimulus was presented at the 135o and 225o locations than when presented at other lo cations. The manatees small head and elongated elliptical body is fashioned in such a way that in the azimuth plane, there is more surface area for the signa l to reflect off of when sounds are presented to the posterior angles of th e body (Figure 3.4). Figure 3.4. Interpretation of how signals presented from the 135o and 225o locations reflect off the manatees elliptically shaped body. The second hypothesis declared that interaur al differences in level cues would be greater with higher frequencies than lower frequencies. ILDs are found when the level of the sound wave that reaches the ear nearest th e source is greater than when it reaches the ear farthest from the source. ILDs are most effective with shorter wavelengths (higher frequencies), especially those that are shor ter than the species inter-meatal distance (Brown & May, 1990; Brown, 1994; Blauert, 1997). 87

PAGE 101

Materials and Methods Subjects The subjects of this study were tw o captive-born male Florida manatees ( Trichechus manatus latirostris ), Hugh and Buffett, that reside at Mote Marine Laboratory and Aquarium in Sarasota, Florid a. All procedures used were permitted through the United States Fish and Wildlife Service (Permit # MA837923-6) and approved by the Institutional Animal Ca re and Use Committee of Mote Marine Laboratory and Aquarium. At the inception of this study Hugh was 23 years of age, weighed 547 kg, and was 310 cm in length, while Buffett was 20 years of age, weighed 773 kg, and was 334 cm in length. They were housed in a 265,000 lit er exhibit that was composed of three inter-connected sections : a 3.6 x 4.5 x 1.5 m Medical Pool, a 4.3 x 4.9 x 1.5 m Shelf Area, and a 9.1 x 9.1 x 3 m Exhibit Area. Both animals had acquired an extensive tr aining history over the previous seven years and had been behaviorally conditioned for husbandry procedures (Colbert et al., 2001) and studies which investigated lung cap acity (Kirkpatrick et al., 2002), serum and urine creatinine levels as a function of rel ease conditions (Manire et al., 2003), visual acuity (Bauer et al., 2003), faci al vibrissae tactile sensitivit y (Bauer et al., 2005), auditory evoked potentials (Mann et al., 2005) and f our-choice (Colbert, 2005) and eight-choice sound localization studies (Chapter 2). 88

PAGE 102

Experimental Design An eight alternative forced-choice discrimination paradigm was used to measure the head/body related tran sfer functions (HBRTFs) of two Florida manatees, Trichechus manatus latirostris Testing was conducted in the center of the exhibit area of the manatee habitat where eight underwater speake rs (Aquasonic AC 339) were positioned in a 6.10 m diameter circumference at 0o, 45o, 90o, 135o, 180o, 225o, 270o and 315o (Figure 3.5). Each speaker was suspended from a 1.88 cm diameter PVC rod at a depth of 1.5 m. The rods were bolted to aluminum beams that radiated out from two suspension supports spanning the Exhibit Area, and were designed to pivot so th at the speaker at the bottom of the rod could be pushed backwards while the top of the rod tilted forward in a pendulum motion. 3.05 m 270o 315o 225o 180o 45 o90o 135o 0o 4.5 x 4.9 x 1.5 Shelf Area 3.6 x 4.5 x 1.5 m Medical Pool 9.1 x 9.1 x 3 m Exhibit Area Figure 3.5. Testing setup for the manatee body related transfer function experiment. Subjects stationed facing 0o and test speakers were suspended from pivoting rods at 45o, 90o, 135o, 180o, 225o, 270o and 315o. The blue octagon represents the Test Trainers locatio n, the green square represents the Data Recorders location, and the orange triangle represents the Stationing Trainers location. A 23 cm x 1.5 m stationing apparatus wa s constructed from 2.54 cm diameter polyvinyl chloride (PVC) pipe and positioned in the center of the circular array, 3.05 m 89

PAGE 103

away from the speakers. Each subject had b een previously trained to position the top of its rostrum, approximately 10 cm posterior to the nostrils, up against a stationing bar located at the bottom of the stationing apparatus in res ponse to an individualized stationing tone. For this experiment, the st ationing apparatus was modified such that two hydrophones (HTI 96 min; sensitivity -164 dBV/ Pa from 0.2 Hz to 37 kHz) were suspended next to but not touching each of th e subjects external auditory meatus (Figure 3.6). Figure 3.6. Stationing apparatus used to measure manatee body related transfer functions. The black circle represents the speaker that played the stationing tones. The subject pressed the crease of his rostrum up against the gray stationing bar on the bottom. Two hydrophones were suspended next to but not touching each external auditory meatus. W at e r Lin e 1 .5 m Tr a in e r Pl at f o r m 2 3 c m H yd r op h o n e 90

PAGE 104

The subject remained stationed facing 0o until a 0.2-30 kHz, 3,000 ms test signal was played from one of the eight test speakers. U pon hearing the test signal, the subject would swim to and push the speaker from wh ich he believed the sound originated. If correct, a secondary reinforcer signal (1.4-12 kHz with a peak at 5.3 kHz for Buffett, 1.211 kHz with a peak at 2.7 kHz for Hugh) wa s emitted from the test speaker and the subject returned to the stationing device to be fed a primary reinforcement of food (apples, beets, and carrots). If incorrect the stationing tone was played from the stationing apparatus speaker and the subject re -stationed correctly with no reinforcement given to await a minimum of 30 seconds before the initiation of the next trial. Three people were required to run the expe riment: a Test Trainer, Data Recorder, and Station Trainer (Figure 3.5). The Test Trainer wa s positioned on a platform suspended across the Exhibit Area and ensu red that the subject stationed properly, initiated trials, indicated which speaker the subject selected, and provided reinforcement when the subject selected the correct sp eaker location. The Data Recorder was positioned behind a laptop computer out of sight of the Test Trainer and subject, and set up the sessions experimental conditions, informed the Test Trainer if the subject was correct or incorrect, recorded all data on a tank-side sessi on sheet (see Appendix C), and ran the video equipment. The Station Trai ner was positioned at northeast end of the Medical Pool and was responsible for holdi ng the non-test animal at station throughout the subjects session. Although the personnel an d experimental design protocols put in place to avoid cuing the subject s in the eight choice localiz ation study were replicated, the subjects speaker selection choices were not the topic of this investigation and therefore not recorded. 91

PAGE 105

Testing sessions were run between 0700 and 1000 h. The manatees daily ration of food (72 heads of romaine lettuce and 12 bunches of kale) was fed to the animals from 1200 to 1400 h and was usually consumed by 1700 h, leaving a 14 to 16 hour overnight fast before training was initi ated the following morning. Signal Generation & Programming All signals including each subjects stat ion and secondary reinforcement signals as well as the 0.2-30 kHz, 3,000 ms test stim ulus were programmed in RPvds language, digitally generated by a Tucker-Davis Tec hnologies real-time pro cessor (RP2.1). The signal were amplified with a Hafler power amplifier and the test stimulus was switched to the eight test speakers during the presentation of the stationing tone through a power multiplexer (PM2R). Three separate digital to analog channels were used; one to generate the signal to the stationing speaker at the center of the array and two to record from the hydrophones. MATLAB programming was used to generate blocks of sixteen trials that were counterbalanced between th e eight speaker locations in a quasi-random order, meaning that the test signal loca tion was randomized, but had a criterion of no more than two trials in a row from the same location. Test trials were initiated and comple ted through an electr onic control box which was connected to the RP2 unit, and then in to a Dell laptop computer (model Latitude D505) with Windows XP. Test signals received by each hydrophone were digitally recorded to the real time processor. The laptop computer was used to run the signal generation equipment and to automatically download the parame ters of all hydrophone recordings into separate .wav files. All te st trials were visually recorded through a Sony variable zoom, high resolution, outdoor w eather proof, color dome camera (model SCW92

PAGE 106

CD358DVP) that was attached to the trainers platform directly over the subjects head and connected to a Sony digital video camera (model DCR-TRV50). Raw data were analyzed to remove any sounds of subject movement and only segments in which there was no extraneous noise were kept. Test tr ials were collected until the kept segments from each speaker provided a minimum of 3,000 ms of data. Fast Fourier transforms (FFTs) using 9,766 poi nts were used to convert the signal from the time domain to the frequency domain. The FFTs from each location were averaged together using a 10 Hz frequency resolution. All data analyses were programmed in MatLab (Appendix D). 93

PAGE 107

Results Testing was conducted from April 25-30, 2007. Both subjects readily adapted to the presence of the hydrophones on the stationi ng apparatus and additional conditioning was not needed for them to complete the lo calization task in their usual manner. Three data analyses were conducted per subject. The first compared power spectra received at the left and right hydrophones as a func tion of sound source location with animal absent and animal present conditions. HBRTFs were determined by subtracting the averaged animal present FFT s from the averaged animal absent FFTs. The second determined the magnitude of interaur al level differences for all frequencies. The final analysis determined the magnitude of interaural level differences for specific 0.2-1.5, 0.2-5, and 18-30 kHz bands of frequencies. Comparisons were made between the power spectra acquired at the left hydrophone to that acquired at the right hydropho ne (simulating the manatees left and right ears) when the signals originated from the eight different locations. These data were then compared to the same data colle cted in the absence of the subject at the stationing bar (Figures 3.8 and 3.9). Results demonstrated interaur al level differences for all frequencies ranging from 0.2 to 30 kHz (the output limits of the test speakers) as a function of source location. 94

PAGE 108

Hu g h Right Ear Left Ear Figure 3.8. Comparisons of left vs. right received signals as a function of signal source location and the presence (red line) or absence (blue line) with Hugh. The Y axis represents the amount of signal attenuation (dB) and the x-axis represents frequency (Hz) for each graph in the figure. 0o 315o 90o 270o 225o 180o 135o 45o Animal Present Si g nal Animal Absent Signal Right Ear Right Ear Right Ear Right Ear Right Ear Right Ear Left Ear Left Ear Left Ear Left Ear Left Ear Left Ear Right Ear Left EarAttenuation (dB) Fre q uenc y 95

PAGE 109

Buffett Right Ear Left Ear Figure 3.9. Comparisons of left vs. right received signals as a function of signal source location and the presence (red line) or absence (blue line) with Buffett. The Y axis represents the amount of signal attenuation (dB) and the x-axis represents frequency (Hz). Head/body related transfer functions were then derived from these data by subtracting the averaged animal present FFT s from the averaged animal absent FFTs received by each hydrophone as a function of sound source location. Figure 3.10 0o 315o 90o 270o 225o 180o 135o 45o Animal Present Si g nal Animal Absent Signal Right Ear Right Ear Left Ear Attenuation ( dB) Fre q uenc y Left Ear Left Ear Left Ear Right Ear Right Ear Right Ear Left Ear Left Ear Right Ear Right Ear Left Ear 96

PAGE 110

demonstrates how the test signal was atte nuated across frequenc ies by the subjects (Buffett) head and torso when the sound origin ated on the opposite side. The red lines of the plots represent the averaged animal ab sent minus the animal present signals received by the right hydrophone and the blue lines indicate the same for the left hydrophone. The y-axis corresponds to the amount of signal attenua tion and the x-axis denotes frequency. Results demonstrat ed that signals originating from 90o were attenuated by the subjects body across all frequencies by as much as 8 dB re 1 Pa when received at the left hydrophone. Li kewise, signals or iginating from 270o were attenuated when received at the right hydrophone. Figure 3.10. Left (270o) vs. right (90o) head/body related transfer functions for Buffett. Red lines represent animal absent averaged signals minus animal present signals for the right hydrophone and blue lines represent the same for the left hydrophone. The Y axis represents the amount of attenuation difference (dB) between ears and the x-ax is represents frequency (Hz). Ri g ht Ear ( Animal Absent -Animal Present Si g nals ) Left Ear (Animal Absent -Animal Present Signals) Attenuation Difference (dB) 270o 90o Buffett Frequency (Hz) Similar ILD results were found with both subjects (Figures 3.11 & 3.12). Shadowing effects from the manatee head and torso created signal differences that 97

PAGE 111

covered a broad frequency range starting at frequencies belo w 1 kHz and extending up to the 30 kHz output limits of the test speakers. Figure 3.11. Head/body related transfer functions for Hugh. Red lines represent animal absent averaged signals minus animal present signals for the right hydrophone and blue lines represent the same for the left hydrophone. The Y axis represents the amount of attenuation difference (dB) between ears and the xaxis represents frequency (Hz). 0o 315o 90o 270o 225o 180o 135o 45o Hu g hRi g ht Ear ( Animal Absent Minus Animal Present Si g nals ) Left Ear ( Animal Absent Minus Animal Present Si g nals ) Difference 98

PAGE 112

Buffett Figure 3.12. Head/body related transfer functions for Buffett. Red lines represent animal absent averaged signals minus animal present signals for the right hydrophone and blue lines represent the same for the left hydrophone. The Y axis represents the amount of attenuation difference (dB) between ears and the x-axis represents frequency (Hz). Right Ear (Animal Absent Minus Animal Present Signals) Left Ear (Animal Absent Minus Animal Present Signals) 0o 315o 90o 270o 225o 180o 135o 45o Difference 99

PAGE 113

The magnitude of the level differences (d B) found between the signals received at the left and right hydrophones as a function of frequency and sound source location were found by subtracting the animal absent averag ed signals minus animal present signals received from the right hydrophone from the animal absent averaged signals minus animal present signals for the right hydr ophone (Figures 3.13 & 3.14). Results showed that received signals had a large amount of variability in de cibel gain and loss depending upon the frequency and sound source location. 100

PAGE 114

Buffett 0o 45o315o 270o90o 225o135o 180o Figure 3.13. Interaural level difference magnitudes between the left and right hydrophones for all frequencies with Buffett. Animal absent averaged signals minus animal present signals for the right hydrophone were subtracted from th e animal absent averaged signals minus animal present signals for the right hydrophone. The ILD spectrum represents the difference in a decibel scale. The x-axis represents frequency (Hz) and the y-axis represen ts level gain or loss (dB). A positive ILD indicates level in the right ear was higher than in the left ear. 101

PAGE 115

Hugh 0o 45o315o 270o90o 225o135o 180o Figure 3.14. Interaural level difference magnitudes between the left and right hydrophones for all frequencies with Hugh. Animal absent averaged signals minus animal present signals for the right hydrophone were subtracted from th e animal absent averaged signals minus animal present signals for the right hydrophone. The ILD spectru m represents the difference in a decibel scale. The x-axis represents frequency (Hz) and the y-axis represen ts level gain or loss (dB). A positive ILD indicates level in the right ear was higher than in the left ear. The magnitude of interaural level differe nces (dB) as a function of sound source locations were also calculated for specific 0.2-1.5, 0.2-5 and 18-30 kHz bands of frequencies (Figure 3.15; Table 3.1). Pos itive ILDs, which indicated that level was higher in the right ear, were found when the te st signal originated to the right of the 102

PAGE 116

subjects and negative ILDs were found when it originated to their left. The 0.2-1.5 kHz frequency range was the same low frequency test stimuli used in the 8-choice manatee sound localization investigat ion (Chapter 2). The 0.25 kHz frequency range was composed of stimuli with wavelengths longer than the manatees in termeatal distance. The 18-30 kHz frequency range included stimul i that were shorter than the manatees intercochlear distance. Results for both subj ects demonstrated that ILD magnitudes were greatest with the higher 18-30 kHz freque ncy band, however the lowest 0.2-1.5 kHz frequency band had larger magnitudes than the 0.2-5 kHz frequency band (Figure 3.15). 103

PAGE 117

Figure 3.15. Interaural level difference magnitudes between the left and right hydrophones with 0.2-1.5, 0.2-5 and 18-30 kHz bands of frequencies. Animal absent averaged signals minus animal present signals for the right hydrophone were subtracted from the animal absent averaged signals minus animal present signals for the right hydrophone The x-axis represents sound source location and the y-axis represents level gain or loss (dB). A positive ILD indicat es level in the right ear was higher than in the left ear. Buffett-6.5 -6.0 -5.5 -5.0 -4.5 -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 0o45o90o135o180o225o270o315oSpeaker LocationGain or Loss (dB) 0.2-1.5 kHz 0.2-5 kHz 18-30 kHz Hugh -6.5 -6.0 -5.5 -5.0 -4.5 -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 0o45o90o135o180o225o270o315oSpeaker LocationGain or Loss (dB) 0.2-1.5 kHz 0.2-5 kHz 18-30 kHz 104

PAGE 118

ILD magnitudes were greatest when th e test signal originated at 90o and 270o, and were smallest when they originated at 0o. When these data were compared in contralateral pairs, ILDs were overwhelmingl y greater when signals originated in the back (Table 3.1). Although ILD magnitude wa s greatest when signals originated from the 90o and 270o pair, greater magnitudes between the frequency bands were equally distributed between these two locations. Table 3.1. Level differences (in dB) between subjects of averaged animal absent minus animal present signals for left and right hydrophones with 0.2-1.5, 0.2-5 and 18-30 kHz bands of frequencies. Data are presented in contralateral pairs with locations in the posterior 180o in italics. A positive ILD indicates level in the right ear was higher than in the left ear. Shaded areas represent the larger ILD of the pair. 0.2-1.5 kHz 0.2-5 kHz 18-30 kHz Hugh Buffett Hugh Buffett Hugh Buffett 0o 0.46 dB -0.22 dB 0.21 dB 0.46 dB 0.05 dB 0.79 dB 180o -0.04 dB -1.06 dB -0.37 dB -0.72 dB -4.00 dB -1.51 dB 45o 0.83 dB 0.24 dB 0.30 dB 0.57 dB 0.86 dB 2.26 dB 225o -3.61 dB -2.68 dB -2.01 dB -2.34 dB -4.50 dB -2.39 dB 315o -0.79 dB -0.76 dB -0.38 dB -0.32 dB -1.13 dB -2.80 dB 135o 2.36 dB 1.7 dB 1.02 dB 1.71 dB 2.41 dB 2.88 dB 90o 2.32 dB 2.54 dB 0.93 dB 1.61 dB 3.34 dB 4.07 dB 270o -1.54 dB -2.59 dB -0.08 dB -0.74 dB -6.14 dB -4.72 dB 105

PAGE 119

Discussion The results from this study provided information about the means by which manatees were able to generate interaural level difference cues via head and torso filtering effects. In order to capitalize on the previously learned behaviors of stationing in the center of a 6.10 m diameter circumfere nce with eight test speakers positioned at 45o apart (Chapter 2), the subj ects, Hugh and Buffett, comp leted localization trials, however, their speaker location selections we re not recorded and only head/body related transfer functions were measured. Due to the subjects experience at the localization task, testing was completed in only four sessions. Results indicated that leve l differences were found for all frequencies, starting below 1 kHz and extending up to the 30 kHz output limits of the test speakers, as a function of source location due to the shadowing effect of the subjects head and torso. This is remarkable given that a 1 kHz fre quency has a wavelength of 1.5 m in water. These findings demonstrate that ILDs are relevant cues which manatees may be able to detect and suggest that manatees, like the dolphin, likely use the same interaural differential sound cues as terrestrial mammals. The first hypothesis posited that subj ects would have greater sound shadows present when the test stimulus was presented at the 135o and 225o locations than when presented at other locations. Specific ILDs were determined for both subjects with 0.21.5, 0.2-5, and 18-30 kHz bands of frequencies, with positive ILDs indicating that the received level in the right ear was greater than the left (Table 3.1). Results showed test 106

PAGE 120

signals originating at 90o and 270o provided greater ILD cues than those originating from other locations, refuting this hypothesis. Inte restingly, ILDs were greater when the signal originated behind the subject at 180o, 225o and 135o than in front of them at 0o, 45o and 315o. Manatees have a unique body shape. Thei r head is small in comparison to the remainder of their large torso which is elong ated and elliptically-shaped. At the caudal end of their body is their paddle, or tail, whic h is laterally compresse d and less than 2 cm thick along its edge. During this study, Hugh weighed 547 kg and was 310 cm in length, while Buffett weighed 773 kg and was 334 cm in length. The circumference of Hughs head was 90 cm while the widest part of hi s torso, located at the umbilicus, was 204 cm. Buffetts head circumference was 101 cm and his torso was 237 cm. The ~ 2.3 : 1 body to head size ratio demonstrates that the to rso provides more surface area for sounds to reflect, refract, or scatter off of or be absorbed by than the head. When positioned horizontally in the azimuth plane, which is typi cal for this species, the shape of their torso provides more shadowing effects when sounds originate from the lateral and posterior angles of the body as compared to an terior angles (Figure 3.12). 107

