The University of Southern Mississippi
NON-VOCAL COMMUNICATION IN THE ATLANTIC SPOTTED DOLPHIN
(STENELLA FRONTALIS) AND THE INDO-PACIFIC BOTTLENOSE DOLPHIN
(TURSIOPS ADUNCUS)
by
Robin Duncan Paulos
A Thesis Submitted to the College of Education and Psychology of The University of Southern Mississippi in Partial Fulfillment of the Requirements for the Degree of Master of Arts
Approved:
______Director
______
______
______
______Dean, College of Education and Psychology
May 2004 ABSTRACT
NON-VOCAL COMMUNICATION IN THE ATLANTIC SPOTTED DOLPHIN
(STENELLA FRONTALIS) AND THE INDO-PACIFIC BOTTLENOSE DOLPHIN
(TURSIOPS ADUNCUS)
by Robin Duncan Paulos
May 2004
The goal of this project was to increase our understanding of non-vocal communication in the social lives of two species of dolphins, the Atlantic spotted dolphin
(Stenella frontalis) and the Indo-Pacific bottlenose dolphin (Tursiops aduncus). The non- vocal behaviors produced by dolphins in various social contexts and the reactions of other dolphins to these behaviors were described and analyzed. Although no specific patterns of behavior (defined as sequences comprised of greater than two behaviors) were found in either species based on lag sequential analysis, significant associations between target events and non-vocal behaviors were identified. Several behaviors were significantly associated with three “target events” (depart, join, and contact) in both study groups including contact behaviors. Spotted dolphins were more likely to touch another individual after joining than before departing (76.47% vs. 28.57%). The Indo-
Pacific bottlenose dolphins were equally likely to touch another after joining or before departing (44.44% vs. 57.14%). However, there was no significant difference between species in the use of these behavioral combinations. Both species engaged in reciprocated contact in only two behavioral contexts (mingle and play). The spotted
ii dolphins, however, use a wider variety of contact behaviors than do the Indo-Pacific
bottlenose dolphins.
Serendipitously, a behavior previously undocumented in the literature was
observed in both species of dolphin during data analyses for this study. This behavior
was labeled an “oscillating swim”. This swim behavior is defined as a dolphin moving
forward while its entire body is involved in a rolling-type movement. The oscillating
swim was documented in five social contexts (general social, travel, forage, play, and inquisitive) for both species, but the use of this behavior varied with age (χ2 = 24.241; df
= 3; p<.001) as well as by sex (χ2 = 3.98; df = 1; p<.05).
iii DEDICATION
This thesis is respectfully and lovingly dedicated to Terry and Hope Paulos, and
Jennie and Robert Duncan. Your unending support, indefatigable enthusiasm, and constant encouragement kept me going when the light became dim. Thank you for being such wonderful role models and tireless cheerleaders.
iv ACKNOWLEDGMENTS
This work could not have been completed without the advice and support provided by a number of people. I would first like to thank the Chair of my committee,
Stan Kuczaj. Without his generosity, guidance, and artful prodding this project might still be in the planning stage. I would also like to thank him for his flexibility and commitment in making it possible for me to spend time in the field. Sincere thanks are also extended to my other committee members, Tammy Greer and Sheree Watson, for their sage advice and comments throughout this process. Special thanks go to committee member, Kathleen Dudzinski, not only for her counsel, comments, and encouragement, but for her willingness to disseminate and share data used in this project. Without her data and her support, this project could not have even commenced. I also thank her for providing opportunities for me to gain invaluable field experience. Finally, I wish to acknowledge with gratitude the technical assistance provided by Andrew Wright and
Peter Sugarman in the creation of macros used in the analyses in this project (without which I would still be calculating Z-scores).