PAGE 121

Figure 3.16. Simulated shadow effects created from the manatee head and body. Sounds originating from posterior angles of the body, such as shown from speaker (b) have more surface area to reflect off and create a larger shadow effect than th ose originating from anterior angles of the body such as shown from speaker (a). b a b a Previous studies have sugge sted that the ability to localize a sound source may be influenced by multimodal sensory systems and be a function of visual orientation responses (Brown, 1994; Heffner, 1997) or fo r those species possessing muscularized pinnae, a Preyer reflex (Francis, 1979; Ehre t, 1983). Investigations with terrestrial mammals have shown that the pinna provides elevation cues (Batteau, 1967; Wright et al., 1974) and has substantial e ffects on HRTFs with wavelengt hs smaller than the pinna size (Mehrgardt & Mellert, 1977; Wightman & Kistler, 1997). Studies with humans demonstrate that the pinna has a stronger effect on HRTFs than the torso (Kuhn & Gurnsey, 1983; Kuhn, 1987). Manatees, like dol phins, lack pinnae and therefore do not 108

PAGE 122

benefit from the spectral cues they provide. Dolphins however, unlike manatees, seem to have compensated for this deficit with their ability to receive sona r echoes through the fat channels located in their lower jaw. This mechanism likely functions as a pinna analogue (Brill, 1988; Ketten et al., 1992; Mhl et al ., 1999) and provides relevance for the ITDs and ILDs found by Moore et al. (1995). Investigations with manatees suggested that they were able to localize sounds over a wide frequency range including thos e in the frequency range of boats and conspecifics (Gerstein, 1999; Co lbert, 2005) and within all 360o (Chapter 2). Results from the 360o testing paradigm demonstrated that although subjects were able to localize from points behind them and made few front to back confusions, the number and dispersal of errors was greater than when test signals originated behind them. These results suggest that manatees ut ilize visual orienting responses to assist with localization from locations within their visual field increasing their accuracy, but still have compensated for the absence of these res ponses when sounds originated behind them through the use of amplified ILD cues produced by their body shape. The second hypothesis declared that IL Ds would be greater with higher frequencies than lower frequencies. Results found that ILDs were greater for the higher 18-30 kHz frequency range than the 0.2-1.5 or 0.2-5 kHz frequency ra nges (Figure 3.15; Table 3.1). These results sugge st that frequencies above 18 kHz, provide more salient cues for localization than t hose below it, supporting this hypo thesis. ILDs have been shown to be most effective with higher freque ncies, especially those that are shorter than the intermeatal distance fo r terrestrial species (Brown & May, 1990; Brown, 1994; Blauert, 1997) and the intercochlear distance for cetaceans (Dudok van Heel, 1962; 109

PAGE 123

Ketten et al., 1992). Ketten et al. (1992) found that the manatees average intermeatal distance was 0.278 m and intercochlear distan ce was 0.082 m. When both distances are considered, it appears that frequencies 5.4 kHz (wavelength = 0.78 m) or higher would provide more effective ILD cues when usi ng the intermeatal distance, and those 18 kHz (wavelength = 0.08 m) or higher would be more effective when using the intercochlear distance. ILD magnitudes were calculated with high frequency bands (18-30 kHz) having wavelengths shorter than the manatee s intercochlear distance and low frequency bands (0.2-5) having wavelengths longer than th eir intermeatal distan ces. Results from these analyses show that th e higher frequency band produced more effective ILD than the lower frequency and followed a pattern sim ilar to terrestrial mammals and cetaceans. Surprisingly, results found for the lowest 0.2-1.5 kHz frequency band used in the eightchoice manatee localization experiment (Cha pter 2) deviate from the typical mammalian pattern and demonstrate greater ILDs than th e wider range of low frequencies from 0.2-5 kHz. Since Florida manatees spend a signifi cant amount of time grazing in shallow water, interpretation of ILD cues from hi gher frequencies might be hindered due to attenuation from refraction, reflection, scatte ring and absorption caused by objects in the environment as well as the surface of the water and bottom terrain (Medwin & Clay, 1998). The combined effects of multiple reflective sound paths can sometimes be as loud as or louder than sound tr aveling directly from the source. The precedence effect (also called the Haas Effect or Law of th e First Wavefront) provides a solution for this problem however, by weighing the preceding so und more heavily than the reflection or echoes that arrive shortly ther eafter (Blauert, 1997). 110

PAGE 124

The information gained from this study demonstrates how different frequencies of a 0.2-30 kHz test signal, presented at 45o angles within the 360o of the azimuth plane, are filtered by the manatees body to provide ILD cues which may be used to facilitate sound localization. All measurements were obt ained when the subjects head and body were stationary, however in the natural environmen t, animals have the ability to move as sounds occur allowing the monaural and intera ural characteristics of the sound to change at the inner ear thereby reduc ing the cone of confusion a nd magnifying ILD cue strength (Blauert, 1997). These are the first head/ body related transfer function data collected for any Sirenian species and future investigations should be conducted to supplement this knowledge. Although ILDs are typically mo re effective with frequencies having wavelengths shorter than an animals intermeatal or intercochlear distance, the anomalous results found for the 0.2-1.5 kHz frequency band that pr oduced greater ILDs than the 0.2-5 kHz range warrants further research. Most HRTF investigations introduce sound sources that are at least 1 m away from the subject because HRTFs become independent of distance beyond this. As sound s ources originate at distances less than 1 m, ILDs increase dramatically and ITDs rema in constant. Investigations in which sound sources originate at a distance less than 1 m would be beneficial to determine if this pattern will hold true for mana tees. Manatees live in a habitat where acoustic stimuli may occur above and below them as often as around them. Field tests that measured manatee responses to controlled boater appr oaches found that manatees increased swim speed and oriented to deeper channel waters as boats approached (Nowacek et al., 2004). Algazi et al. (2002) compared HRTFs with acoustic measurements in the horizontal, 111

PAGE 125

median and frontal planes and demonstrated that reflections and shadows from the human torso provided important eleva tion cues. The study of mana tee HRTFs using test signals that originate in the eleva tion plane would provide valuable information about the salience of interaural difference cues in th is dimension. This information may provide insight into the importance of determini ng sound source directionality from dangerous sources such as boats at the surface. 112

PAGE 126

References Cited Ahnelt, P.K. & Bauer, G.B. (2000). Unpublis hed data reported in Ahnelt, P.K. & Kolb, H. (2000). The mammalian photore ceptor mosaic-adaptive design. Progress in Retinal and Eye Research 19, 711-777. Ahnelt, P.K. & Kolb, H. (2000). The mamm alian photoreceptor mosaic-adaptive design. Progress in Retinal and Eye Research 19, 711-777. Algazi, V.R., Avendano, C. & Duda, R.O. (2001). Elevation localization and headrelated transfer function anal ysis at low frequencies. Journal of the Acoustical Society of America 109, 1110. Algazi, V.R., Duda, R.O., Duraiswami R., Gumerov, N.A. & Tang, Z. (2002). Approximating the head-related transfer function using simple geometric models of the head and torso. Journal of the Acoustical Society of America 112 (5), 2053-2064. Anderson, S. (1970). Directional hearing in th e harbor porpoise, Phocoena phocoena. In G. Pilleri (Ed.), Investigations on Cetacea, Vol II. Berne, Benteli Ag. Au, W. (1993). The Sonar of Dolphins NY: Springer-Verlag. Babushina, E.S. & Poliakov, M.A. (2004). The underwater and airborne sound horizontal localization by the northern fur seal. Biophysics 49, 723. Bachteler, D. & Dehnhardt, G. (1999). Ac tive touch performance in the Antillean manatee: Evidence for a functional differen tiation of the facial tactile hairs. Zoology, 102, 61-69. Batteau, D.W. (1967). The role of the pinna in human localization. Proceedings of the Royal Society London, 168, 158. Bauer, G.B., Colbert, D.E., Gaspard, J.C., Littlefield, B. & Fellner, W. (2003). Underwater visual acuity of Florida manatees ( Trichechus manatus latirostris.). International Journal of Comparative Psychology 16, 130-142. Bauer, G.B., Gaspard III, J.C., Colbert, DE., Leach, J.B. & Reep, R. (2005, March). Tactile Discrimination of Textures by Florida Manatees, Trichechus manatus latirostris. Paper presented at the 12th annua l International Conference on Comparative Cognition, Melbourne, Florida. 113

PAGE 127

Blauert, J. (1997). Spatial Hearing: The Psychophysics of Human Sound Localization Massachusetts Institute of Technology Press, Massachusetts. Bodson. A., Miersch, L., Mauck, B. & Dehnha rdt, G (2006). Unde rwater auditory localization by a swim ming harbor seal ( Phoca vitulina ) Journal of the Acoustical Society of America 120 (3) 1550-1557. Branstetter, B.K. (2005). Sound localization and auditory pe rception by an echolocating bottlenose dolphin (Tursiops truncatus ). Unpublished Doctoral Dissertation, University of Hawaii, Manoa. Branstetter, B.K. & Mercado III, E. (2006). Sound localization by cetaceans. International Journal of Comparative Psychology 19, 25. Branstetter, B.K., Mevissen, S.J., Herman, L.M., Pack, A.A. & Roberts, S.P. (2003). Horizontal angular discrimination by an echolocating bottlenose dolphin Tursiops truncatus Bioacoustics 14, 15. Branstetter, B.K., Mevissen, S.J., Pack, A.A ., Herman, L.M., Roberts, S.R. & Carsrud, L.K. (2007). Dolphin ( Tursiops truncatus ) echoic angular discrimination: Effects of object separation and complexity. Journal of the Acoustical Society of America 121 (1), 626-635. Brill, R.L. (1988). The jaw-hearing dolphin: Preliminary behavioral and acoustical evidence. In P.E. Nachtigall, & P.W.B. Moore (Eds) Animal Sonar Processes and Performance Plenum Press, New York, pp. 281-287. Brown, C.H. (1994). Sound Localization. In R.R. Fay and A.N. Popper (Eds.) Comparative Hearing: Mammals New York, Springer-Verlag, pp. 57-97. Brown, C.H., Beecher, M.D., Moody, D.B. & St ebbins, W.C. (1980). Localization of noise bands by old world monkeys. Journal of the Acoustical Society of America 68, 127-132. Brown, C.H. & May, B.J. (1990). Sound localiza tion and binaural processes. In M.A. Berkley and W.C. Stebbins (Eds.) Comparative Perception Vol I, New York, John Wiley and Sons, pp. 247-284. Brungart, D.S. & Rabinowitz, W.R. (1999). A uditory localization of nearby sources. Head-related transfer functions. Journal of the Acoustical Society of America 106, 1465. Bullock, T.H., Domning, D.P. & Best, R.C. (1980). Evoked potentials demonstrate hearing in a manatee ( Trichechus inunguis ). Journal of Mammalogy 61 (1), 130133. 114

PAGE 128

Bullock, T.H., OShea, T.J. & McClune, M.C. (1982). Auditory evoked potentials in the West Indian manatee (Sirenia: Trichechus manatus). Journal of Comparative Physiology, 148, 547-554. Burda, H., Bruns, V. & Muller, M. (1990). Sensory adaptations in subterranean mammals. In E. Nevo & O. A. Reig (Eds.), Evolution of Subterranean Mammals at the Organismal and Molecular Levels. Wiley-Liss, NY, pp. 269-293. Carlile, S. & Pettigrew, A.G. (1987). Directi onal properties of the auditory periphery in the guinea pig. Hearing Research 31, 111. Carlile, S. (1990). The auditory periphery of the ferret.: Directional response properties and the pattern of interaural level differences, Journal of the Acoustical Society of America 88, 2180. Casseday, J. H. & Neff, W. D. (1973 ). Localization of pure tones. Journal of Acoustical Society of America 54, 365-372. Chapla, M., Nowacek, D., Rommel, S. & Sadler, V. (2007). CT scans and 3D reconstructions of Florida manatee (Trich echus manatus latirostris) heads and ear bones. Hearing Research 228, 123-135. Cohen, J.L., Tucker, G.S. & Odell, D.K. ( 1982). The photoreceptors of the West Indian manatee. Journal of Morphology, 173, 197-202. Colbert, D.E., Fellner, W., Bauer, G.B., Mani re, C. & Rhinehart, H. (2001). Husbandry and research training of two Florida manatees, Trichechus manatus latirostris. Aquatic Mammals 27 (1), 16-23. Colbert, D. (2005). Sound Localization Abilities of Two Florida Manatees, Trichechus manatus latirostris Unpublished Masters Thesis, Un iversity of South Florida, Tampa, FL. Coles, R.B. & Guppy, A. (1986). Biophysical aspects of directi onal hearing in the Tammar wallaby, Macropus Eugenii Journal of Expe rimental Biology, 121, 371. Don, M. & Starr, A. (1972). Lateralizat ion performance of the squirrel monkey ( Samiri sciureus ) to binaural click signals. Journal of Neurophysiology 35, 493-500. Dudok van Heel, W.H. (1962) Sound in cetacea, Netherlands Journal of Sea Research 1, 407-507. Ehret, G. (1983). Psychoacoustics. In: J.F. Willot & I.L. Springfield (Eds) The auditory psychobiology for the mouse Charles C. Thomas. 115

PAGE 129

Firzlaff, U. & Schuller. G. (2003). Spectral dire ctionality of the extern al ear of the lesser spear-nosed bat, Phyllostomus discolor. Hearing Research 181, 27. Florida Department of Highway Safety and Motor Vehicles (2007). Annual vessel statistics http://www.hsmv.state.fl.us/dmv/vslfacts.html Francis, R.L. (1979). The Preyer reflex audi ogram of several rodents and its relation to the absolute audiogram in the rat. Journal of Auditory Research 19, 217-233. Fuzessery, Z.M. (1996), Monaural and binaural spectral cues created by the external ears of the pallid bat. Hearing Research 95, 1. Gentry, R.L. (1967). Underwater auditory localization in the California seal lion ( Zalophus californianus ), Journal of Auditory Research 7, 187-193. Gerstein, E. (1999). Psychoacoustic Evaluations of the West Indian manatee (Trichechus manatus latirostris) Unpublished Doctoral Di ssertation, Florida Atlantic University, Boca Raton, FL. Gerstein, E., Gerstein, L., Forsythe, S. & Bl ue, J. (1999). The unde rwater audiogram of the West Indian manatee (Trichechus manatus ). Journal of the Acoustical Society of America 105, 3575-3583. Gerstein, E.R. (2002). Manatees, bioacoustics and boats. American Scientist 90, 154163. Griebel, U. & Schmid, A. (1996) Color vision in the manatee ( Trichechus manatus). Vision Research, 36, 2747-2757. Griebel, U. & Schmid, A. ( 1997). Brightness discrimination ab ility in the West Indian manatee ( Trichechus manatus). Journal of Experimental Biology, 200, 15871592. Hartmann, W.M. (1999). How we localize sound. Physics Today 11, 24-29. Hartley, R.V.L. & Fry, T.C. (1921). The binaural localization of pure tones. Physical Review 18, 431. Heffner, R.S. (1997). Comparative study of sound localization and its anatomical correlates in mammals. Acta Otolaryngologica 532, 46-53. Heffner, R.S. & Heffner, E.H. (1982). Hearing in the elephant ( Elephas maximus): Absolute sensitivity, frequency disc rimination, and sound localization. Journal of Comparative and Physiological Psychology 96 (6), 926-944. 116

PAGE 130

Heffner, R.S. & Heffner, E.H. (1984). Sound localization in large mammals: Localization of complex sounds by horses. Behavioral Neuroscience 98 (3), 541555. Heffner, R.S. & Heffner, E.H. (1985). S ound localization in wild Norway rats ( Rattus Norvegicus). Hearing Research, 19, 151-155. Heffner, R.S. & Heffner, E.H. (1988a). Sound localization and use of binaural cues by the gerbil ( Meriones unguiculatus ). Behavioral Neuroscience 102 (3), 422-428. Heffner, R.S. & Heffner, E.H. (1988b). Sound lo calization acuity in the cat: Effect of azimuth, signal duration and test procedure. Hearing Research 36, 221-232. Heffner, R.S. & Heffner, E.H. (1988c). S ound localization in a predatory rodent, the Northern Grasshopper mouse ( Onychomys leucogaster ). Journal of Comparative Psychology, 102 (1), 66-71. Heffner, R.S. & Heffner, E.H. (1989). Sound lo calization, use of binaural cues and the superior olivary complex in pigs. Brain Behavior Evolution 33, 248-258. Heffner, R.S. & Heffner, E.H. (1992a). Evolut ion of sound localizati on in mammals. In, D. Webster, R. Fay and A. Popper (Eds.) The Biology of Hearing SpringerVerlag, pp. 691-715. Heffner, R.S. & Heffner, E.H. (1992b). H earing in large mammals: Sound localization acuity in cattle (Bos Taurus ) and goats (Capra hircus ). Journal of Comparative Psychology, 106 (2), 107-113. Heffner, R.S. & Masterson, R.B. (1990). Sound localization in mammals: Brain stem mechanisms. In M. A. Berkley, & W. C. Stebbins (Eds.) Comparative Perception New York, John Wiley and Sons, pp. 285-314. Holt, M.M., Schusterman, R.J., Southall, B. L. & Kastak, D. (2004). Localization of aerial broadband noise by pinnipeds. Journal of the Acoustical Society of America, 115, 2339-2345. Houben, D. & Gourevitch, G. (1979). A uditory lateralization in monkeys: An examination of two cues serving directional hearing. Journal of Acoustical Society of America 66, 1057-1063. Irvine, D.R. (1987). Interaural intensity differences in the cat: Changes in sound pressure intensity level at the two ears associated with azimuthal displacements in the frontal horizontal plane. Hearing Research 26, 267. Isley, T.E. & Gysel, L.W. (1975). S ound source localizatio n by the red fox. Journal of Mammalogy 56, 397-404. 117

PAGE 131

Jen, P.H. & Chen, D.M. (1988). Directionality of sound pressure transformation at the pinna of echolocating bats. Hearing Research 34, 101. Kastelein, R.A., de Haan, D. & Verboom, W.C. (2007). The influence of signal parameters on the sound source localizat ion ability of a harbor porpoise ( Phocena phocena ). Journal of the Acoustical Society of America 122 (2), 1238-1248. Keller, C.H., Hartung, K. & Takahashi, T.T. (1998). Head-related tr ansfer functions of the barn owl: Measurement and neural responses. Hearing Research 118, 13. Kelly, J.B. & Potas, M. (1986). Directional responses to sounds in young gerbils, Meriones unguiculatus Journal of Comparative Psychology 100, 37. Ketten, D.R., Odell, D.K. & Domning, D.P. ( 1992). Structure, function,and adaptation of the manatee ear. In J. A. Thomas, R. A. Kastelein, & A. Y. Supin (Eds.) Marine Mammal Sensory Systems. Plenum Press, NY, pp.77-79. Kirkpatrick, B., Colbert, D.E., Dalpra, D., Ne wton, E.A.C., Gaspard, J., Littlefield B. & Manire, C.A. (2002). Florida Red Tides, Manatee Brevetoxicosis, and Lung Models. In K. A. Steidinger, J. H. La ndsberg, C. R. Tomas, and G. A. Vargo (Eds.) Xth International Conference on Harmful Algae. Florida Fish and Wildlife Conservation Commission and Intergove rnmental Oceanographic Commission of UNESCO. Klishen, V.O., Diaz, R.P., Popov, V.V. & Supin, A.Y. (1990). Some characteristics of hearing of the Brazilian manatee, Trichechus inunguis. Aquatic Mammals 16 (3), 139-144. Knudsen, E.I. & Konishi, M. (1979. Mechanis ms of sound localization in the barn owl ( Tyto alba ). Journal of Comparative Physiology, 133, 13. Kuhn, G.F. (1977). Model for the interaural time differences in the azimuthal plane. Journal of the Acoustical Society of America 62, 157. Kuhn, G.F. (1987). Physical acoustics and measurements pertaining to directional hearing. In W.A. Yost & G. Gourevitch (Eds.) Directional Hearing Springer Verlag, New York, pp. 3. Kuhn, G.F. & Guernsey, R.M. ( 1983). Sound pressure distribution about the human head and torso. Journal of the Acoustical Society of America 73, 95. Maki, K. & Furukawa, S. (2005). Acous tical cues for sound localization by the Mongolian gerbil, Meriones unguiculatus Journal of the Acoustical Society of America 118 (2), 872-886. 118

PAGE 132

Manire, C.A., Walsh, C.J., Rhinehart, H.L., Colbert, D.E., Noyes, D.R. & Luer, C.A. (2003). Alterations in blood and urine parameters in two Florida manatees, Trichechus manatus latirostris from simulated conditi ons of release following rehabilitation. Zoo Biology 22, 103-120. Mann, D., Colbert, D.E., Gaspard, J.C. III, Casper, B., Cook, M.L.H., Reep, R.L. & Bauer, G.B. (2005). Temporal resolution of the Florida manatee ( Trichechus manatus latirostris ) auditory system. Journal of Comparative Physiology 191, 903-908. Mass, A.M., Odell, D.K., Ketten, D.R. & Supin, A.Y. (1997). Ganglion layer topography and retinal resolution of the Caribbean manatee Trichechus manatus latirostris Doklady Biological Sciences, 355, 392-394. Masterson, B., Heffner, H. & Ravizza, R. (1969). The evolution of human hearing. Journal of the Acoustical Society of America 45, 966-985. Masterson, B., Thompson, G.C., Bechtol d, J.K. & RoBards, M.J. (1975). Neuroanatomical basis of binaural phase-difference analysis for sound localization: A comparative study. Journal of Comparative Physiological Psychology, 89, 379-386. Medwin, H. & Clay, C.S. (1998). Fundamentals of Ac oustical Oceanography NY: Academic Press. Mehrgardt, S. & Mellert, V. ( 1977) Transformation characteristics of the external human ear. Journal of the Acoustical Society of America 61, 1567. Middlebrooks, J.C., Makous, J.C. & Green, D.M. (1989). Directional sensitivity of sound-pressure intensity levels in the human ear canal. Journal of the Acoustical Society of America 86, 89. Middlebrooks, J.C. & Green, D.M. (1991). Sound localization by human listeners. Annual Review of Psychology, 42, 135-159. Mills, A.W. (1958). On the minimum audible angle. Journal of the Acoustical Society of America 30, 127-246. Mills, A.W. (1972). Auditory loca lization. In J. V. Tobias (Ed.) Foundations of Modern Auditory Theory : Vol. II, (pp. 3030-348). NY: Academic Press. Mhl, B., Au, W.W.L., Pawloski, J. & Nachtig all, P.E. (1999). Dolphin hearing: Relative sensitivity as a function of point of a pplication of a contact sound source in the jaw and head region. Journal of the Acoustical Society of America 105, 34213424. 119

PAGE 133

Moiseff, A. (1989). Binaural disparity cu es available to the barn owl for sound localization. Journal of the Acoustical Society of America 59, 1222. Moore, P.W.B. (1974). Underwater localiza tion of click and pulse d pure-tone signals by the California sea lion ( Zalophus californianus ). Journal of the Acoustical Society of America 57 (2), 406-410. Moore, P.W.B. & Au, W.L. (1975). Underwat er localization of pul sed pure tones by the California sea lion ( Zalophus californianus). Journal of the Acoustical Society of America, 58 (2), 721-727. Moore, P.W.B. & Brill, R.L. (2001). Binaural hearing in dolphins. The Journal of the Acoustical Society of America 109 (5), 2330-2331. Moore, P.W.B. and Pawloski, D.A. (1993). Interaural time discrimination in the bottlenose dolphin. The Journal of the Acoustical Society of America, 94 (3), 1829-1830. Moore, P.W.B., Pawloski, D.A. & Dankiewi cz, (1995). Interaural time and intensity difference thresholds in the bottlenose dolphin ( Tursiops truncatus ). In R.A. Kastelein, J.A. Thomas & P.E. Nachtigall (Eds.) Sensory Systems of Aquatic Mammals De Spil Publishers, The Netherlands, pp 1124. Musicant, A.D., Chan, J.C. & Hind, J.E. (1990) Direction-dependent spectral properties of cat external ear: New data and cross-species comparisons. Journal of the Acoustical Society of America 87, 757. Nowacek, D.P., Casper, B.M., Wells, R.W ., Nowacek, S.M. & Mann, D.A. (2003). Intraspecific and geographic variat ion of West Indian manatee ( Trichechus manatus spp.) vocalizations. Journal of the Acoustical Society of America 114 (1), 66-69. Nowacek, S.M., Wells, R.S., Owen, E.C.G., Speakman,T.R., Flam, R.O. & Nowacek, D.P. (2004). Florida manatees, Trichechus manatus latirostris respond to approaching vessels. Biological Conservation 119, 517-523. Obrist, M.K., Fenton, M.B., Eger, J.L. & Schl egel, P.A. (1993). What ears do for bats: A comparative study of pinna sound pressure transformation in chiroptera. Journal of Experimental Biology, 180, 119-152. Phillips, D.P., Calford, M.B., Pettigrew, J.D., Aitkin, L.M. & Semple,M.N. (1982). Directionality of sound pressure tr ansformation at the cats pinna. Hearing Research 8, 13. 120

PAGE 134

Piggins, D.J., Muntz, W.R.A. & Best, R.C. (1983). Physical and mo rphological aspects of the eye of the manatee, Trichechus inunguis Marine Behaviour and Physiology 9, 111-130. Popov, V.V. & Supin, A.Y. (1990). Electrophys iological studies on hearing in some cetaceans and a manatee. In J.A. Thomas, & R. A. Kastelein (Eds.), Sensory Abilities of Cetaceans: Laboratory and Field Evidence NY: Plenum Press. Reep R.L. Marshall, C.D. & Stoll, M.L. (2002). Tactile hairs on the postcranial body in Florida manatees: a mammalian lateral line? Brain Behavior and Evolution 59, 141-154. Renaud, D.L. & Popper, A.N. (1975). S ound localization by the bottlenose porpoise Tursiops truncatus Journal of Experimental Biology 63, 569-585. Reynolds, J.E. III. (1979). The semisocial manatee. Natural History 88 (2), 44-53. Rice, J.J., May, B.J., Spirou, G.A. & Young, E. D. (1992). Pinna-based spectral cues for sound localization in cat. Hearing Research 58, 132. Richardson, W., Greene, C., Malme, C. & Thompson, D. (1995). Marine Mammals and Noise. San Diego, CA: Academic Press. Roth, G.L., Kochhar, R.K. & Hind, J.E. (1980). Interaural time differences: Implications regarding the neurophysiol ogy of sound localization. Journal of the Acoustical Society of America, 68, 1643. Searle, C.L., Braida, L.D., Cuddy, D.R. & Davis, M.F. (1975). Binaural pinna disparity: Another auditory localization cue. Journal of the Acoustical Society of America 57, 448. Spezio, M.L., Keller, C.H., Marrocco, R.T. & Takahashi, T.T. (2000). Head-related transfer functions of the rhesus monkey. Hearing Research 144, 73. Stevens, S.S. & Newman, E.B. (1936). The localization of actual sources of sound. American Journal of Psychology. 48, 297-306. Strutt, J.W. (Lord Rayleigh) (1907). On our perception of sound direction. Philosophical Magazine, 13, 214. Terhune, J.M. (1974). Directional hearing of a harbor seal in air and water. Journal of the Acoustical Society of America 56 (6), 1862-1865. Urick, R.J. (1996). The Principles of Underwater Sound 3rd Edition (423pp.). CA: Peninsula Publications. 121

PAGE 135

Wakeford, O.S. & Robinson, D. E. (1974). La teralization of tonal stimuli by the cat. Journal of the Acoustical Society of America 55, 649-652. Walls, G.L. (1963). The vertebrate eye and it s adaptive radiation NY: Hafner. West, J.A., Sivak, J.G., Murphy, C.J. & K ovacs (1991). A comparative study of the anatomy of the iris and ciliary body in aquatic mammals. Canadian Journal of Zoology, 69, 2594-2607. Wright, D., Hebrank, J.H. & Wilson, B. ( 1974). Pinna reflections as cues for localization. Journal of the Acoustical Society of America ,. 56, 957. Wightman, F.L. & Kistler, D.J. (1997). Factors effecting the rela tive salience of sound localization cues. In R.H. G ilkey & T.R. Anderson (Eds.) Binaural and Spatial Hearing in Real and Virtual environments, Lawrence Erlbaum, Mahwah, NJ, pp. 1. Yost, W.A. (2000). Sound localizat ion and binaural hearing. In Fundamentals of hearing: An introduction. San Diego, CA: Academic Press (pp. 179-192). 122

PAGE 136

Chapter Four: Potential Sound Conduction Pathways for the Florida Manatee, Trichechus manatus latirostris Abstract Behavioral investigations have demonstr ated that manatees possess the capacity to detect and localize sounds over a wide range of frequencies (Bullock et al., 1980; 1982; Klishen et al., 1990; Popov & Supin, 1990; Gerstein, 1999; Gerstein et al., 1999; Colbert, 2005; Mann et al., 2005). Paradoxi cally, anatomical investigations have established that the manatees external and mi ddle ear is formed in a manner atypical of most mammalian species (Ketten et al., 1992; Chapla et al., 2007). The external auditory meatus is occluded and separated from the tympanic membrane making it an unlikely channel for sound transmission, the tympanoperio tic complex is located intracranially but not ossified to the skull, and the ossi cles are massive. Several sound conduction pathways outside the traditional pinna-to-cochlea conduit have been proposed to explain these anatomical anomalies, however the sp ecific means by which manatees hear remains unknown. Auditory evoked potential (AEP) tec hniques, using 15 kHz (154.9 dB re 1 Pa) and 24 kHz (158.8 dB re 1 Pa ) carrier tone bursts that were amplitude modulated (AM) with a 600 Hz rate, were used to map po ssible sound conduction pathways with four Florida manatees ( Trichechus manatus latirostris) Voluntary AEP measurements were obtained from positions on the heads of tw o subjects, Hugh and Buffett, while all portions of their body, excluding th e electrodes, were positioned in the water. Restrained 123

PAGE 137

evoked potential measurements were obtained from various positions on the heads and torsos of three subjects, Hugh, Mo, and Bock, wh ile all portions of th eir bodies were in air. Data included in this study were colle cted prior to the development of a formal methodological plan to inves tigate the possible existen ce of manatee sound conduction pathways and should be considered with cau tion. Transducer pos itions were coded by video analysis and results were derived th rough the compilation and organization of the data already collected. Results demonstrated that all four subjects, regardless of being positioned in air or in water, produced AEPs at every position the transducer was placed on their bodies, however no obvious sound conduction pathway was identified. Estimated effective sound pressure levels between body positions we re found to be proportionally the same. AEP amplitudes were usually greater with the 24 kHz carrier when tested in both the inair and in-water mediums, however patterns between carriers at id entical body positions were highly variable between s ubjects. In-water testing dem onstrated identical or similar AEP amplitudes at six of seven common pos itions with the 24 kHz carrier, however amplitudes were inconsistent for all but one of the common positions with the 15 kHz carrier. In-air testing showed that Bock a nd Mo had similar AEP amplitudes at four of five common positions with both carriers, however Hugh shared only one similar AEP amplitude out of the nine positions common to the three subjects with the 15 kHz carrier. Evoked potentials, averaged together fr om positions along the vertebral column and lateral ribs that were more than 20 cm caudal to the scapula, were greater than those averaged together from positions at and dorsal to the meatus, those averaged from 124

PAGE 138

125 positions along the zygomatic process, and those averaged from positions along the vertebral column and lateral ribs that we re cranial to 20 cm behind the scapula.

PAGE 139

Introduction The Florida manatee ( Trichechus manatus latirostris ) is an endangered species protected by both the Marine Mammal Protect ion Act (1972) and the Endangered Species Act (1973). Investigations of the manatee s sensory processes re veal that vision and touch likely function best with tasks in close proximity, while audition functions effectively for tasks that are both nearby and at a distance (C hapters 2 & 3). The hearing range of the manatee has b een assessed through the development of an audiogram and by utilizing auditory evoke d potential techniques. Gerstein et al. (1999) obtained a behavioral audiogram fo r two manatees which showed hearing thresholds that ranged from 0.5 kHz for one subject and 0.4 kHz for the other. The frequency range of best hearing was between 10 kHz and maximum sensitivity was ~50 dB re: 1 Pa at 16 and 18 kHz, decreasing by ~20 dB per octave from 0.8 to 0.4 kHz and 40 dB per octave above 26 kHz. Auditory evoked potential measurements have been obtained in several studies. Bulloc k et al. (1980; 1982) a nd Popov & Supin (1990) found that the highest frequency detection reach ed 35 kHz when tested in air and Klishen et al. (1990) found it reached 60 kHz when tested in water. More recently, Mann et al. (2005) found that detection reached 40 kHz when tested in water, results similar to those found by Bullock (1980; 1982), Popov & Supin (1990). These results indicate that manatees are able to detect conspecific vo calizations which typically range between 2.5 5.9 kHz (Nowacek, et al., 2003) and boat engi ne noises which typi cally range between 0.01 kHz (Gerstein, 2002; Richardson et al., 1995). 126

PAGE 140

The localization abili ties of the manatee have been investigated though fourchoice discrimination paradigms that presen ted acoustic stimuli in the frontal 180 (Gerstein, 1999; Colbert 2005) and an eight-choi ce paradigm with stimuli presented in all 360 of the horizontal plane surrounding a st ationary subject (Chapter 2). Results indicated that they were pr oficient at localizing broadba nd stimuli over a wide range of frequencies (0.2-24 kHz) as well as those restricted to high (6-20 kHz) and low frequencies (0.2 kHz) at va rious durations and levels as well as tonal stimuli. Performance accuracy was decreased with lo wer levels, decreased durations and tonal stimuli, but still remained above chance leve ls. These results suggested that manatees were able to localize frequencies having wave lengths that were both shorter and longer than their interaural distances. Head related transfer functions were measur ed with two manatees to investigate if interaural level difference (ILD) cues facilitated their abil ity to determine sound source directionality (Chapter 3). Results indicated that ILDs we re found for all frequencies as a function of source location a nd that the torso provided greater shadowing effects than the head. ILDs appeared to be magnified wh en sounds originated behind the subjects and may have compensated for their inability to visually orient towards these locations as stimuli were presented. The findings of these behavioral investig ations might suggest that the manatee auditory system is built and functions similarly to that of typical mammalian species. This assumption however, is false and anatomical examination of the manatee ear has provided evidence to the contrary. Unlike most terrestrial mammals, but similar to fully aquatic marine mammals such as cetaceans, the manatees external pinna flange is absent 127

PAGE 141

(Ketten et al., 1992; Chapla et al., 2007). The entrance to the external auditory meatus (EAM) is only 1 mm in diameter and the EAM is ~61 mm long and has a ~3.6 mm diameter at its widest point (Chapla et al., 2007). In cont rast to most terrestrial and marine mammals, the manatees EAM is occl uded by cellular debris and reaches a blind end that is separated from the tympanic membrane which makes it an unlikely channel for sound transmission (Ketten et al., 1992; Chapla et al., 2007) (Figure 3). Figure 4.1. Diagrammatic illustration of manatee auditory anatomy based on multiple cross-sections through the transverse plane. eam, external auditory meatus; eao, external auditory opening; h, hyoid bones; htr, hypotympanic recess; mec, middle ear cavity; pb, periotic bone; sq, squamosal bone; tb, tympanic bone; tm, tympanic membrane. Figure used with permission from Chapla et al., 2007. The manatee middle ear is composed of a large bilobed peri otic bone and the tympanic bone. The two are connected to one another at two small locations that may act as a hinge and are called the tympanoperiotic complex (Fleischer, 1978; Ketten et al., 1992). Cetaceans have a tympanoperiotic comple x that is located extracranially and has no bony attachment to the skull (Ketten, 1992) Manatees have a tympanoperiotic 128

PAGE 142

complex that is located intracranially, howev er it is not ossified to the skull and the periotic is only attached to the occipital bone by a small ~ 5 mm cartilage disc and to the squamosal bone by an even smaller ~2.5 mm carti lage disc (Ketten et al., 1992; Chapla et al., 2007). The middle ear cavity is composed of tw o main sections, the epitympanic recess and the mesotympanum (Ketten et al., 1992). Th e epitympanic recess is filled with soft tissue and surrounds the short arm of the incus. The mesoty mpanum is divided into two chambers that are separated by a membranous septum. The lateral chamber contains the ossicular chain and abuts the tympanic membrane. The medial chamber includes the round window and is bordered ventrally by soft tissue and dorsomedially and caudally by the skull. The hypotympanic recess connects to the middle ear cavity, is bound ventrally by soft tissues, and allows air to pass betw een it and the nasopharynx via the eustachian tube (Fleischer, 1978; Ketten et al., 1992; Domning, 2001). The tympanic membrane is multilayered, has an elliptical shape, and is laterally convex (Ketten et al., 1992). The ossicles are massive with most of their 5400 mg mass centered in the large malleus h ead (Figure 3.2). The cross-sec tional area of the ossicular chain has been found to be proportional to th e area of the tympanic membrane for most terrestrial mammals (Nummela, 1995) and fo r cetaceans (Nummela et al., 1999). A greater tympanic membrane area intensif ies the amount of energy collected by the membrane so the ossicular chain must beco me modified and enlarged as a means to endure the increased membrane vibrati on forces. Numm ela (1995) found area proportions to be ~ 0.1, for species ranging in size from the elephant to the shrew, however the area proportion was found to be greatly increased to 0.4 for seals whose 129

PAGE 143

ossicular chain was significantly larger than the area of the tympanic membrane. The manatee ossicular chain shows a similar deviation and is ove rly massive compared to the tympanic membrane area as well as the other ty mpanic and periotic ear bones (Chapla, et al., 2007). Figure 4.2. Manatee ossicles from the right ear. (A) Malleus, incus, and stapes; (B) malleus magnified; (C) stapes magnified; 1. tip of incus sh ort arm (weakly fuses within epitympan ic recess of periotic); 2. malleus rostral ossification (connects with tympanic bone); 3. malleus caudal process (attachment site for tympanic membrane); 4. malleus medial process (attachment site for tensor tympani muscle); 5. ligamentary vestige of the stapedial artery (traverses stapes dorsoventral foramen); 6. stapedial footplate (abuts the oval window); f, facets by which malle us and incus articulate (arrows denote small f acets anddotted line in B defines outline); i, incus; k, cartilaginous keel (das hed line marks the border of the keel and the malleus manubrium); mh, malleus head; mm, malleus manubrium; s, stapes; sm, stapedius muscle; tt, tensor tympani muscle. Figure used with permission from Chapla et al., 2007. The anatomical anomalies found in the manatees outer and middle ears of indicate that their auditory sy stem functions in a manner dissi milar to that of the typical terrestrial or marine mammal species. Se veral sound conduction pathways outside the traditional pinna-to-cochlea course have been proposed that use the zygomatic process, cranial tissues, cranial bones, vertebrae and/or lungs to directly stimulate the tympanic membrane. 130

PAGE 144

The manner by which sound is transmitted between water and soft body tissue, or between soft body tissue and bone, is complex and somewhat dependent upon the elastic, reflective and absorptive properties of the tissues and bones. Th e velocity of sound traveling through terrestrial mamm al soft tissues has been found to be linearly related to the density of the tissues (Mas t, 2000). Soft tissue densit y varies ~10% from that of seawater and velocity varies ~15% (A royan, 1996). Tissues containing increased structural elements such as collagen reta in higher densities and sound velocities than water (Goold & Clarke, 2000), while those composed of greater fat or lipid content retain lower densities and sound velocities (Mast, 2000). Investigations with cetaceans have shown that fatty tissue has a density and s ound velocity less than that of sea water, muscle has a density and sound velocity similar to that of seawater, and connective tissue has a density and sound velocity greater than th at of seawater (Solde villa et al., 2005). Cetaceans have acoustic fats located in their mandible and melon that are less dense than the surrounding blubber tissues. Soldevilla et al. (2005) suggested that the velocity change which occurs as sound travel s from seawater to blubbe r tissue and then to the acoustic fat, likely plays an important im pedance matching role as sound is channeled to the middle ear complex. Investigations with manatees have show n that the tympanoperiotic complex is connected to the squamosal bone, which in tu rn, is connected to the zygomatic process (Figure 4.3) (Ketten et al., 1999; Chapla et al., 2007). Kette n et al. (1999) found that the zygomatic process differed from all othe r cranial bones and was a lipid-filled bony sponge that may serve a unique function to enhance sound transmission much like the acoustic fat found with cetaceans. The zygomatic process was found to have 131