v TABLE OF CONTENTS
ABSTRACT………………………………………….……………………………ii
DEDICATION…………………………………………………………………….iv
ACKNOWLEDGEMENTS………………………………………………………..v
LIST OF ILLUSTRATIONS…………………………………………………….viii
LIST OF TABLES………………………………………………………………….x
LIST OF ABBREVIATIONS………………………………………………..……xi
CHAPTER
I. INTRODUCTION………………………………………………….1
Non-vocal Communication in Humans Communication in Non-human Animals Communication in Dolphins
II. THE PRESENT STUDY………………………………………….23
III. METHODS………………………………………………………..25
The Sample Animal Variables Behavioral Context Behavior Codes Video Data Collection Analysis
IV. RESULTS………………………………………………………...36
Sequences of Behavior Behavioral Associations Adventitious Finding: Oscillating Swim
vi V. DISCUSSION……………………………………………….…….61
Sequences of Behavior Behavioral Associations Adventitious Finding: Oscillating Swim
VI. CONCLUSION…………………………………………………...76
Limitations and Future Study
APPENDIXES…………………………………………………………………….79
REFERENCES………………………………………………………………..…115
vii LIST OF ILLUSTRATIONS
Figure
1. Geographic location of video data acquisition for Stenella frontalis……….…..26
2. Geographic location of video data acquisition for Tursiops aduncus………..…27
3. Diagram illustrating ‘skeleton’ transitions (solid lines) and intervening transitions (dotted lines) between behaviors C, B, A, and target event………………….…35
4. Three-behavior sequence depicting pattern that occurred more than once...... 39
5. Number of total occurrences of contact-depart and the number of occurrences in which the dolphin is both the initiator of contact and the ‘departer’ in both study groups…………………………………………………………….………43
6. Number of total occurrences of contact-depart in the Bahamas study group and the number of occurrences in which the dolphin is both the initiator of contact and the ‘departer’ for each age class……………………………….…..43
7. Number of total occurrences of contact-depart in the Japan study group and the number of occurrences in which the dolphin is both the initiator of contact and the ‘departer’ for each age class…………………………………..44
8. Number of depart occurrences by behavioral context that were preceded by a contact behavior in the Bahamas study group……………………...…….46
9. Number of depart occurrences by behavioral context that were preceded by a contact behavior in the Japan study group………………………………..46
10. Number of total occurrences of join-contact and the number of occurrences in which the dolphin is both the initiator of contact and the ‘joiner’……….....49
11. Total number of identified dolphins in the Bahamas study group by age class and the number of dolphins that were both the joiner and initiator of contact………………………………………………………………………….50
12. Total number of identified dolphins in the Japan study group by age class and the number of dolphins that were both the joiner and initiator of contact……………………………………………………………………….…51
viii 13. Total number of join events in the Bahamas study group by bbc and the number of times that contact followed those join events…………………..53
14. Total number of join events in the Japan study group by bbc and the number of times that contact followed those join events……………..…....53
15. Total number of identified dolphins and number observed performing oscillating swim by sex and site……………………………………………..…58
16. Total number of identified dolphins and number of dolphins observed performing the oscillating swim in the Bahamas study group by age………….59
17. Total number of identified dolphins and number of dolphins observed performing the oscillating swim in the Japan study group by age……………..59
ix LIST OF TABLES
Table
1. Spot Class Designations and Associated Color Pattern Changes in Atlantic Spotted Dolphins…………………………………………………….………...29
2. Number of Seconds by Site, Year, and Broad Behavioral Context …………..36
3. Number of Target Events for Atlantic Spotted Dolphins in the Bahamas and Indo-pacific Bottlenose Dolphins in Japan by Broad Behavioral Context…....37
4. Proportion of Animals in Each Age Group by Year…………………………..38
5. Number of Dolphins Included in the Sequential Analyses by Site and Age Group……………………………………………………………………..40
6. List of Significant Behaviors (Z> 3.291) Before (lags -1, -2, or -3) the Target Event depart…………………………………………………………....42
7. Percentage of depart Occurrences that were Preceded by a Contact Behavior……………………………………………………………………….45
8. Results for the Pairwise Comparisons (for contact-depart) using Holm’s Sequential Bonferroni Method……………………………………..…47
9. List of Significant (Z> 3.291) Behaviors After (lags +1, +2, or +3) the Target Event join…………………………………………………………..….48
10. Percentage of join Occurrences that were Followed by a Contact Behavior………………………………………………………………….…...52
11. Results for the Pair-wise Comparisons (for join-contact) using Holm’s Sequential Bonferroni Method……………………………………………...... 54
12. List of Significant (Z> 3.291) Behaviors that Occurred before (lags -1, -2, or -3) and After (lags +1, +2, or +3) the Target Event contact…..56
13. Results for the Age Pair-wise Comparisons using Holm’s Sequential Bonferroni Method………………………………………………………….…60
x LIST OF ABBREVIATIONS bbc broad behavioral context
DCP dolphin communication project
MA /I F Mystic Aquarium/ Institute for Exploration
MIK Mikura Iruka Kenkyukai sag significantly occurring behavior
xi
CHAPTER I
INTRODUCTION
Communication is an integral part of daily life for social species (Altmann, 1967;
Cullen, 1972; Smith, 1977). Wthout a system to convey information reliably and
consistently, a social system could not exist (Marler, 1977; Otte, 1974). Imagine a world
in which a word had no specific meaning, or for example, one day the word “embrace”
indicated a sign of affection and the next day it was indicative of aggression. Although in
human language word meanings do sometimes change, having a communication system
that is relatively stable allows us to effectively convey information about events and our
surroundings to others. W e can be reasonably sure that when we hear the word
“embrace” today, it will have the same meaning tomorrow.
Many definitions of “communication” have been used in the study of human and
animal behavior. Definitions for the word abound in the literature (for review see
Hauser, 1996). Almost every author discussing communication provides a definition of the concept within the context of their study. These definitions are alike in that they all involve a transfer of information, but some are too restrictive. For example, Kimura
(1993) offers a definition that restricts the recipient of the signal to the same species as
the communicating animal. According to K imura, communication refers to the
“behaviors by which one member of a species conveys information to another member of
the species” (p. 3). However, information exchange occurs between as well as within 2 species (e.g. in the Bahamas between Atlantic spotted dolphins, Stenella frontalis, and
Indo-Pacific bottlenose dolphins, Tursiops aduncus, Dudz inski, 1996; Herzing, 1997;
Herzing, Moewe, & Brunnick, 2003). For Dusenbery (1992), “the term ‘true
communication’ is restricted to cases in which the transmitting organism engages in
behavior that is adaptive principally because it generates a signal and the interaction
mediated by the signal is adaptive to the receiving organism as well” (p. 37). However,
this adaptive component may not always be present when communication occurs (e.g., in
a verbal expletive uttered by a person who just pounded his thumb with a hammer and his
friend’s offering of sympathy).