PAGE 145

significantly lower density than other bones (Fawcett, 1942; Caldwe ll & Caldwell, 1985; Domning & de Buffrenil, 1991), however the li pids it contained were composed almost entirely of triacylgly cerols (Ames et al., 2002) and not the isovaleric acid typical of cetacean acoustic fat by which sounds are condu cted (Varanasi & Malins, 1971). Chapla et al. (2007) suggest that the distinctive com position of the zygomatic process causes it to be less rigid than other dense bones. This increased elasticity may enhance sound wave propagation along the zygomatic process to stimulate the tympanic membrane and tympanic bone. zp A B Figure 4.3. Right lateral view of a three dimensional reconstruction of a CT scanned manatee head (MSW0058). (A) Skin layer showing relation of external auditory opening (eao) to eye. (B) Skeletal layer showing relation of squamosal bone (sq, pink) and the zygomatic process (zp, pink) which is filled with fats and blood vessels to external auditory meatus (eam), eye, and tympanic bone (tb). Figure edited from Fig. 4 of, and used with permission from, Chapla et al., 2007. Chapla et al. (2007) found that the soft ti ssues of the manatee head have a density similar to that of seawater suggesting that sound waves could propagate easily from one 132

PAGE 146

medium to the other. In addition, these au thors suggest that these soft tissues are arranged in a manner that allows sounds to be transmitted to the mana tees inner ear with a minimal amount of reflection (Figure 4.4). Sound waves with azimuth angles between 45o and 90o (this includes the area surrounding the external auditory meatus) and elevation angles between ~43o and 73o should be able to propagate through these soft tissues, without reflecting off of the squamosal bone, to stimulate the tympanic membrane directly. Figure 4.4. Potential sound pathways where a sound wave will experience the least amount of reflection. (A) Dorsal view showing that soundwaves with angles of incidence between 45o and 90o from the midsaggital line, on both the left and right sides of the head, may reach the ear without reflecting off of the squamosal. (B) Rostral view with an axial cut at the level of the ear showing that sound waves with angles of incidence (measured from the horizontal) between 43o and 64o (on the right side of the head) and between 55o and 73o (on the left side of the head) have only dermis and fatty tissue to pass through in order to reach the ears. The external auditory meatus is also present within this area, but is composed only of soft tissue and has no connecti on with the tympanic membrane. Airs paces, teal; cartilage, fuchsia; fatty tissue, yellow; malleus, green; muscle, dark red; periotic, purple; salivary glands, dark blue; skin, gray; squamosal, pink; tympanic bone, light blue; tympanic membrane, orange. Figure used with permission from Chapla et al., 2007. Sound can also be transmitted to the inner ear through bone. Sound waves can cause the bones of the skull to vibrate thr ough inertial, compressional, and osseotympanic movements (Tonndorf, 1966; Yost, 2000; Gelf and, 2004). In humans, inertial bone conduction occurs with assistance from the mi ddle ear and with frequencies below 800 Hz which causing the skull to vibrate as one unit while the ossicular chain lags behind it 133

PAGE 147

due to its inertia. The lagging motion of the o ssicular chain causes the stapes to stimulate the oval window in a manner identical to air conduction. Comp ressional bone conduction occurs with assistance from the inner ear a nd with frequencies above 800 Hz and causes the temporal bone to vibrate in such a manne r that the cochlear capsule compresses and expands simultaneously. These compressions cause the round window to oscillate and send a traveling wave through the cochlea. Osseotympanic bone conduction occurs with the assistance of the outer ear and at frequencies below 1,000 Hz which causes the external auditory canal to vibrate and radiate along the length of the canal to stimulate the tympanic membrane. An occlusion effect o ccurs when the external auditory canal is blocked. Bone-conducted sound vibrations are prevented from radiating out of the ear canal and are instead reflected back toward the tympanic membrane. The occlusion effect has been found to boost sound pressure in the ear canal by 20 dB with frequencies below 500 Hz. Investigations with cetacean s have suggested that high frequencies may cause the thinner portion of the tympanic bone to vibrat e with greater amplit ude than the thicker portion (Hemila et al., 1999). These vibrations would in turn be conducted to the ossicular chain causing the tympanic bone to act much like a tuni ng fork (Fleischer, 1978). Chapla et al., (2007) suggest that the manatees massive ossicular chain may have evolved to function in a similar ma nner by increasing movement relative to the tympanoperiotic complex and skull. Several investigations have revealed that the lungs and skeletal system of snakes and turtles play important roles in heari ng (Hartline, 1971; Lenhardt, 1982; Lenhardt et al., 1983). These reptiles have lungs that lie in a horizontal plane along the bodys length 134

PAGE 148

instead of the transverse plane typical fo r mammals. This positioning was found to facilitate sound wave vibrations to be received by the lungs and skeletal system and then be conducted to the ears. The position of the manatees pleural cavity and lungs also diverge from the typical mammalian arrangement and are similar to those found in many reptiles (Chapla et al., 2007). The manatee l ungs extend the full leng th of the body cavity and lie dorsal to the heart (Figure 4.5) (Rommel & Lowenstine, 2000). The pleural cavities are supported by two separate diap hragms (hemidiaphragms) instead of the typical single mammalian diaphragm (Rom mel & Reynolds, 2000). The cranial portion of the hemidiaphragms are attached to the first three ribs and extend from the sixth cervical vertebra to the 26th vertebra, spanning an incredib le 40% of the total body length (Rommel & Reynolds, 2000). Chapla et al. ( 2007) suggest that vi brations from sound waves may be transmitted through the lungs, ribs and/or spinal column to the skull and ear bones. Rommel & Reynolds (2000) further suggest that the separation of the hemidiaphragms may provide individualiz ed cues to aid in sound localization. 135

PAGE 149

Figure 4.5 Schematic illustrations of the manatee diaphragm and lung. A: In left lateral view, the manatee lung is a relatively flattened and elongate structure that occupies the dorsal region of the pleural cavity. B: In cross section the manatee body is an ellipse, with the diaphragm stretched almost horizontally from the hypapophyses at the midline to the ribs at mid shaft. Figure used with permission from Rommel & Reynolds, 2000. Behavioral investigations with the manat ee have demonstrated that they are able to hear and localize sounds over a wide range of frequencies (C hapter 2) and that interaural level cues likely assist in locali zation (Chapter 3). The unusual anatomy of the manatee ear however, causes sp eculation as to how these so unds are received by their auditory system. Auditory evoked potential (AEP) techniques may be a valuable tool for clarifying which sound conduction pathways are most prominent for the manatee. AEPs are neural electrical firing responses that spontaneously occur when an acoustic stimulus is received. Although indivi dual AEPs have amplitudes that only range up to several microvolts, they can be detected through electrodes placed on the head (Ferraro & Durrant, 1994). To amplify these AEPs, amplitude modulated tones are presented rapidly which result in an envel ope following response (EFR) in which neural 136

PAGE 150

responses are phase-locked with the stimulus and averaged together to make it easily distinguishable from electrical noise (Dolphin, 1996, 1997). AEP techniques have been traditionally used to determine hearing thresholds and have been used with birds (Lucas et al., 2002), terrestrial mammals (Corwin et al., 1982) cetaceans (Ridgway, et al., 1981; Szymanski et al., 1999; Cook et al., 2006), and manatees (Bullock et al., 1980, 1982; Klis hen et al., 1990; Popov & Supin, 1990; Mann et al., 2005). Bullock et al. (1982) conducted a cursory in vestigation to determine if specific areas of the manatee head had increased acoustic sensitivity when they measured the in-air hearing thresholds of four West Indian manatees ( Trichechus manatus ) using AEP techniques. Results suggested that the area surrounding the external auditory meatus showed only a slightly higher sensitivity than a consid erable area in front of it, suggesting that acoustic energy may be received over a large area (Bullock et al., 1982). The manatees ability to detect and locali ze sounds and the atypical anatomy of its ear seems paradoxical and signifies an area that merits further research. Results found by Bullock et al. (1982) represent a lim ited sound conduction pathway evaluation but introduce the potential for utiliz ing AEP techniques to more fully evaluate the existence of sound conduction pathways outside of the traditional pinna-to-cochlea conduit. The objective of this study was to evaluate if AEP measurements that were previously obtained using in-air and in-water when ac oustic stimuli were presented on various positions of the heads and torsos of four manatees would identify the existence of specific sound conduction pathways. 137

PAGE 151

Hypotheses Two hypotheses were made. The first posited that auditory evoked potentials would be of greater magnitude at the position of the external auditory meatus than at the zygomatic process. The zygomatic process is a lipid-filled bony sponge (Ketten et al. 1999) that has a lower density than other bones (Fawcett, 1942; Caldwell & Caldwell, 1985; Domning & de Buffrenil, 1991). It has be en suggested that the uniqueness of this bone may enhance sound transmission much like the acoustic fat found in the cetacean mandible and melon (Varanasi & Malins, 1971) however the lipids in the zygomatic process differ from that of cetacean acoustic fa t (Ames et al., 2002). Chapla et al. (2007) suggested that manatee intercranial tissue arrangements n ear the external auditory meatus had impedance properties similar to that of water and had a minimal amount of reflection which may facilitate sound transm ission to their inner ears (Figure 4.4). The second hypothesis stated that auditory evoked potentials w ould be of greater magnitude at points along the vertebral column and lateral ribs that are more than 20 cm caudal to the scapula than those located crania l to, at the level of, or up 20 cm caudal to the scapula. Hartline (1971), Lenhardt (1982), and Lenhardt et al. (1983) found that several reptiles (snakes and turtles) had l ungs that lay in a hori zontal plane along the bodys length through which vibratory stimulation was transferred from the lungs and skeletal system to the ears. Rommel & Re ynolds (2000) found that the manatees lungs also lie in a horizontal plan e along the bodys length, and ar e composed of two pleural cavities that are ventrally supported by he midiaphragms. The hemidiaphragms are 138

PAGE 152

attached to the first three ribs and extend back to the 26th vertebra which accounts for approximately 40% of the manatees total body length (Figure 4.5). Chapla et al. (2007) suggest that vibrations from sound waves may be transmitted through the manatees lungs, ribs and/or spinal column to the sku ll and ear bones. The manatees large scapulas may provide a deaf-spot that sound waves reflect from. 139

PAGE 153

Materials and Methods Subjects The subjects for this study include d four male Florida manatees ( Trichechus manatus latirostris ), Hugh and Buffett who reside at Mote Marine Laboratory and Aquarium in Sarasota, Florida, and Mo and Bock who reside at Walt Disney Worlds The Living Seas at EPCOT in Lake Buena Vista, Florida. All procedures used with these subjects were permitted through the United St ates Fish and Wildlife Service (Permit # MA837923-6) and approved by the Institutional Animal Care and Use Committees of each facility. Hugh and Buffett were both captive-born animals. Hugh was 20 years of age, weighed 547 kg, and was 310 cm in length, while Buffett was 17 years of age, weighed 773 kg, and was 334 cm in length. They were housed in a 265,000 lit er exhibit that was composed of three inter-connected sections : a 3.6 x 4.5 x 1.5 m Medical Pool, a 4.3 x 4.9 x 1.5 m Shelf Area, and a 9.1 x 9.1 x 3 m Exhibi t Area (Figure 4.6). Both animals had acquired an extensive training history and were conditioned to voluntarily participate in a prior auditory evoked potential study, ma king them excellent candidates for this investigation (Mann et al., 2005) In addition, they had been behaviorally conditioned for husbandry procedures (Colbert et al., 2001) a nd studies which investigated lung capacity (Kirkpatrick et al., 2002), seru m and urine creatinine levels as a function of release conditions (Manire et al., 2003), visual acuity (Bauer et al., 20 03), facial vibrissae tactile 140

PAGE 154

sensitivity (Bauer et al., 2005), as well as a four-choice (Colbert 2005) and an eightchoice sound localization study (C hapter 2). Voluntary evok ed potentials measurements were obtained from the cranial regions of Hugh and Buffett while all portions of their body, excluding the electrodes, were in th e water. Restrained evoked potential measurements were also collected from H ugh when he was dry-docked in the drained medical pool with all por tions of his body in air. 4.5 x 4.9 x 1.5 m Shelf Area 3.6 x 4.5 x 1.5 m Medical Pool 9.1 x 9.1 x 3 m Exhibit Area Figure 4.6. Testing setup for voluntary auditory evoked potential measurements used to map sound conduction pathways with subjects at Mote Marine Laboratory & Aquarium. Subjects stationed facing the northeast wall of the Shelf Area. The blue octagon represents the Test Trainers location, the green square represents the Data Recorders location, and the orange triangle represents the Subject Handler. Lines a, b, and c represent the reference, recordin g and ground electrode leads respectiv ely, that travel to the amplifier housed in a water resistant case (yellow rectangle) which was connected to the Workstation. The blue line represents the transducer. a b c Mo and Bock were both orphaned shortly af ter birth. Mo was 10 years of age, weighed 458 kg and was 280 cm in length. Bock was 4 years of age, weighed 346 kg and was 247 cm in length. They were housed w ith a variety of fish species in a 465,605 liter exhibit that was composed of two inter-c onnected sections: a 14.17 m x 7.16 m x 3.27 m Public Display Pool and a 8.23 m x 4.42 m x 3.70 m adjoining Off-display Medical Pool (Figure 4.7). All evoked potential measur ements obtained from these subjects were 141

PAGE 155

collected when they were restrained in th e drained off-display me dical pool with all portions of their bodies in air. Figure 4.6. Testing setup for restrained auditory evoked potential measurements used to map sound conduction pathways with subjects at Walt Di sney Worlds The Living Seas at EPCOT. Subjects were restrained out of water in the Off-display Medical Pool. Gray shaded areas represent walkways and the dashed line represents a gate that c onnects the two pools under the walkway. 8.23 m x 4.42 m x 3.70 m Off-display Medical Pool 14.17 m x 7.16 m x 3.27 m Public Display Pool Experimental Design All data included in this study were collected between September, 2003 and February, 2005, prior to the development of a formal methodological plan to investigate the possible existence of manatee sound c onduction pathways. Transducer positions were coded by video analysis and results were derived through the compilation and organization of these data. Voluntary AEP measurements with Hugh and Buffett had been previously obtained (Mann et al., 2005). The subjects had been trained to station motionless at a target placed against the northeast wall of the shelf area (Figure 4.1). Through a process 142

PAGE 156

of counter-conditioning (Pearce & Dickin son 1975; Domjan 2003), they were desensitized to surgical scr ub preparation of the skin whic h consisted of isopropyl alcohol and betadine scrubs that were alternated three times each, and insertion of two 27-gauge needle electrodes (Rochester Electro-Medical). The recording electrode was inserted 0.7.0 cm into the skin above the cranium, approximately 5 cm cranial to the back of the skull and the reference electrode was inserted to the same skin depth approximately 20 cm caudal to the recording el ectrode. A third electrode, the ground, was placed in the water. The aversive properties of the needle insertions were countered by the immediate presentation of food reinforcement including ap ples, carrots, beets and monkey biscuits if they remained motionless. Subjects were tr ained to remain still for a duration of 2 minutes during recording bouts and while th e transducer was placed underwater at different positions located cranial to the scapula (Figure 4.7). Figure 4.7. Voluntary AEP measurements with Hugh. Three personnel were needed to obtain voluntary AEP measurements, including a Test Trainer, Data Recorder and Subject Handler (Figure 4.5). The Test Trainer 143

PAGE 157

maintained the subjects proper behavior, pe rformed the surgical scrubs, inserted the electrodes and positioned the transducer. The Data Recorder initiated and recorded each trial via the computer. The Subject Handler wa s positioned in the water to the right of the subject with his/her knee positioned under th e subjects sternum to ensure that the electrodes remained above water and provide d the subject primar y food reinforcement between recording bouts. Restrained measurements with Hugh, Mo and Bock required that each subject be dry-docked out of the water and confined as much as possible to avoid movement. Subjects were placed on closed cell foam pa ds and their skin was kept moist with wet towels or water from a garden hose. The surgical scrub procedures and needle electrode positions were identical to those used w ith voluntary recordings, however the ground electrode was also inserted in to a surgically scrubbed area approximately 10 cm lateral to the reference electrode. The transducer wa s positioned on different locations of their entire bodies (Figure 4.8). Figure 4.8. Restrained AEP measurements with Bock. 144

PAGE 158

Restrained AEP procedures required 710 personnel including the Test Trainer, Data Recorder and numerous Subject Handlers The Test Trainer was responsible for performing the surgical scrubs, inserting the electrodes, and positioning the transducer. The Data Recorder was responsible for in itiating and recordin g each trial via the computer. The Subject Handlers were respon sible for keeping the subject as motionless as possible and hi skin moist. Signal Generation and Programming A Tucker-Davis Technologies AEP wo rkstation and laptop computer (Dell Latitude D505) with SigGen and BioSig softwa re were used to present and collect all evoked potential data. The same workstation ha d been previously us ed to investigate cetacean AEPs (Cook, 2006; Cook, et al., 2006). Signals were generated with a 100 kHz sample rate, amplified by a Hafler amplifier (P1000) and delivered via a piezoce ramic transducer (ITC-1042) that was embedded in a suction cup (VI-SIL V-1062, Rhodi a, Inc.) constructed of a silicone-based material that had an acoustic impedance sim ilar to water (Brill et al. 2001). A 15 kHz carrier tone burst was pr esented at 154.9 dB re 1 Pa and a 24 kHz carrier tone burst was presented at 158.8 dB re 1 Pa. These tone bursts had 5 ms cosine-squared rise-fall times that were amplitude modulated (AM) with a 600 Hz rate that were 40.96 ms and presented 14.5 times per second. The 600 Hz AM rate was found to have produced the largest AEPs with 15 and 24 kHz carrier fre quencies in a study that investigated the temporal resolution of Hugh and Buffett (Mann et al., 2005). The fr equencies and levels that were presented were chosen beca use they produced the largest AEPs. 145

PAGE 159

AEP electrical responses received thr ough the electrodes were returned to a differential amplifier (TDT-RP2.1) that was housed in a water-r esistant case that could be easily positioned next to the subject. Thes e signals were differentially amplified and averaged with an acquisition sample rate of 25 kHz. Amplified signals were sent via fiber optic cable to the TDT Workstation for da ta analysis using the BioSig software. Evoked potentials were collected in re sponse to 200,000 presentations of the stimulus. Underwater calibration was performed from within the BioSig software by playing the test signal from the transducer and recording the r eceived level from a hydrophone (HTI 96 min; sensitivity -164 dbV/ Pa from 2 to 37 kHz) that was positioned 2 cm away from it and 10 cm below the surface. Evoked potential magnitudes were calc ulated by performing a Hanning window on the (EFR) signal followed by a 2,048-point Fast Fourier Transform (FFT) with a measuring amplitude at 600Hz. Equivalent s ound pressure levels (SPL) were estimated for evoked potential measurement amplitudes within each carrier frequency and at each transducer location by dividing previously obtained input/out put functions (Mann et al., 2005) by 20 log. 146

PAGE 160

Results Voluntary evoked potential measurements were obtained from Hugh and Buffett for positions on the body that were cranial to the scapula while underwater on March 3, 2004. Restrained evoked potential measuremen ts were obtained for positions over the entire body while in air for Hugh on September 17, 2003, Mo on February 17, 2004, and Bock on February 22, 2005. All four subjects, whether pos itioned in air or in wate r, produced EFRs at the 600 Hz AM rate with both the 15 kHz (Figure 4.9) and 24 kHz carrier sign als (Figure 4.10). Figure 4.9. A typical auditory evoked potential found at the 600 Hz AM frequency using the 15 kHz carrier. Frequency is defined along the X-axis and signal strength (dB volts) is defined along the Y-axis. Results are from Bocks rib 5 cm caudal to the scap ula. Top shows complete measurement, bottom shows same AEP signal magnified. 147

PAGE 161

Figure 4.10. A typical auditory evoked potential found at the 600 Hz AM frequency using the 24 kHz carrier. Frequency is defined along the X-axis and signal strength (dB volts) is defined along the Y-axis. Results are from the hinge of Bocks chin. Top shows complete measurement, bottom shows same AEP signal magnified. A total of 17 in-water AEPs were measur ed for Hugh, with nine derived from the 15 kHz carrier frequency a nd eight from the 24 kHz carrier frequency (Figure 4.11 & 4.12). EFR amplitudes and SPLs were determined for positions both dorsal and ventral to the eye, the hinge and center of the lower jaw, both meatuses and the vertebrae located between the scapulas. The pattern of EFR amplitudes and equivalent SPLs varied between carrier frequencies but were higher w ith the 24 kHz carrier frequenc y for all positions except 5 cm dorsal to the left eye, the left meatus and the vertebrae between the scapulas. Measurements found with the 15 kHz carrier frequency ranged from 1.4 nV / 118.2 dB (5 148

PAGE 162

cm ventral to the right eye) to 8.2 nV / 126.6 dB (5 cm dorsal to the left eye). Measurements found with the 24 kHz carrier fr equency ranged from 2 nV / 119 dB (left meatus and 5 cm ventral to the right eye) to 7.7 nV / 126 dB (5 cm ventral to the left eye). Hugh In-Water AEPs 0 1 2 3 4 5 6 7 8 95 cm Dorsal to Left Eye 5 cm Ventral to Left Eye 5 cm Ventral to Right Eye 5 cm Ventral to Right Eye Left Jaw Hinge Right Lower Jaw Left MeatusRight Meatus Vertebrae Between ScapulaLocationEFR Amplitude (nV) 15 kHz 24 kHz Figure 4.11. In-water auditory evoked potential response measurements for Hugh. Body locations the transducer was positioned on are lis ted along the X-axis and envelope following response amplitudes (nV) are defined along the Y-axis. Measurements obtained with the 15 kHz carrier are denoted by triangles while those obtained with the 24 kH z carrier are demarked by squares. *Note that some locations may be duplicated or absent for one or both carrier frequencies. 149

PAGE 163

Hugh In-Water Estimated SPL Differences114 116 118 120 122 124 126 1285 cm Dorsal to Left Eye 5 cm Ventral to Left Eye 5 cm Ventral to Right Eye 5 cm Ventral to Right Eye Left Jaw Hinge Right Lower Jaw Left MeatusRight Meatus Vertebrae Between ScapulaLocationSound Pressure Level (dB) 15 kHz 24 kHz Figure 4.12. Estimated in-water sound pressure level measurements for Hugh. Body locations the transducer was positioned on are listed along the X-axis and the sound pressure levels (dB) are defined along the Y-axis. Measurements obtained with the 15 kHz carrier are denoted by triangles while those obtained with the 24 kHz carrier are demarked by squares. *Note that some locations may be duplicated or absent for one or both carrier frequencies. A total of 16 in-water AEPs and SPLs were measured for Buffett, eight each from the 15 and 24 kHz carrier frequencies (Fi gure 4.13 & 4.14). EFR amplitudes and SPLs were determined for positions both dorsal and ventral to the eye, the jaw hinges and center of the lower jaw, both meatuses and th e vertebrae located between the scapulas. The pattern of EFR amplitudes and SPLs between carrier frequencies was similar; those found with the 24 kHz carrier frequency were higher or equal (center of lower jaw and left meatus) to those found with the 15 kHz frequency carrier. Measurements found with the 15 kHz carrier frequency ranged from 0.3 nV / 11.9 dB (5 vertebrae between the scapulas) to 4.8 nV / 122.4 dB (hi nge of the left jaw). Meas urements found with the 24 kHz carrier frequency ranged from 1.4 nV / 119.7 dB (hinge of right jaw) to 7.1 nV / 125.3 dB (hinge of left jaw). 150

PAGE 164

Buffett In-Water AEPs0 1 2 3 4 5 6 7 85 cm Dorsal to Eye 5 cm Ventral to Eye Left Jaw Hinge Left Lower Jaw Right Jaw Hinge Left MeatusRight MeatusVertebrae Between ScapulaLocationEFR Amplitude (nV) 15 kHz 24 kHz Figure 4.13. In-water auditory evoked potential re sponse measurements for Buffett. Body locations the transducer was positioned on are lis ted along the X-axis and envelope following response amplitudes (nV) are defined along the Y-axis. Measurements obtained with the 15 kHz carrier are denoted by triangles while those obtained with the 24 kH z carrier are denoted by squares. Buffett In-Water Estimated SPL Differences112 114 116 118 120 122 124 1265 cm Dorsal to Eye 5 cm Ventral to Eye Left Jaw Hinge Left Lower Jaw Right Jaw Hinge Left MeatusRight MeatusVertebrae Between ScapulaLocationSound Pressure Level (dB) 15 kHz 24 kHz Figure 4.14. Estimated in-water sound pressure le vel measurements for Buffett. Body locations the transducer was positioned on are listed along the X-axis and the sound pressure levels (dB) are defined along the Y-axis. Measurements obtained with the 15 kHz carrier are denoted by triangles while those obtained with the 24 kHz carrier are denoted by squares. 151

PAGE 165

EFR amplitudes of the 7 common positions between Hugh and Buffett, when tested in-water and using both the 15 and 24 kHz carrier frequencies, were plotted together to examine similarities and discrepancies (Figure 4.15). Patterns within the 15 kHz carrier were inconsistent and amplitudes va ried by as much as 6.5 nV (5 cm dorsal to the eye) at all locations except the righ t meatus which only had a 0.7 nV difference between subjects. Patterns within the 24 kHz carrier were id entical or similar (within 0.6 nV) at all locations except the hinge of the left jaw which had a 4.1 nV difference between subjects (note that the data for the left lower jaw with the 24 kHz carrier is absent for Hugh). Hugh and Buffett In-Water AEP Comparison0 1 2 3 4 5 6 7 8 9 5 cm Dorsal to Eye 5 cm Ventral to Eye Left Jaw HingeLeft Lower JawLeft MeatusRight MeatusVertebrae Between ScapulaLocationEFR Amplitude (nV) Buffett 15 kHz Buffett 24 kHz Hugh 15 kHz Hugh 24 kHz Figure 4.15. In-water auditory evoked potential response measurement comparison of common positions for Hugh and Buffett. Body locations the transduc er was positioned on are listed along the X-axis and envelope following response amplitudes (nV) are defined along the Y-axis. Measurements obtained with the 15 kHz carrier are denoted by teal triangles fo r Hugh and maroon triangles for Buffett while those obtained with the 24 kHz carrier are denoted by teal squares for Hugh and maroon squares for Buffett. A total of 18 in-air AEPs and SPLs were measured for Bock, nine each from the 15 and 24 kHz carrier frequencies (Figure 4.16 & 4.17). EFR amplitudes and SPLs were determined for positions caudal to the nares, th e hinge of the jaw, ve ntral to the eye, the 152

PAGE 166

meatus, on the zygomatic process ventral to the meatus, caudal to the meatus, on the lateral side of the ribs caudal to the scapul a and on the last vertebrae of the tail. The pattern of EFR amplitudes and SPLs between carrier frequencies was similar; those found with the 24 kHz carrier frequency were higher than those found with the 15 kHz frequency carrier for all positions excep t ventral to the meat us on the zygomatic process and 10 cm caudal to the meatus. M easurements found with the 15 kHz carrier frequency ranged from 2 nV / 119 dB (ventral to the meatus on the zygomatic process) to 9.7 nV / 128.5 dB (hinge of the jaw). M easurements found with the 24 kHz carrier frequency ranged from 0.7 nV / 117.4 dB (10 cm caudal to the m eatus) to 15.1 nV / 135.2 dB (hinge of jaw). Bock In-Air AEPs 0 2 4 6 8 10 12 14 16Caudal to Nares Midline Jaw Hinge5 cm Ventral to Eye MeatusZP 5 cm Ventral to Meatus 5 cm Caudal to Meatus 10 cm Caudal to Meatus Lat. Ribs Caudal to Scapula Last Vertebrae on TailLocationEFR Amplitude (nV) 15 kHz 24 kHz Figure 4.16. In-air auditory evoked potential response measurements for Bock. Body locations the transducer was positioned on are lis ted along the X-axis and envelope following response amplitudes (nV) are defined along the Y-axis. Measurements obtained with the 15 kHz carrier are denoted by triangles while those obtained with the 24 kHz carrier are denoted by squares. ZP denoted zygomatic process and Lat. denoted lateral. 153

PAGE 167

154 Bock In-Air Estimated SPL Differences105 110 115 120 125 130 135 140 Caudal to Nares Midline Jaw Hinge5 cm Ventral to Eye MeatusZP 5 cm Ventral to Meatus 5 cm Caudal to Meatus 10 cm Caudal to Meatus Lat. Ribs Caudal to Scapula Last Vertebrae on TailLocationSound Pressure Level (dB) 15 kHz 24 kHz Figure 4.17. Estimated in-air sound pressure level m easurements for Bock. Body locations the transducer was positioned on are listed along the X-axis and the sound pressure levels (dB) are defined along the Yaxis. Measurements obtained with the 15 kHz carrier are denoted by triangles while those obtained with the 24 kHz carrier are denoted by squares. ZP denoted zygomatic process and Lat. denoted lateral. A total of 32 in-air AEPs and SPLs were measured for Hugh at only the 15 kHz carrier frequency (Figure 4.18 & 4.19). EFR amplitudes and SPLs were determined for positions surrounding the jaw, eye, meatus, zygomatic process, scapula, the vertebrae midway down the length of the sp ine and the tail, as well as every 10 cm along the lateral side of the ribcage. AEPs were found at every position tested but amplitudes were highest at the vertebrae located midway dow n the length of the tail ( 15 nV / 135 dB) and the spine (13.3 nV / 132.9 dB). Amplitudes were lowest at the lateral side of the ribs, 10 cm caudal to the scapula (1.4 nV / 118.2 dB) and the meatus (2.3 nV / 119.3 dB).

PAGE 168

Hugh In-Air AEPs0 2 4 6 8 10 12 14 16C e n te r o f R ig h t L o w e r J a w C a u d a l E n d o f L o w e r J a w J a w H in g e 5 c m D o r s a l & C a u d a l to J a w H in g e 5 c m V e n tr a l a n d 5 c m C r a n ia l to E y e 5 c m V e n tr a l to E y e 5 c m C a u d a l to E y e M e a t u s Z P V e n tr a l to M e a tu s Z P 5 c m V e n t r a l t o M e a tu s Z P 1 0 c m V e n tr a l to M e a tu s L a t R ib s C r a n ia l t o S c a p u l a S c a p u la L a t R ib s C a u d a l t o S c a p u l a L a t R ib s 1 0 c m C a u d a l t o S c a p u l a L a t R ib s 2 0 c m C a u d a l t o S c a p u l a L a t R ib s 3 0 c m C a u d a l t o S c a p u l a L a t R ib s 4 0 c m C a u d a l t o S c a p u l a L a t R ib s 5 0 c m C a u d a l t o S c a p u l a L a t R ib s 6 0 c m C a u d a l t o S c a p u l a L a t R ib s 7 0 c m C a u d a l t o S c a p u l a V e r te b r a e M i d w a y D o w n B a c k L a t R ib s 8 0 c m C a u d a l t o S c a p u l a L a t R i b s 9 0 c m C a u d a l t o S c a p u l a L a t R ib s 1 0 0 c m C a u d a l to S c a p u l a L a t R ib s 1 1 0 c m C a u d a l to S c a p u l a L a t R ib s 1 2 0 c m C a u d a l to S c a p u l a L a t R ib s 1 3 0 c m C a u d a l to S c a p u l a L a t R ib s 1 4 0 c m C a u d a l to S c a p u l a L a t e r a l T o r s o 1 0 c m C r a n ia l t o C r e a s e L a t e r a l T o r s o C r e a s e V e r te b r a e M id w a y D o w n T a i l LocationEFR Amplitude (nV) 15 kHz Figure 4.18. In-air auditory evoked potential response measurem ents for Hugh. Body locations the transducer was positioned on are listed along the X-axis and envelope following response amplitudes (nV) are defined along the Y-axis. All measurements were obtained with the 15 kHz carri er frequency. ZP denoted zygomatic process and Lat. denoted lateral. 155

PAGE 169

156 0 Hugh In-Air SPL Differences105 110 115 120 125 130 135 140C e n t e r o f R ig h t L o w e r J a w C a u d a l E n d o f L o w e r J a w J a w H in g e 5 c m D o r s a l & C a u d a l t o J a w H i n g e Z P 5 c m V e n t r a l a n d 5 c m C r a n ia l t o E y e Z P Ve n t r a l to E y e 5 c m C a u d a l t o E y e M e a t u s Z P V e n tr a l t o M e a tu s Z P 5 c m V e n t r a l t o M e a tu s Z P 1 0 c m V e n tr a l to M e a t u s L a t R i b s C r a n ia l t o S c a p u l a S c a p u l a L a t R i b s C a u d a l t o S c a p u l a L a t R i b s 1 0 c m C a u d a l t o S c a p u l a L a t R i b s 2 0 c m C a u d a l t o S c a p u l a L a t R i b s 3 0 c m C a u d a l t o S c a p u l a L a t R i b s 4 0 c m C a u d a l t o S c a p u l a L a t R i b s 5 0 c m C a u d a l t o S c a p u l a L a t R i b s 6 0 c m C a u d a l t o S c a p u l a L a t R i b s 7 0 c m C a u d a l t o S c a p u l a Ve r t e b r a e M id w a y D o w n B a c k L a t R i b s 8 0 c m C a u d a l t o S c a p u l a L a t R i b s 9 0 c m C a u d a l t o S c a p u l a L a t R i b s 1 0 0 c m C a u d a l t o S c a p u l a L a t R i b s 1 1 0 c m C a u d a l t o S c a p u l a L a t R i b s 1 2 0 c m C a u d a l t o S c a p u l a L a t R i b s 1 3 0 c m C a u d a l t o S c a p u l a L a t R i b s 1 4 0 c m C a u d a l t o S c a p u l a L a t e r a l T o r s o 1 0 c m C r a n i a l to C r e a s e L a t e r a l T o r s o C r e a s e Ve r t e b r a e M i d w a y D o w n T a i l LocationSound Pressure Level (dB) 15 kHz Figure 4.19. Estimated in-air sound pressure level measurements for Hugh. Body locations the transducer was positioned on are listed along the X-axis and the sound pressure levels (dB) are defined along the Y-axis. Measurements were obtained only with the 15 kHz carrier. ZP denoted zygomatic process and Lat. denoted lateral.

PAGE 170

A total of 27 in-air AEPs and SPLs were measured for Mo, with 14 derived from the 14 kHz carrier frequency and 13 from the 24 kHz carrier frequency (Figure 4.20 & 4.21). EFR amplitudes and SPLs were determ ined for positions on the center of the lower jaw, areas surrounding the eye and meatus, the last vertebrae on the tail, as well as positions along the lateral side of the ribcage. The pattern of EFR amplitudes and SPLs varied and amplitudes with the 24 kHz carrier frequency were not always higher than those found with the 15 kHz frequency carrier. Amplitudes with the 24 kHz carrier frequency was higher for all positions except the center of the lower jaw, 5 cm ve ntral to the eye, 15 cm ventral and 10 cm caudal to the eye, and on the last vertebrae of the tail. Measurements found with the 15 kHz carrier frequency ranged from 0.6 nV / 117.2 dB (10 cm ventral and caudal to eye) to 7.7 nV / 126 dB (meatus). Measurements found with the 24 kHz carrier frequency ranged from 1.1 nV / 117.9 dB (center of lower jaw) to 9 nV / 127.6 dB (5 cm caudal to the meatus). 157

PAGE 171

Mo In-Air AEPs0 1 2 3 4 5 6 7 8 9 10 Center of Right Lower Jaw 5 cm Ventral to Eye 5 cm Ventral to Eye 5 cm Caudal to Eye 10 cm Ventral and Caudal to Eye 15 cm Ventral and 10 cm Caudal to Eye 15 cm Ventral and 10 cm Caudal to Eye 5 cm Cranial to Meatus Meatus5 cm Caudal to Meatus Lat. Ribs Caudal to Scapula Lat. Ribs 10 cm Caudal to Scapula Lat. Ribs 1/2 way along ribcage Last Vertebrae on TailLocationEFR Amplitude (nV ) 15 kHz 24 kHz Figure 4.20. In-air auditory evoked potential response measurements for Mo. Body locations the transducer was positioned on ar e listed along the X-axis and envelope following response amplitudes (nV) are defined along the Y-axis. Measurements obtained with the 15 kHz carrier are de noted by triangles while those obtained with the 24 kHz carrier are denoted by squares. ZP denoted zygomatic process and Lat. denoted lateral. *Note that some locations may be duplicated or absent for one or both carrier frequencies. 158

PAGE 172

159 Mo In-Air Estimated SPL Differences 112 114 116 118 120 122 124 126 128 130Center of Right Lower Jaw 5 cm Ventral to Eye 5 cm Ventral to Eye 5 cm Caudal to Eye 10 cm Ventral and Caudal to Eye 15 cm Ventral and 10 cm Caudal to Eye 15 cm Ventral and 10 cm Caudal to Eye 5 cm Cranial to Meatus Meatus5 cm Caudal to Meatus Lat. Ribs Caudal to Scapula Lat. Ribs 10 cm Caudal to Scapula Lat. Ribs 1/2 way along ribcage Last Vertebrae on TailLocationSound Pressure Level (dB) 15 kHz 24 kHz Figure 4.21. Estimated in-air sound pressure level measurements for Mo. Body locations the transducer was positioned on are l isted along the X-axis and the sound pressure levels (dB) are defined along the Y-axis. Measur ements obtained with the 15 kHz carrier are denoted by triangle s while those obtained with the 24 kHz carrier are denoted by squares. ZP denoted zygomatic pr ocess and Lat. denoted lateral. *Note that some locations may be duplicated or absent for one or both carrier frequencies.

PAGE 173

EFR amplitudes of the 12 positions that were common between at least two of the three subjects, when tested in-air, were pl otted together to examine similarities and discrepancies between Hugh, Mo & Bock (Fi gure 4.22). Comparisons were made with the 15 and 24 kHz carrier frequencies for Mo and Bock, but only included the 15 kHz carrier frequency with Hugh. Patterns within the 15 kHz carrier were similar and had less than 1 nV difference only when positioned 5 cm ventral to the eye between Mo and Bock, the meatus between Hugh and Bock, 5 cm caudal to the meatus between Mo and Bock, and on the last vertebrae on the tail between Mo and Bock. Patterns within the 24 kHz carrier were similar (within 0.9 nV) at all locations except 5 cm ventral to the eye which had a 3.7 nV difference between Mo and Bock. 160

PAGE 174

Hugh, Mo & Bock In-Air AEP Comparisons0 2 4 6 8 10 12 Center of Lower Jaw Jaw Hinge5 cm Ventral to Eye 5 cm Caudal to Eye Right MeatusZP 5 cm Ventral to Meatus 5 cm Caudal to Meatus Lat. Ribs Caudal to Scapula Lat. Ribs 10 cm Caudal to Scapula Lat. Ribs 1/2 way along ribcage Last Vertebrae on Tail LocationEFR Amplitude (nV) Hugh 15 kHz Mo 15 kHz Mo 24 kHz Bock 15 kHz Bock 24 kHz Figure 4.22. In-air auditory evoked potential response measurement comparison of common positions for Hugh, Mo, and Bock. Bod y locations the transducer was positioned on are listed along the X-axis and envelope following response amplitudes (nV) are defined along the Y-axis. Me asurements obtained with the 15 kHz carrier are denoted by teal triangles for Hugh and red triangles for Mo and blue triangles for Bock while those obtained wi th the 24 kHz carrier are denoted by red squares for Mo and blue squares for Bock (24 kHz carrier not used with Hugh). ZP denoted zygomatic process and Lat. denoted lateral. 161

PAGE 175

EFR amplitudes of the 4 positions common between at least one subject when measured in-air and at least one subject when measured in-water, we re plotted together (Figure 4.23). Comparisons were made with the 15 and 24 kHz carrier frequencies in-air for Mo and Bock but only the 15 kHz carrier frequency with Hugh, and the 15 and 24 kHz carrier frequencies for Hugh and Buffet in-water. Amplitudes were typically greater within the 15 and 24 kHz carrier frequencies when tested in-air with up to a 9 nV and 12.3 nV differences (respectiv ely) found at the jaw hinge between Bock and Buffett. Bock and Hugh demonstrated greater in-air am plitudes at the jaw hinge and 5 cm ventral to the eye, however the remaining amplitude s were more symmetrical between subjects, positions, carrier frequencies and the medium in which testing was conducted. In-Air vs. In-Water AEP Comparisons0 2 4 6 8 10 12 14 16 Center of Lower Jaw Jaw Hinge 5 cm Ventral to Eye Right MeatusLocationEFR Amplitude (nV) In-Air Hugh 15 kHz In-Air Mo 15 kHz In-Air Mo 24 kHz In-Air Bock 15 kHz In-Air Bock 24 kHz In-Water Hugh 15 kHz In-Water Hugh 24 kHz In-Water Buffett 15 kHz In-Water Buffett 24 kHz Figure 4.23. In-water vs.in-air auditory evoked potential response measuremen t comparison of common positions for Hugh, Buffett, Mo and Bock. Body lo cations the transducer wa s positioned on are listed along the X-axis and envelope following response amplitudes (nV) are defined along the Y-axis. Measurements obtained with the in-air 15 kHz carrier are denoted by solid teal triangles for Hugh, solid red triangles for Mo and solid blue triangles for Bock, while those obtained with the in-air 24 kHz carrier are denoted by solid teal squares for Hugh, solid red squares for Mo and solid blue squares for Bock. Measurements obtained with the in-wat er 15 kHz carrier are denoted by sh aded teal triangles for Hugh and shaded maroon triangles for Buffett, while those obtained with the in-water 24 kHz carrier are denoted by shaded teal squares for Hugh and shaded maroon squares for Buffett. 162

PAGE 176

Discussion This investigation used auditory evoked potential techniques to evaluate if sound conduction pathways, outside of the typical mammalian pinna-to-cochlea conduit, may be used by manatees to detect sounds. A 600 Hz signal was used to amplitude modulate 15 and 24 kHz carrier tone bursts. AEP responses were collected as test signals were delivered in-water to several positions on the subjects heads, and in -air to a variety of positions on the subjects heads and torsos. Transducer positions were coded by video analysis and results were derived through the compilation and organization of the data already collected. Hugh and Buffett had been previously c onditioned to remain motionless in the water at a target for a prior AEP investiga tion (Mann et al., 2005) This training was capitalized upon to obtain voluntary in-water AEP measurements from these subjects as test signals were presented to several areas on the head that could be easily reached from the side of the exhibit. In-air measurements were obtained from three subjects, Hugh, Mo and Bock, as the test signal(s) were pr esented to many of the same areas on their heads in addition to numerous ar eas on their torsos. For the in-air testing, the 15 and 24 kHz carrier frequencies were used with Mo and Bock however only the 15 kHz carrier frequency was used with Hugh. Overall findings demonstrated that all subjects, regardless being positioned in air or in water, produced AEPs with the 15 and 24 kHz carriers at the 600 AM rate at every position the transducer was placed on their bodies. Sound pressure levels mirrored 163

PAGE 177

amplitude variations between body positions a nd did not attenuate at positions further away from the meatus. This is an interesting phenomenon, considering some measurements were obtained at a distance of 3 m from the meatus and supports the AEP results which suggest that sound waves may be received across the entire body. Results found between subjects, body posit ions, carrier frequencies and in-air vs. in-water mediums should be interpreted with caution. Data were collected in an unsystematic manner that did not permit many identical comparisons to be made. Results found between subjects when AEPs were collected in-water demonstrated identical or similar amplitudes (within 0.6 nV) at six of the seven common positions using the 24 kHz carrier, however amplitudes were inconsistent for all but one of the common positions using the 15 kHz carrier. Re sults between subjects when collected inair showed that Bock and Mo had similar amplitudes (within 0.9 nV) at four of their five common positions with both the 15 and 24 kH z carriers, however Hugh had only one similar amplitude (0.9 nV difference) with Bo ck out of the nine positions that were common to Hugh, Mo, and Bock. Results found between the 15 and 24 kH z carrier frequencies generally demonstrated that amplitudes were higher with the 24 kHz carrier when tested in both inair and in-water. It is important to recall th at all amplitudes are re presented in nanovolts, so these differences are quite small when compared to differences found with cetacean AEP investigations that were measured in microvolts (Cook et al., 2006). Patterns between carriers at identical positions were highly variable between subjects, with Buffett and Bock showing more similarities than Hugh and Mo. 164

PAGE 178

Of the possible four positions that were common to the four subjects in both the in-air and in-water mediums, amplitude was typically greater within the 15 and 24 kHz carrier frequencies when tested in-air. In comparison to the other subjects, Bock had a considerably greater amplitude with both car riers at one of the four locations and Hugh had greater amplitude with the 15kHz carrier at two of the four locations when tested inair. All remaining amplitudes were more sy mmetrical between subjects, positions, carrier frequencies and the medium in which testing was conducted. The first hypothesis posited that auditory evoked potenti als would be of greater magnitude at the position of the external auditory meatus than at the zygomatic process. Anatomical investigation of the zygomatic proc ess has shown that it lies ventral to the external auditory opening (EAO) and extends cranially to a point about half way between the EAO and the eye. It is connected to th e squamosal bone, which in turn, is connected to the tympanoperiotic complex (Figure 4.3) (K etten et al., 1999; Ch apla et al., 2007). The zygomatic process is a bony sponge filled with lipids and blood vessels that has less density and rigidity than other bones (F awcett, 1942; Caldwell & Caldwell, 1985; Domning & de Buffrenil, 1991; Ketten et al., 1999), however the lipids it contains are not considered acoustic fats (Ames et al., 2002) It may be that the composition of the zygomatic process and its geometric position relative to the ear bones may serve as an acoustic channel. The manatees inter-cranial tissue arrangeme nts near the external auditory meatus has been found to have an impedance similar to that of water with a minimal amount of bone for sound waves to reflect off of (Fi gure 4.4) (Chapla et al., 2007). Sound waves arriving from azimuth angles between 45o and 90o and elevation angles between ~43o and 165

PAGE 179

73o may be able to propagate through the soft tissue surrounding the external auditory meatus area to stimulate the tympanic membrane directly. The AEP amplitudes (nV) for areas surroundi ng the external auditory meatus and zygomatic process were defined from the da ta collected and averaged by the positions defined in the hypothesis (Table 4.1). Note that data for some positions may have been duplicated or never obtained between subjects. Positions defined as caudal to the meatus are likely to have been presented over the edge of the zygomatic process or the squamosal bone as it extends dorsally and ca udally and were included in the zygomatic process average. The one measurement obtained cranial to the meatus was included in the meatus average. Measurements from the jaw hinge should also be considered as the transducer may have been positioned on the ventral edge of the zygomatic process, however this positioning was not certai n and it was averaged on its own. Results did not support the hypothesis that suggested AEPs would be of greater magnitude at the position of the external auditory meatus than at the zygomatic process. The averaged AEPs were found to be iden tical and suggest that one area does not represent a stronger sound c onduction pathway. Interestingl y, averaged measurements obtained from the jaw hinge were of greater magnitude than those found in the areas of the zygomatic process or meatus. 166

PAGE 180

Table 4.1. AEP amplitudes (nV) obtained from the areas surrounding the external auditory meatus and zygomatic process for all subjects. The averaged AEP amplitudes for each position are listed in the end column. Note that data for some positions may be duplicated or absent. In-Water In-Air Hugh Buffett Hugh Mo Bock Location 15 kHz 24 kHz 15 kHz 24 kHz 15 kHz 15 kHz 24 kHz 15 kHz 24 kHz Meatus 4.6 2 1.4 1.4 2.3 7.7 8 3.2 7.7 Meatus 3.6 4.8 2.9 4.4 5 cm Cranial to Meatus 6 4.3 5 cm Caudal to Meatus 4 9 3.6 8.1 10 cm Caudal to Meatus 3.1 0.7 ZP / Just Ventral to Meatus 9.7 ZP / 5 cm Ventral to Meatus 3.6 2 1.2 ZP / 10 cm Ventral to Meatus 2.9 4.3 Jaw Hinge 1.8 3 4.8 7.1 9.7 15.1 Jaw Hinge 0.7 2.6 5.6 The second hypothesis stated that su bjects would demonstrate greater evoked potentials at points along the ve rtebral column and lateral ribs that are more than 20 cm caudal to the scapula than those located crania l to, at the level of, or up to 20 cm caudal to the scapula. The manatee lung structure is composed of two pleural cavities that are supported ventrally by hemidiaphragms which lie in a horizontal plane along the bodys length (Rommel & Reynolds, 2000). The manat ees hemidiaphragms are attached to the first three ribs and extend back to the 26th vertebra which accounts for approximately 40% of the manatees total body length. This anatomical arrangement is not typical for mammals but is similar to that found with many reptiles that use their lungs and skeletal system as a conduit for acoustic vibratory s timulation to be tran sferred to the ears (Hartline, 1971; Lenhardt, 1982; Lenhardt et al., 1983). This arrangement provides a huge surface area for acoustic sound wave vibr ations to be received by the manatees lungs, ribs and/or spinal column and tr ansmitted to the skull and ear bones. The manatees large scapulas however, may pr ovide a deaf-spot that sound waves will reflect from rather than be received through. 167

PAGE 181

The AEP amplitudes (nV) for points along th e vertebral column and lateral ribs were defined from the data collected and av eraged by the areas defined in the hypothesis (Table 4.2). Results showed that evoked potentials from positions along the vertebral column and lateral ribs that were more than 20 cm caudal to the scapula (6.3 nV) were greater than those located in cranial to, at the level of, or up to 20 cm caudal to the scapula (4.4 nV). These findings substantiate the hypothesis and suggest that the spinal column and lateral ribs, positioned caudal to the scapula, may serve as an important conduit for sound transmission to the ear bones. It is important to note that the averages, particularly those that incl ude positions greater than 20 cm caudal to the scapula, are composed of numerous single data points from Hugh. The evoked potentials obtained from Hugh were charact eristically found to be greater than those obtained from other subjects and these results shoul d be interpreted cautiously. 168

PAGE 182

Table 4.2. AEP amplitudes (nV) obtained from points along the vertebral column and lateral ribs that are more than 20 cm caudal to the scapula than those located cranial to, at the level of, or up 20 cm caudal to the scapula. The averaged AEP amplitudes for each area are listed in the end column. Note that data for some positions may be duplicated or absent and many positions have only one data point. In-Water In-Air Hugh Buffett Hugh Mo Bock Location 15 kHz 24 kHz 15 kHz 24 kHz 15 kHz 15 kHz 24 kHz 15 kHz 24 kHz Lat. Ribs Cranial to Scapula 11.9 Scapula 9.9 Lat. Ribs Just Caudal to Scapula 5.8 2.4 6.9 3.6 6.8 Lat. Ribs 10 cm Caudal to Scapula 1.4 3.4 4 Lat. Ribs 20 cm Caudal to Scapula 2.2 Vertebrae Between Scapula 3.7 2.1 0.3 1.5 4.4 Lat. Ribs 30 cm Caudal to Scapula 4.8 Lat. Ribs 40 cm Caudal to Scapula 7 Lat. Ribs 50 cm Caudal to Scapula 4.3 0.9 1.8 Lat. Ribs 60 cm Caudal to Scapula 5.2 Lat. Ribs 70 cm Caudal to Scapula 4.9 Lat. Ribs 80 cm Caudal to Scapula 7.6 Lat. Ribs 90 cm Caudal to Scapula 5.6 Lat. Ribs 100 cm Caudal to Scapula 8.3 Lat. Ribs 110 cm Caudal to Scapula 7.5 Lat. Ribs 120 cm Caudal to Scapula 9.9 Lat. Ribs 130 cm Caudal to Scapula 7 Lat. Ribs 140 cm Caudal to Scapula 7.8 Vertebrae Midway Down Back 13.3 Vertebrae Midway Down Tail 15 Last Vertebrae on Tail 4.3 3.9 4 3.2 6.3 The information gained from this stu dy demonstrates how AEP techniques may be used to evaluate the existence of sound conduction pathway outside of the traditional pinna-to-cochlea conduit. EFRs can be isolat ed using in at the 600 Hz AM rate with 15 and 24 kHz carrier signals, although those in the 24 kHz carrier generally produced potentials with greater amplitudes. AEPs were found for all positions tested with all four subjects, regardless of their being positioned in air or in water. These results suggest that manatees have evolved a way to compensate for their occluded extern al auditory meatus; the exact means by which they have accomp lished this however, remains a conundrum that requires further investigation. AEP techniques offer potential insight for solving this 169

PAGE 183

puzzle and future AEP investigations should incorporate systematic and multiple measurements of identical positions on each subjects body. 170

PAGE 184

References Cited Ames, A.L., Van Vleet, E.S. & Reynolds J.E. (2002). Comparison of lipids in selected tissues of the Florida manatee (Order Sirenia) and bottlenose dolphin (Order Cetacea; Suborder Odontoceti). Comparative Biochemistry and Physiology Part B 132, 625-634. Aroyan, J.L. (1996). Three-Dimensional Numerical Simulation of Biosonar Signal Emission and Reception in the Common Dolphin. Unpublished Ph.D. Dissertation, University of California, Santa Cruz. Bauer, G.B., Colbert, D.E., Gaspard, J.C., Littlefield, B. & Fellner, W. (2003). Underwater visual acuity of Florida manatees ( Trichechus manatus latirostris.). International Journal of Comparative Psychology 16, 130-142. Bauer, G.B., Gaspard III, J.C., Colbert, D.E., Leach, J.B. & Reep, R. (2005, March). Tactile Discrimination of Textures by Florida Manatees, Trichechus manatus latirostris. Paper presented at the 12th annual International Conference on Comparative Cognition, Melbourne, Florida. Brill R.L., Moore P.W.B. & Dankiewicz L.A. (2001) Assessment of dolphin ( Tursiops truncatus ) auditory sensitivity and hearing loss using jawphones. Journal of the Acoustical Society of America 109:1717-1722 Bullock, T.H., Domning, D.P. & Best, R.C. (1980). Evoked potentials demonstrate hearing in a manatee ( Trichechus inunguis ). Journal of Mammalogy 61 (1), 130133. Bullock, T.H., OShea, T.J. & McClune, M.C. (1982). Auditory evoked potentials in the West Indian manatee (Sirenia: Trichechus manatus). Journal of Comparative Physiology, 148, 547-554. Caldwell, D.K. & Caldwell, M.C. (1985). Manatees: Trichechus manatus (Linnaeus, 1758); Trichechus senegalensis (Link, 1795); and Trichechus inunguis (Natterer, 1883). In S.H. Ridgway & R. Harrison (Eds.) Handbook of Marine Mammals; The Sirenians and Baleen Whales Vol. 3. Academic Press, London, pp. 33. Chapla, M., Nowacek, D., Rommel, S. & Sadler, V. (2007). CT scans and 3D reconstructions of Florida manatee ( Trichechus manatus latirostris ) heads and ear bones. Hearing Research 228, 123-135. 171

PAGE 185

Colbert, D. (2005). Sound Localization Abilities of Two Florida Manatees, Trichechus manatus latirostris. Unpublished Masters Thesis, Un iversity of South Florida, Tampa, FL. Colbert, D.E., Fellner, W., Bauer, G.B., Mani re, C. & Rhinehart, H. (2001). Husbandry and research training of two Florida manatees, Trichechus manatus latirostris. Aquatic Mammals 27 (1), 16-23. Cook, M.L.H. (2006). Behavioral and Auditory Evoked Potential (AEP) Hearing Measurements in Odontocete Cetaceans Unpublished Masters Thesis, University of South Florida, Tampa, FL. Cook, M.L.H., Varela, R.A., Goldstein, J.D ., McCulloch, S.D., Bossart, G.D., Finneran, J.J., Houser, D. & Mann, D.A. (2006). Beaked whale auditory evoked potential hearing measurements. Journal of Comparative Physiology A 192, 489-495. Corwin, J.T., Bullock, T.H. & Schweitzer, J. (1982). The auditory brain stem response in five vertebrate classes. Electroencephalography and Clinical Neurophysiology, 54, 629-641. Dolphin, W.F. (1996). Auditory evoked responses to amplitude modulated stimuli consisting of multiple envelope components. Journal of Comparative Physiology A 179, 113-121. Dolphin, W.F. (1997). The envelope following response to multiple tone pair stimuli. Hearing Research 110, 1-14. Domjan, M. (1998). The principles of l earning and behavior. Pacific Grove, CA: Brooks-Cole Publishing Company. Domning, D.P. (2001). Evolution of the Sire nia and Desmostylia. In Mazin, J.M. & de Buffrenil, V. (Eds.) Secondary Adaptation of Tetrapods to Life in the Water Verlag, Munchen, pp. 151. Domning, D.P. & de Buffrenil, V. (1991). Hydr ostasis in the Sireni a: quantitative data and functional interpretations. Marine Mammal Science 7 (4), 331. Fawcett, D.W. (1942). The amedulla ry bones of the Florida Manatee ( Trichechus latirostris ). American Journal of Anatomy 71, 271. Ferraro, J.A. & Durrant, J.D. (1994). Auditory evoked potentials: overview and basic principles. In Katz, J (Ed) Handbook of Clinical Audiology Williams and Wilkens, London, pp317-338. Fleischer, G. (1978). Evolutionary Principles of the Mammalian Middle Ear Springer, Berlin. 172

PAGE 186

Gelfand, A.G. (2004). Hearing: An Introduction to Psychological and Physiological Acoustics, 4th Edition. Marcel Dekk er, New York, pp 98-102. Gerstein, E. (1999). Psychoacoustic Evaluations of the West Indian manatee (Trichechus manatus latirostris) Unpublished Doctoral Di ssertation, Florida Atlantic University, Boca Raton, FL. Gerstein, E., Gerstein, L., Forsythe, S. & Bl ue, J. (1999). The unde rwater audiogram of the West Indian manatee (Trichechus manatus ). Journal of the Acoustical Society of America 105, 3575-3583. Goold, J.C. & Clarke, M.R. (2000). Sound velocity in the head of the dwarf sperm whale, Kogia sima with anatomical and functional discussion. Journal of the Marine Biological Association of the U.K. 80, 535. Hartline, P.H. (1971). Physiological basis for detection of sound and vibration in snakes. Journal of Experimental Biology 54, 349. Hemila, S., Nummela, S. & Reuter, T. (1999). A model of th e odontocete middle ear. Hearing Research 133, 82. Ketten, D.R. (1992). The marine mammal ear: Specializations for aquatic audition and echolocation. In R. Fay, W. Webster & A. Popper (Eds.) The biology of Hearing Springer-Verlag, pp 717-754. Ketten, D.R., Odell, D.K. & Domning, D.P. ( 1992). Structure, function,and adaptation of the manatee ear. In J.A. Thomas, R. A. Kastelein & A.Y. Supin (Eds.) Marine Mammal Sensory Systems. Plenum Press, NY, pp.77-79. Kirkpatrick, B., Colbert, D.E., Dalpra, D., Ne wton, E.A.C., Gaspard, J., Littlefield B. & Manire, C.A. (2002). Florida Red Tides, Manatee Brevetoxicosis, and Lung Models. In K. A. Steidinger, J. H. La ndsberg, C. R. Tomas, and G. A. Vargo (Eds.) Xth International Conference on Harmful Algae Florida Fish and Wildlife Conservation Commission and Intergove rnmental Oceanographic Commission of UNESCO. Klishen, V.O., Diaz, R.P., Popov, V.V. & Supin, A.Y. (1990). Some characteristics of hearing of the Brazilian manatee, Trichechus inunguis. Aquatic Mammals 16 (3), 139-144. Lenhardt, M.L. (1982). Bone conduction in turtles. Journal of Auditory Research 22, 153. 173

PAGE 187

Lenhardt, M.L., Bellmund, S., Byles, R.A., Harkins, S.W. & Musick, J.A. (1983). Marine turtle reception of bone-conducted sound. Journal of Auditory Research 23, 119 125. Lucas, J.R., Freeberg, T.M., Krishnan, A. & L ong, G.R. (2002). A comparative study of avian auditory brainstem responses: Correlations with phylogeny and vocal complexity and seasonal effects. Journal of Comparative Physiology A 188, 981992. Manire, C.A., Walsh, C.J., Rhinehart, H.L., Colbert, D.E., Noyes, D.R. & Luer, C.A. (2003). Alterations in blood and urine parameters in two Florida manatees, Trichechus manatus latirostris from simulated conditi ons of release following rehabilitation. Zoo Biology 22, 103-120. Mann, D., Colbert, D.E., Gaspard, J.C. III, Casper, B., Cook, M.L.H., Reep, R.L. & Bauer, G.B. (2005). Temporal resolution of the Florida manatee ( Trichechus manatus latirostris ) auditory system. Journal of Comparative Physiology 191, 903-908. Mast, T.D. (2000). Empirical relationships between acoustic parameters in human soft tissues. Acoustic Research Letters Online, 1 (2), 37. Nowacek, D.P., Casper, B.M., Wells, R.W ., Nowacek, S.M. & Mann, D.A. (2003). Intraspecific and geographic variat ion of West Indian manatee ( Trichechus manatus spp.) vocalizations. Journal of the Acoustical Society of America 114 (1), 66-69. Nummela, S. (1995). Scaling of the mammalian middle ear. Hearing Research 85, 18 30. Nummela, S., Wagar, T., Hemila, S. & Reuter T. (1999). Scaling of the cetacean middle ear. Hearing Research 133, 71. Pearce, J.M. & Dickinson, A. (1975). Pa vlovian counterconditioning: changing the suppressive properties of s hock by association with food. Journal of Experimental Psychology: Animal Behavior Proceedings 1, 170. Popov, V.V. & Supin, A.Y. (1990). Electrophys iological studies on hearing in some cetaceans and a manatee. In J.A. T homas & R. A. Kastelein (Eds.), Sensory Abilities of Cetaceans: Laboratory and Field Evidence NY: Plenum Press. Richardson, W., Greene, C., Malme, C. & Thompson, D. (1995). Marine Mammals and Noise. San Diego, CA: Academic Press. Ridgway, S.H., Bullock, T.H., Carder, D.A., Seeley, R.L., Woods, D. & Galambos, R. (1981). Auditory brainstem response in dophins. Neurobiology 78, 1943-1947. 174

PAGE 188

Rommel, S.A. & Lowenstine, L. J. (2000). Gross and microsc opic anatomy. In Dierauf, L.A. & Gulland, M.D. (Eds.), CRC Handbook of Marine Mammal Medicine, Second Edition. CRC Press, Boca Raton, FL, pp 129-158. Rommel, S. & Reynolds, J.E. III (2000). Diaphr agm structure and function in the Florida manatee ( Trichechus manatus latirostris ). Anatomical Record 259(1):41. Ruggero, M.A. & Temchin, A.N. (2002). The roles of the external, middle, and inner ears in determining the bandwidth of hearing. Proceedings of the National Academy of Sciences of the United States, 99.20: 13206(5). Soldevilla, M.S., McKenna, M.F., Wiggins, S.M., Shadwick, R.E., Cranford, T.W. & Hildebrand, J.A. (2005). Cuviers beaked whale ( Ziphius cavirostris ) head tissues: physical properties and CT imaging. Journal of Expe rimental Biology, 208, 2319. Szymanski, M.D., Bain, D.E., Kiehl, K. Pennington, S., Wong, S. & Henry, K.R. (1999). Killer whale ( Orcinius orca) hearing: Auditory brain stem responseand behavioral audiograms. Journal of the Acoustical Society of America 106, 1134-1141. Tonndorf, J. (1966). Bone conduction: st udies in experimental animals. Acta Otolaryngologica, Stockholm Supplement, 213, pp. 1. Varanasi, U. & Malins, D.C. (1971). Unique lipids of the porpoise ( Tursiops gilli ): Differences in triacyl glycerols and wax esters of acoustic (mandibular canal and melon) and blubber tissues. Biochimica Biophysica Acta 231 (2), 415. Yost, W.A. (2000). Fundamentals of hearing: An introduction 4th Edition. Academic Press, San Diego, CA, pp.71-77. 175

PAGE 189

Chapter Five: The Importance of Understa nding the Auditory Sensory System of the Florida Manatee, Trichechus mana tus latirostris: Concluding Remarks The Florida manatee ( Trichechus manatus latirostris ) is protected by both the Marine Mammal Protection Act (1 972) and the Endangered Spec ies Act (1973). It is the only marine mammal known to annually migrate from turbid saltw ater habitats to freshwater springs during the winter months and reverse th is pattern during the summer months (Reynolds & Wilcox, 1986; Reynolds & Odell, 1991). The manatee is a semisocial species, often grazing or traveling alone, but able to find conspecifics for socialization or reproductive purposes (Reynolds, 1979). Manatee mortality caused specifically by watercraft remains relatively stable ranging between 19-31% of the annual mortalities (Table 1.1) (Florida Fish and Wildlife Research Institute, 2007). While field research has provided info rmation about the manatees social structure, habitat usage, and annual migratory behaviors. Sensory biology investigations have indicated that the manat ees auditory system almost certainly plays a principal role in their ability to find one another, determine directionality and avoi d danger in their vast habitat. The manatees hearing range has b een found to be quite wide, spanning between 0.2-40 kHz (Bullock et al., 1980, 1982; Popov & Supin, 1990; Gerstein et al., 1999; Mann et al., 2005) and perhaps as high as 60 kHz (Klishen et al., 1990). Previous investigations have shown that manatees have the capacity to localize broadband and tonal stimuli of various durations and leve ls within a 0.2-20 kHz frequency range from 176

PAGE 190

four locations in the frontal 180 of the azimuth plane (Gerstei n, 1999; Colbert 2005). These capabilities indicate that manatees are able to detect and localize, at least from some directions, conspecific vocalizati ons which typically range between 2.5.9 kHz (Nowacek, et al., 2003) and r ecreational boat engine noise which generally range between 0.01 kHz (Gerstein, 2002; Richardson et al., 1995). Given this information, it could be assumed that the manatee auditory system is constructed and functions similarly to that of typical mammalian sp ecies, however, this assumption is inaccurate. The manatees external and middle ears have been found to be unusually structured. The external auditory meatus is occluded and separated from the tympanic membrane making it an unlikel y channel for sound transmission, the tympanoperiotic complex is located intracrania lly but not ossified to the skull, and the ossicles are massive (Kette n, 1992; Ketten et al., 1992; Chapla et al., 2007). Although much has been learned about the manatees auditory system, a plethora of questions remain. The primary objective of this dissertation was to address some of these uncertainties by ascertaining if manatees have the ability to determine sound source directionality within all 360o of the azimuth plane and to identify the possible means by which they do so. Chapter Two investigated the manatees abil ities to localize test signals that were systematically varied across dimensions of bandwidth, duration and level as they originated from 45o angles within all 360o of the azimuth plane at a distance of 3.05 m. Test signals included a tonal stimulus and three broadband stimuli, one of which spanned a wide range of frequencies, one that was restricted to higher frequencies that had wavelengths shorter than a manatees interaural time distances, and one that was 177

PAGE 191

restricted to lower frequencies that had wa velengths longer than a manatees interaural time distances (Table 2.3). Both subjects performed well above the 12.5% chance level for all of the broadband stimuli, however performance decreas ed dramatically (14% and 20%) with the 4 kHz tonal stimulus. Both were able to lo calize the broadband stim uli at a short duration that prohibited head movement and over a large level range and little contralateral confusion occurred. Accuracy decreased w ith the shorter durati on and when signals originated from the posterior locations. Erro rs were typically located at the speakers neighboring the test speaker but became somewhat more dispersed when they originated from 135o and 180o for Buffett and more widely dispersed when they originated from 180o and 225o for Hugh. Although accuracy was lower when signals came from behind them, the subjects were able to localize from these positions (with the exception of Hugh at 180o) without the aide of visually orienti ng towards these areas and front to back confusions were minimal. Results from this study indicate that manatees have good directional hearing capabilities, at least with broadband s ounds which are typical in their natural environment, in all azimuth angles relati ve to their bodies, including those in the frequency range of boats and conspecifics. Their ability to localize may be a function of visual orientation responses when sounds or iginate in their visual field (Brown, 1994; Heffner, 1997), however it is likely that some type of interaural cue(s) are also interpreted to assist with discriminations from all angles but particularly from those outside of their visual field. 178

PAGE 192

Chapter Three investigated how different fr equencies of a test signal, presented in different locations on the azimuth plane, are filtered by the manatees head and torso by measuring head/body related transfer functi ons. Head/body related transfer functions were determined by subtracting the averaged animal present FFTs from the averaged animal absent FFTs and the magnitude of inte raural level differences was derived for all frequencies in addition to specific 0.2-1.5, 0.2-5 and 18-30 kHz bands of frequencies. These are the first head/ body related transfer function data collected for any Sirenian species and results demonstrated th at interaural level di fferences (ILD) were present for all frequencies as a function of source location. ILDs were of the greatest magnitude with frequencies in the 18-30 kH z noise band which had wavelengths shorter than the manatees intercochlear distance, however the 0.2-1.5 kHz noise band, which had wavelengths longer than the manatees intermeatal distance, produced greater ILDs than the wider 0.2-5 kHz noise band of low freq uencies. Test signals originating at 90o and 270o provided the greatest ILD cues however, ILDs were greater when the signal originated behind the subjects than when it originated in front of them. Results from this study suggest that manatees are ab le to utilize ILD cues to localize sounds via head and torso filtering effects. The amplified ILD cues produced by their unique body shape when sounds originate from the lateral and posterior angles of the body may compensate for the inability to utilize visual orienting responses when sounds originated from these angles. Although ILDs are typically found with wavelengths shorter than a sp ecies interaural distances, manatees may also have the ability to utilize ILDs with wavelengths longer than thei r interaural distances. 179

PAGE 193

Chapter Four utilized auditory evoked pot ential (AEP) techniques to investigate the possible existence of sound conduction pathways that manatees may use as a means to overcome outer ear limitations and benefit from the middle ears unique structure and geometry. AEPs were collected in-water for positions on the manatee head, and in-air for positions on the head and torso using 15 a nd 24 kHz carrier tone bursts that were amplitude modulated (AM) with a 600 Hz rate. Results demonstrated that AEPs were found at every position the transducer was placed on their bodies, regardless of whether th ey were positioned in water or air. AEP amplitudes were usually greater with the 24 kHz carrier however patterns between carriers at identical body positions were highly variable between subjects. Data from the 24 kHz carrier showed that identical or si milar AEP amplitudes were found at six of seven positions that were common between subjects when tested in-water, and four of five positions when tested in-air. Data from the 15 kHz carrier showed that AEP amplitudes were inconsistent for all but one of the common positions when tested inwater, and two subjects (out of the three) had similar amplit udes at four of five common positions, however the third subject had similar amplitudes at only one of nine common positions. Evoked potentials, averaged together from positions along the vertebral column and lateral ribs that were more than 20 cm caudal to the scapula, were greater than those averaged together from positions at and dorsal to the meatus, those averaged from positions along the zygomatic process, and those averaged from positions along the vertebral column and lateral ribs that we re cranial to 20 cm behind the scapula. Results indicate that manatees demonstrat e AEPs from all parts of their body and have evolved a means to compensate for thei r occluded external auditory meatus. The 180

PAGE 194

increased AEP amplitudes found along the vertebral column and lateral ribs which were more than 20 cm caudal to the scapula may suggest that the unique structure of the manatees plural cavities a nd ribs may facilitate bone conduction to the inner ear, however the data included in this stu dy should be considered cautiously. The information gained from the individual experiments presented in Chapters 2, 3, and 4 provide valuable knowledge about how the manatees ability to localize sounds of different frequencies, durati ons and intensities in all 360o of the azimuth plane, how interaural intensity cues ma y facilitate this ability, and how sounds may be received across their entire body and not through only one primary sound conduction pathway. Additional consideration should be given however to the information that can be learned when the results of all three experiments are taken into account together. Investigations with many terrestrial speci es have shown that ILDs provide cues for sound localization, but they are typi cally only found with frequencies having wavelengths shorter than in termeatal or intercochlear distances (Brown & May, 1990; Brown, 1994; Blauert, 1997). Anatomical i nvestigation of the manatee ear has shown that the tympanoperiotic complex is located intracranially, thereby creating a shorter intercochlear distances than woul d be found if it were located extracranially as is the case with cetaceans (Ketten et al., 1992; Chapla et al., 2007). Using the manatees intercochlear measur ement, it appears that only frequencies of 18 kHz or higher would provide useful ILDs. Results from the sound localization study however, demonstrate that manatees are ab le to localize frequencies well below this range in all azimuth directions. Head/body rela ted transfer measuremen ts show that ILDs can be found for all frequencie s as a function of sound sour ce location and suggest that 181

PAGE 195

ILD cues are being used to f acilitate this localization abi lity. Comparisons between ILD magnitudes found from contralateral speaker pairs when noise bands of 18-30 kHz (wavelengths < intercochlear distance ), 0.2-5 kHz and 0.2-1.5 kHz (both with wavelengths > intermeatal distance) and the sound localization sele ction distributions found with the 18-24 kHz and 0.2-1.5 broadband test signals demonstrate how ILDs may facilitate localization (Fi gures 5.1, 5.2, 5.3 & 5.4). ILD ma gnitudes were greatest when the test signal originated at 90o and 270o and were smallest when they originated at 0o. These results suggest that ILDs play an impor tant role in sound lo calization especially when sounds originate from the lateral and poste rior half of the manatees large body. 182

PAGE 196

0.2-1.5 kHz 18-30 kHz Figure 5.1 Contralateral comparison between ILD magnitudes and sound localization selection distributions at 0o and 180o. ILD differences are shown in 0.2-1.5 kHz, 0.2-5 kHz, and 18-30 kHz noise bands. Sound localization selection distributions are shown for the 0.2-1.5 kHz and 18-24 kHz test signals. Hugh 0.46 dB 0o 0.05 dB 0.79 dB Buffett -0.22 dB Hugh -0.04 dB 180o -4.00 dB -1.51 dB Buffett -1.06 dB 0.2-1.5 kHz 18-24 kHz 45 315 270 90 135 225 ((0)) 180 0.2-1.5 kHz 18-24 kHz 45 315 270 90 135 225 0 ((180)) 45 315 270 90 135 225 0 ((180)) 45 315 270 90 135 225 ((0)) 180 183

PAGE 197

Figure 5.2 Contralateral comparison between ILD magnitudes and sound localization selection distributions at 45o and 225o. ILD differences are shown in 0.2-1.5 kHz, 0.2-5 kHz, and 18-30 kHz noise bands. Sound localization selection distributions are shown for the 0.2-1.5 kHz and 18-24 kHz test signals. Hugh 0.83 dB 45o 0.86dB 2.26 dB Buffett 0.24 dB Hugh -3.61 dB 225o -4.50 dB -2.39 dB Buffett -2.68 dB 0.2-1.5 kHz 18-30 kHz 0.2-1.5 kHz 18-24 kHz 0.2-1.5 kHz 18-24 kHz ((45)) 315 270 90 135 225 0 180 45 315 270 90 135 ((225)) 0 180 ((45)) 315 270 90 135 225 0 180 45 315 270 90 135 ((225)) 0 180 184

PAGE 198

Figure 5.3 Contralateral comparison between ILD magnitudes and sound localization selection distributions at 90o and 270o. ILD differences are shown in 0.2-1.5 kHz, 0.2-5 kHz, and 18-30 kHz noise bands. Sound localization selection distributions are shown for the 0.2-1.5 kHz and 18-24 kHz test signals. Hugh 2.32 dB 90o 3.34 dB 4.07 dB Buffett 2.54 dB Hugh -1.54 dB 270o -6.14 dB -4.72 dB Buffett -2.59 dB 0.2-1.5 kHz 18-30 kHz 0.2-1.5 kHz 18-24 kHz 0.2-1.5 kHz 18-24 kHz 45 315 270 ((90)) 135 225 0 180 45 315 ((270)) 90 135 225 0 180 45 315 ((270)) 90 135 225 0 180 45 315 270 ((90)) 135 225 0 180 185

PAGE 199

Figure 5.4 Contralateral comparison between ILD magnitudes and sound localization selection distributions at 315o and 135o. ILD differences are shown in 0.2-1.5 kHz, 0.2-5 kHz, and 18-30 kHz noise bands. Sound localization selection distributions are shown for the 0.2-1.5 kHz and 18-24 kHz test signals. Hugh -0.79 dB 315o -1.13 dB -2.80 dB Buffett -0.76 dB Hugh 2.36 dB 135o 2.41 dB 2.88 dB Buffett 1.70 dB 0.2-1.5 kHz 18-30 kHz 0.2-1.5 kHz 18-24 kHz 0.2-1.5 kHz 18-24 kHz 45 ((315)) 270 90 135 225 0 180 45 315 270 90 ((135)) 225 0 180 45 315 270 90 ((135)) 225 0 180 45 ((315)) 270 90 135 225 0 180 186

PAGE 200

ILD magnitudes were also greatest with the high frequency 18-30 kHz stimuli, however the 0.2-1.5 kHz stimuli had larger ma gnitudes than the 0.2-5 kHz stimuli (Figure 3.15). These results are surprising given that the wavelengths of these frequencies are longer than the manatees intera ural distances. In humans, an occlusion effect occurs when the external auditory canal is bloc ked and bone-conducted sound vibrations cannot radiate out of the ear canal (Gelfand, 2004). The sound vibrations are instead reflected back toward the tympanic membrane and ha ve been found to boost sound pressure in the ear canal by 20 dB with frequencies below 500 Hz. Since the manatees external auditory meatus is occluded, this effect ma y serve an important role for amplifying ILDs with frequencies longer than th eir interaural distances. When considering the anatomical means by which manatees are able to detect and localize sounds, Chapla et al. (2007) have suggested that the large 3.1 x 104 mm3 airspace which ventrally surrounds each cochlear capsule and the independent hypotympanic recesses may play an important role in both task s. Investigations have demonstrated that tympanoperiotic complex of cetaceans are isolat ed and shielded from the dorsal, medial and posterior surfaces of the skull by air-fi lled sinuses which provide reflective barriers to the passage of sounds between the ears (Dudok van Heel, 1962; Fleischer, 1980; Oelschlger, 1986; Houser et al., 2004). The separation of the tympanoperiotic complexes from each other and the skull li kely facilitates sound localization by enhancing interaural level differences resulting from the shadowing effects of the cetaceans head and torso (H ouser et al., 2004). Aroyan ( 1996) found that airspaces within soft tissues, such as those found su rrounding the manatee cochlear capsules, act as acoustic energy reflectors. Similar to cetaceans, it may be that the resonance vibrations 187

PAGE 201

found in the airspaces surrounding the manatees cochlear cap sules and the independent hypotympanic recesses serve to isolate the middle ears from one another. This anatomical design may facilitate sound localiz ation by providing the means to interpret ILD cues caused by the shadowing effects of the manatees head and torso. The unusual arrangement of the manatees pleural cavities, wh ich are supported by two independent hemidiaphragms, instead of the typical single mammalian diaphragm may also play an important role in determ ining the means by which manatees detect and localize sounds. The airspaces of the lungs may also act as acoustic energy reflectors and the resonance vibrations found in these airsp aces may be transmitted to the ribs and/or spinal column and to the skull and ear bones (Chapla et al., 2007). Rommel & Reynolds (2000) have further suggested that the separation of the hemidiaphragms may provide additional cues to aid in sound localization. The results from the investigations in cluded in this manuscript have provided critical information about the manatees ab ility to localize sounds and the means by which it may accomplish this to find conspecifics, determine directionality and avoid danger in its vast habitat. This information augments our knowledge of how the manatees auditory sensory system assimilate s information and reac ts to environmental stimuli and should be considered when maki ng conservation management decisions about this endangered species. Additional knowledge about the manatee auditory sensory system however, could be gained through future investigations. Manatee sound localization investigations (Gerstein, 1999; Colbert, 2005; Chapter 2) and head/body related transfer functi on measurements (Chapter 3) have only investigated within the azimuth plane to date Controlled boater approaches to free188

PAGE 202

ranging manatees demonstrated that subjects typically increased their swim speed and oriented to deeper channel waters as boats approached (Nowacek et al., 2004). Localization ability assessments and head/body related transfer function measurements obtained in the vertical plane may demonstrat e that ILD cues hold equal or more salience to those in the azimuth plane. This inform ation may provide insight into the importance of determining sound source directionality above and below the animals and may partially explain why they increase de pth in response to boater approaches. Auditory evoked potential techniques have b een shown to be a valuable tool for assessing possible sound conducti on pathways (Chapter 4). Future investigations should incorporate systematic and multiple meas urements of identical positions on each subjects body. In addition, th e resonance frequencies for the manatees lungs and air spaces surrounding the cochlear capsules should be measured to determine if these anatomical characteristics facilitate mo re pronounced sound conduction pathways. Finally, anatomical investiga tions of the human inner ear have demonstrated that the tonotopic organization of the cochlea plays an important role in determining the range of hearing (Ruggero & Temchin, 2002). Ba ndwidth of hearing in the cochlea is determined by the tonotopic frequency map found along the lengt h of the basilar membrane by which higher frequencies st imulate the base stereocilia and lower frequencies stimulate the apex stereocilia. Studies with ceta ceans have shown that functional morphometric analyses of basila r membrane measurements and auditory ganglion cell density counts within the coch lea provide a reliable estimate of hearing sensitivity (Wever, et al., 1971; Parks, et al., 2004). Mann et al. (2005) measured the manatees auditory system temporal re solution through envelope following response 189

PAGE 203

techniques and found they have a temporal resolution of 600 Hz which is approximately half the 1,200 Hz resolution of cetaceans (D olphin et al., 1995; Supin & Popov, 1995) but over ten times the 50 Hz resolution of hu mans (Kuwada et al., 1986). Although the manatees resolution is half th at of the dolphins, it is sti ll impressive considering that manatees cannot echolocate which is what th e dolphins high temporal resolution is thought to be an adaptation for. The anatom y and physiology of the manatees inner ear has not been investigated to date, but may provide informa tion about how their auditory ganglion cell density may be correlated to their range of heari ng, frequency resolution abilities, and possibly an increased sensitivity to timing accuracy. 190

PAGE 204

References Cited Aroyan, J.L. (1996). Three-dimensional numerical si mulation of biosonar signal emission and reception in the common dolphin. Unpublished Ph.D. Dissertation, University of California, Santa Cruz, CA. Blauert, J. (1997). Spatial Hearing: The Psychophysics of Human Sound Localization Massachusetts Institute of Technology Press, Massachusetts. Bullock, T.H., Domning, D.P. & Best, R.C. (1980). Evoked potentials demonstrate hearing in a manatee ( Trichechus inunguis ). Journal of Mammalogy 61 (1), 130133. Bullock, T.H., OShea, T.J. & McClune, M.C. (1982). Auditory evoked potentials in the West Indian manatee (Sirenia: Trichechus manatus). Journal of Comparative Physiology, 148, 547-554. Brown, C.H. (1994). Sound Localization. In R.R. Fay and A.N. Popper (Eds.) Comparative Hearing: Mammals New York, Springer-Verlag, pp. 57-97. Brown, C.H. & May, B.J. (1990). Sound localiza tion and binaural processes. In M.A. Berkley and W.C. Stebbins (Eds.) Comparative Perception Vol I, New York, John Wiley and Sons, pp. 247-284. Chapla, M., Nowacek, D., Rommel, S. & Sadler, V. (2007). CT scans and 3D reconstructions of Florida manatee ( Trichechus manatus latirostris ) heads and ear bones. Hearing Research 228, 123-135. Colbert, D. (2005). Sound Localization Abilities of Two Florida Manatees, Trichechus manatus latirostris. Unpublished Masters Thesis, University of South Florida, Tampa, FL. Dolphin W.F., Au, W.W.L., Nachtigall, P.E. & Pawloski, J. (1995). Modulation rate transfer functions to low-frequency ca rriers in three spec ies of cetaceans. Journal of Comparative Physiology A 177, 235. Dudok van Heel, W. H. (1962) Sound and cetaceans. Netherlands Journal of Sea Research 1 (4), 407. 191

PAGE 205

Fleischer, G. ( 1980). Low-frequency receiver of the middle ear in mysticetes and odontocetes. In R.G. Busnel and J.F. Fish (Eds.) Animal Sonar Systems, Plenum, New York, pp. 891. Florida Fish & Wildlife Res earch Institute (2007). Annual mortality rates http://research.myfwc.com/featur es/view_article.asp?id=12084 Gelfand, A.G. (2004). Hearing: An Introduction to Psychological and Physiological Acoustics, 4th Edition. Marcel Dekk er, New York, pp 98-102. Gerstein, E. (1999). Psychoacoustic Evaluations of the West Indian manatee (Trichechus manatus latirostris) Unpublished Doctoral Di ssertation, Florida Atlantic University, Boca Raton, FL. Gerstein, E., Gerstein, L., Forsythe, S. & Bl ue, J. (1999). The unde rwater audiogram of the West Indian manatee (Trichechus manatus ). Journal of the Acoustical Society of America 105, 3575-3583. Gerstein, E.R. (2002). Manatees, bioacoustics and boats. American Scientist 90, 154163. Heffner, R.S. (1997). Comparative study of sound localization and its anatomical correlates in mammals. Acta Otolaryngologica 532, 46-53. Houser, D.S., Finneran, J., Carder, D., Van Bonn, W., Smith, C., H oh, C., Mattrey, R. & Ridgway, S. (2004). Structural and f unctional imaging of bottlenose dolphin ( Tursiops truncatus) cranial anatomy. Journal of Expe rimental Biology 207, 3657. Ketten, D.R. (1992). The marine mammal ear: Specializations for aquatic audition and echolocation. In R. Fay, W. Webster & A. Popper (Eds.) The Biology of Hearing Springer-Verlag, pp 717-754. Ketten, D.R., Odell, D.K. & Domning, D.P. ( 1992). Structure, function,and adaptation of the manatee ear. In J.A. Thomas, R. A. Kastelein & A.Y. Supin (Eds.) Marine Mammal Sensory Systems. Plenum Press, NY, pp.77-79. Klishen, V.O., Diaz, R.P., Popov, V.V. & Supin, A.Y. (1990). Some characteristics of hearing of the Brazilian manatee, Trichechus inunguis. Aquatic Mammals 16 (3), 139-144. Kuwada, S., Batra, R., Maher, V.L. (1986) Scalp potentials of normal and hearingimpaired subjects in response to sinu soidally amplitude-modulated tones. Hearing Research 21179. 192

PAGE 206

Mann, D., Colbert, D.E., Gaspard, J.C. III, Casper, B., Cook, M.L.H., Reep, R.L. & Bauer, G.B. (2005). Temporal resolution of the Florida manatee ( Trichechus manatus latirostris ) auditory system. Journal of Comparative Physiology 191, 903-908. Nowacek, D.P., Casper, B.M., Wells, R.W ., Nowacek, S.M. & Mann, D.A. (2003). Intraspecific and geographic variat ion of West Indian manatee ( Trichechus manatus spp.) vocalizations. Journal of the Acoustical Society of America 114 (1), 66-69. Nowacek, S.M., Wells, R.S., Owen, E.C.G., Speakman,T.R., Flam, R.O. & Nowacek, D.P. (2004). Florida manatees, Trichechus manatus latirostris respond to approaching vessels. Biological Conservation 119, 517-523. Oelschlger, H.A. (1986). Comparative mor phology and evolution of the otic region in toothed whales (Cetacea, Mammalia). American Journal of Anatomy, 177, 353 368. Parks, S., Ketten, D., OMalley, J., & Arruda, J. (2004). Hearing in the North Atlantic right whale: Anatomical predictions The Journal of the Acoustical Society of America 115 (5), 2442. Popov, V.V. & Supin, A.Y. (1990). Electrophys iological studies on hearing in some cetaceans and a manatee. In J.A. T homas & R. A. Kastelein (Eds.), Sensory Abilities of Cetaceans: Laboratory and Field Evidence NY: Plenum Press. Richardson, W., Greene, C., Malme, C. & Thompson, D. (1995). Marine Mammals and Noise. San Diego, CA: Academic Press. Reynolds, J.E. III. (1979). The semisocial manatee. Natural History 88 (2), 44-53. Reynolds, J.E. & Odell, D.K. (1991). Manatees and Dugongs (192 pp.). NY: Checkmark Books. Reynolds, J.E. III, & Wilcox, J.R. (1986). Distribution and abundance of the West Indian manatee, Trichechus manatus around selected Florida power plants following winter cold fronts: 1984-1985. Biological Conservation 38, 103-113. Rommel, S. & Reynolds, J.E. III (2000). Diaphr agm structure and function in the Florida manatee ( Trichechus manatus latirostris ). Anatomical Record 259(1):41. Ruggero, M.A. and A.N. Temchin. (2002). The roles of the external, middle, and inner ears in determining the bandwidth of hearing. Proceedings of the National Academy of Sciences 99 (20), 13206. Supin, A.Y., Popov, V.V. (1995). Envelope-following response and modulation transfer function in the dolphins auditory system. Hearing Research 92, 38. 193

PAGE 207

Wever, E., McCormick, J., Palin, J. & Ridgewa y, S. (1971). The cochlea of the dolphin, Tursiops truncatus : Haircells and ganglion cells. Proceedings of the National Academy of Science 68 (12), 2908-2912. 194

PAGE 208

Appendices 195

PAGE 209

Appendix A: RPvds language used to ge nerate signals used in the manatee sound localization experiment. In the manatee sound localization expe riment, RPvds language was used to generate signals and record trial informati on via the button box that interfaced with the computer and signal generation equipment. Specific RPvds language was developed to generate each subjects call to station (Figure A-1), the init iation of each trial (Figure A2), each subjects secondary bridge reinforcement if correct (Figure A-3) and documentation of incorrect selections (Figure A-4). Figure A-1. RPvds langua ge used to generate each subjects call to station. 196

PAGE 210

Appendix A: (Continued) Figure A-2. RPvds language used to generate the initiation of each trial. Figure A-3. RPvds language us ed to generate each subjects secondary bridge when correct. 197

PAGE 211

Appendix A: (Continued) Figure A-4. RPvds language used to document when incorrect selections were made. 198

PAGE 212

Appendix B: Computer protocols used for setting up the manatee sound localization and head/body related tran sfer function experi mental conditions. A graphical user interface, programmed in Visual C, was designed to run each phase of the experimental conditions (Figure B-1). A drop-down subject menu was designed to distinguish which subject was being tested, and this selection automatically referenced and played that animals stat ioning and reinforcement tones throughout the block. A notes section allowed any comments to be digitally recorded relative to that block. The set-up section defined how many speak er locations were to be tested, how many trials were to be run from each of thos e speakers, and how many of the test sounds could be played from the same location in a row. In addition, broa d-band noise bursts or tonal signals were defined as were the fre quency range to be tested, the sound duration, the dB level and if the sounds were to be automatically digi tally recorded. The correct experimental conditions were incorporated for each portion of the session, including the warm-up, testing, and cool-down trials. In all portions of a sound localization and head/body related transfer function session, ei ght speaker locations and a maximum of two trials in a row per locati on were held constant. In the warm-up trials, one trial was set up per speaker for a total of eight trials. The noise button was selected and the frequency range was defined from 24,000-200 Hz. The sound duration was defined as 3 seconds. In the testing trials, all of the settings for the conditions being tested were defined and the number of trials per speaker was cha nged from one to two, for a total of sixteen trials. These settings were maintained until five blocks were co mpleted per condition. 199

PAGE 213

Appendix B: (Continued) In the cool-down trials, one trial was set up per speaker for a tota l of eight trials. The noise button was selected and the frequency range was defined from 24,000 to 200 Hz. The sound duration was defined as 3 seconds. In the head/body related transfer function experiment, two trials were set up per speaker for a total of sixteen trials. The noise button wa s selected and the frequency range was defined from 30,000 to 200 Hz. The sound duration was defined as 3 seconds. The speaker section provided informa tion about which speaker location each test sound was played from. If needed, a manual switching check box was included, which allowed the Data Recorder to select the location of the test sound to be played, rather than the randomized loca tion generated by the program. The status section defined and digitall y recorded how many trials had been completed within the block, and of those, how many were correct and how many were wrong. The start button initiated the block of sixteen trials once the subject and conditions were defined, and th e stop button was used only if the block had to be ended prior to the completion of the twelve trials. 200

PAGE 214

Appendix B: (Continued) Figure B-1: The graphical user interface screen (progr ammed in Visual C) used to setup the experimental conditions and automatically download the results into an Excel file during the testing sessions. 201

PAGE 215

Appendix C: Data recording protocols used to document each sound localization session. All of the sessions general information was documented on a data sheet (Figure C-1). This included the date and the identities of the Test Trainer, Data Recorder and Stationing Trainer. Specific information was documented for all for portions of the session (warm-up, test blocks and cool-down trials) per subj ect including frequency(ies), duration and level of the stimulus, start and end times of the session and each block, the location of each trials test sound, if the subjec t was correct or incorr ect and, if incorrect, the location the subject erroneously selected. In the data sheet shown, the 4 kHz tonal probes were randomly distributed on four of the sixteen trials of each block and are denoted by the shaded cells. Additional information was included for each test block including the video tape number and counter start and stop times, the number of times the te st subject left or attempted to leave in that block, the number of times the test subject was interrupted by the other animal, the amount of time the other animal was on task a nd the test subjects behavioral rating from a scale of one to five, where one indicated that the animal did very poorly and was not able to complete the task and five indicated that he did an excellent job. A comment section was also provided to add additional information if needed. 202

PAGE 216

Appendix C: (Continued) Sound Localization Task Date: Hugh Buffett W A R M U P S W A R M U P S Frequency 0.2-24 kHz Duration 3000 ms Frequency 0.2-24 kHz Duration 3000 ms Trainer: Speaker Correct? Comments Trainer: Speaker Correct? Comments # Speakers: 8 1 # Speakers: 1 Begin Time: 2 Begin Time: 2 Times Left: 3 Times Left: 3 Leave Attempts: 4 Leave Attempts 4 #Interrupted 5 #Interrupted: 5 Buf On Task: 6 Hugh On Task: 6 End Time: 7 End Time: 7 Rating: 8 Rating: 8 T E S T I N G T E S T I N G Frequency 4kHz Probes (0.2-24) Duration -200 ms Frequency 4kHz Probes (0.2-24) Duration -200 ms Trainer: Speaker Correct? Comments Trainer: Speaker Comments # Speakers: 8 1 3 # Speakers: 1 3 Begin Time: 2 5 Begin Time: 2 5 Tape #: 3 4 Tape #: 3 4 Tape Start: 4 7 Tape Start: 4 7 Times Left: 5 2 Times Left: 5 2 6 6 6 6 Leave Attempts: 7 1 Leave Attempts: 7 1 8 1 8 1 #Interrupted: 9 3 #Interrupted: 9 3 10 0 10 0 Buf On Task: 11 2 Hugh On Task: 11 2 12 7 12 7 Tape End: 13 4 Tape End: 13 4 End Time: 14 5 End Time: 14 5 Rating: 15 6 Rating: 15 6 16 0 16 0 COOL-DOWNS COOL-DOWNS Frequency 0.2-24 kHz Duration 3000 ms Frequency 0.2-24 kHz Duration 3000 ms Speaker Correct? Comments Speaker Correct? Comments Times Left: 1 Times Left: 1 Leave Attempts: 2 Leave Attempts 2 #Interrupted 3 #Interrupted: 3 Rating: 4 Rating: 4 = Tonal Probe @ 4 kHz = Tonal Probe @ 4 kHz Figure C-1. The tank-side data-recording sheet used to document each session. 203

PAGE 217

Appendix D: MATLAB program used to determine and chart the manatee head/body related transfer functions. The data collected for the manatee head/body related transfer function investigation was analyzed and charted via the MatLab program below. Four data analyses were conducted per subject. The fi rst compared FFT ratios received at the left and right hydrophones as a func tion of sound source location with animal absent and animal present conditions. The second determined head/bo dy related transfer functions by subtracting the averaged animal present FFTs from the averaged animal absent FFTs. The third determined the magnitude of interaural level differences for all frequencies. The final analysis determined the magnitude of intera ural level differences for specific 0.2-1.5, 0.2-5, and 18-30 kHz bands of frequencies. MatLab Program: %Code for calculating the FFT's for each fftpts segment of the kept signals fftpts=488; %200Hz frequency resolution (srate/150) cc1=[]; cc2=[]; directoryname = uigetdir; cd(directoryname); filenames = dir(directoryname); % allows a directory to pop up to select all files from animal present trials for n=3:length(filenames); load(filenames(n).name); npts=length(channel1chunk); x=floor(npts/fftpts); cc1=[cc1 channel1chunk(1:x*fftpts)]; cc2=[cc2 channel2chunk(1:x*fftpts)]; end %[filename, pathname] = uigetfile({'*.mat'},'File Selector Manatee Present'); % allows a directory to pop up to select one animal present file %cd (pathname); %load (filename); aa1=[]; aa2=[]; directoryname = uigetdir; 204

PAGE 218

Appendix D: (Continued) cd(directoryname); filenames = dir(directoryname); % allows a directory to pop up to select all files from animal absent trials for n=3:length(filenames); load(filenames(n).name); npts=length(channel1); x=floor(npts/fftpts); aa1=[aa1 channel1(1:x*fftpts)]; aa2=[aa2 channel2(1:x*fftpts)]; end srate=97656.25; %sample rate binwidth=srate/fftpts; hpts=fftpts/2; npts=length(cc1); w=hann(fftpts)'; % windowing mALLFFTS1=[]; mALLFFTS2=[]; for n=0:floor(npts/fftpts)-1; %floor rounds down to keep whole number, ceil rounds up startindex=(n*fftpts)+1; %start of each fftpts segment endindex=startindex+fftpts-1; %end of each fftpts segment SIGNAL1=fft(cc1(startindex:endindex).*w,fftpts); %channel 1(manatee's left ear);Calculates FFT SIGNAL1_f=abs(SIGNAL1); % Absolute value, calculates magnitude at each frequency, gets rid of phase info SIGNAL1_s=SIGNAL1_f/hpts; % Scales the results appropriately for yaxis mALLFFTS1=[mALLFFTS1;SIGNAL1_s(1:hpts)]; SIGNAL2=fft(cc2(startindex:endindex).*w,fftpts); %channel 2 (manatee's right ear);Calculates FFT SIGNAL2_f=abs(SIGNAL2); SIGNAL2_s=SIGNAL2_f/hpts; mALLFFTS2=[mALLFFTS2;SIGNAL2_s(1:hpts)]; end %Averages FFTs for manatee absent recordings npts=length(channel1); npts=length(channel2); ALLFFTS1=[]; ALLFFTS2=[]; for n=0:floor(npts/fftpts)-1; %floor rounds down to keep whole number, ceil rounds up startindex=(n*fftpts)+1; %start of each fftpts segment endindex=startindex+fftpts-1; %end of each fftpts segment SIGNAL1=fft(aa1(startindex:endindex).*w,fftpts); %channel 1(manatee's right ear);Calculates FFT SIGNAL1_f=abs(SIGNAL1); % Absolute value, calculates magnitude at each frequency, gets rid of phase info SIGNAL1_s=SIGNAL1_f/hpts; % Scales the results appropriately for yaxis ALLFFTS1=[ALLFFTS1;SIGNAL1_s(1:hpts)]; 205

PAGE 219

206 Appendix D: (Continued) SIGNAL2=fft(aa2(startindex:endindex).*w,fftpts); %channel 2 (manatee's right ear);Calculates FFT .*w = windows each point on waveform SIGNAL2_f=abs(SIGNAL2); SIGNAL2_s=SIGNAL2_f/hpts; ALLFFTS2=[ALLFFTS2;SIGNAL2_s(1:hpts)]; end F=(0:hpts-1)*binwidth; % Creates frequency scale for x axis MAchannel1=(20*log10(mean(ALLFFTS1))); % Variable for Manatee Absent Left ear MPchannel1=(20*log10(mean(mALLFFTS1))); % Variable for Manatee Present Left ear MAchannel2=(20*log10(mean(ALLFFTS2))); % Variable for Manatee Absent Right ear MPchannel2=(20*log10(mean(mALLFFTS2))); % Variable for Manatee Present Right ear Lear=(MAchannel1)-(MPchannel1); % Subtracts Manatee Absent from Manatee Present in Left ear Rear=(MAchannel2)-(MPchannel2); % Subtracts Manatee Absent from Manatee Present in Right ear HRTFoverlay; figure(2); % plots frequency averages of left & right ears hold off; plot (F(1:165)/1000,Lear(1:165)); %/1000 to plot in kHz hold on; plot (F(1:165)/1000,Rear(1:165), 'r'); % blue (manatee's left ear), red (manatee's right ear) xlabel( 'Frequency (kHz)' ) ylabel( 'dB'); figure(3) %plots the signal from the right ear subtracted from the left ear Dear=Lear-Rear; %diff between ears plot(F(1:165)/1000,Dear(1:165)) xlabel( 'Frequency (kHz)' ) ylabel( 'dB'); binHz=200; % 200 Hz bins startf=floor(200/binHz)+1; % bin corresponding to 200 Hz endf=floor(1500/binHz)+1;; %Bin corresponding to 1500 Hz; this is what was used for localization lowfdiff=mean(Dear(startf:endf)) startf=floor(200/binHz)+1; % bin corresponding to 200 Hz endf=floor(5000/binHz)+1;; %Bin corresponding to 5000 Hz; this freq has a wavelenght close to intermeatal distance lowfdiffintermeatal=mean(Dear(startf:endf)) startf=floor(18000/binHz)+1; % bin corresponding to 18000 Hz endf=floor(30000/binHz)+1;; %Bin corresponding to 30000 Hz highfdiff=mean(Dear(startf:endf))

PAGE 220

About the Author Debborah E. Colbert graduated from Ne w College, the Honors College for the State of Florida, in May 1999 with a Bachelor of Arts de gree in Psychology. She was awarded the American Association of Univer sity Womens Scholarship and completed an undergraduate thesis entitled Basic Husbandry Training of Two Florida Manatees ( Trichechus manatus latirostris ) during her senior year. Her advisors were Dr. Gordon Bauer and Dr. Heidi Harley. She graduated with a Master of Arts Degree in Psychology, the Cognitive and Neural Science Program, from the University of South Florida in May 2005. Her thesis was titled Sound Localization Abiliti es of Two Florida Manatees, Trichechus manatus latirostris , and her major advisor was Dr. Sarah Partan. She has held the Secretary Officer positi on for the International Marine Animal and Training Association (IMATA) from 2003 through 2006 and is currently Chair of IMATAs Research and Conservation Committee. Debi currently lives with her husband Larry and three daughters, Katie, Alyssa a nd Lauren in Myakka City, Florida.