Vauclair (1996) offers a parsimonious definition that is less restrictive than those
of Kimura and Dusenbery. He defines communication as an exchange of information
between a sender and a receiver using a code of specific signals that usually serve to meet
common challenges (e.g., reproduction, predator defense, foraging, and promotion of
group cohesiveness in group-living species). Moreover, these signals may be broken
down into three fundamental parts and examined as components of behavioral
interactions. According to Vauclair (1996), these three essential pieces are: a
communicator to provide the signal, a signal, and a recipient to receive the signal.
W ithout any one of these integral pieces, the process of communicating cannot be
completed.
These components are further mitigated by the group composition of the
interacting individuals. Animals do not coordinate their activities with one another by
chance. They do so on the basis of information that they receive from each other and
their environment. Social interaction occurs when two or more individuals interact and 3 have an impact on each other. Some communication occurs in the context of such social interactions when the “message” is based on a shared code that develops through a shared history, both phylogenetically and ontogenetically (Fehr & Exline, 1987). The use of this shared code is a means of establishing a predictable continuity. Therefore, the comprehension of social meaning is equally dependent upon understanding of the code employed, the context in which aspects of the code are used, and the stability of the code over time. For humans as well as all other animals, being successful as social creatures depends on our accurately anticipating and predicting the behavior of conspecifics.
Communication as a whole is a system of co-adaptation responsible for maintaining society and therefore helping to perpetuate human as well as non-human animal existence.
W hen thinking about human communication, we generally think of language or vocal behavior as the predominant mode of signal exchange. Even in the “information age” of the 21st century one cannot deny the implications of technology in the symbolic transmission of information. Communication via printed words would, for our purposes, be considered “verbal” communication as it requires the use of words in the conveyance of the signal. DeV ito and Hecht (1990) define “non-verbal” communication as involving messages other than words. These messages may be conveyed through the use of the physical body (e.g., facial expressions, gestures, or bodily movements) or through the use of external influences such as proximity to the receiver, timing of the message, vocal characteristics, olfactory cues, or objects. Both “verbal” and “non-verbal” generally describe methods of communication employed by humans. “ V ocal” communication is interchangeable with “verbal” communication in that it also can refer to human speech. 4
It is, however, more broadly defined as “sounds produced by expelling air from the lungs, first past some sort of vibrating mechanism and then one or more resonating chambers or tubes” (Smith, 1977, p. 31). By definition, this includes not only the speech, screams, and laughter of humans, but also a wide variety of sounds produced by other animals
(e.g., roars of lions and whistles of dolphins). “Non-vocal” communication therefore may refer to means of communication involving body parts not used to produce sounds
(e.g., body posture or facial expression in humans or animals). In general, “verbal” and
“non-verbal” are terms used to describe human communication and “vocal” and “non- vocal” are used with reference to other animals. To avoid confusion, the terms “vocal” and “non-vocal” communication will be used exclusively henceforward. These definitions of vocal and non-vocal communication are not incontrovertible (for reviews
see Hauser, 1996; Hinde, 1972; Scherer & Ekman, 1982; Siegman & Feldstein, 1987).
However, for our purposes they provide a reasonable distinction in fundamental meaning of these two terms.
In addition, a third category of communicative behavior (non-vocal acoustic) will be addressed. Non-vocal acoustic behavior involves the production of sound, but not through the vocal pathway (e.g., a finger snap by a human or a jaw pop or tail slap from a dolphin).
Non-vocal Communication in Humans
Non-vocal communication plays an important role in our interactions with other
people from a very young age. Gestures emerge in young children prior to the
development of language (Acredolo & Goodwyn, 1988; Bates, 1976). Even people blind
from birth use gestures when speaking (Iverson & Goldin-Meadow, 1998). It has been 5 suggested that an estimated two-thirds of the meaning in any human social situation is obtained from non-vocal cues (Birdwhistell, 1955). Although this estimate is a matter of some debate (see Dittmann, 1971; Jolly, 2000), it is generally agreed that most human communicative acts have associated non-vocal attributes (Burgoon, Buller, & Woodall,
1996).
In general, human non-vocal communication may serve several functions. Ekman and Friesen (1969) propose five functional categories for human non-vocal communication. These functional categories include redundancy (duplicating the verbal message), complementation (magnifying or elaborating on the verbal message), substitution (replacing the verbal message), emphasis (highlighting the verbal message), and contradiction (sending a signal with the opposite meaning of the verbal message).
Ekman and Friesen’s categorical approach represents an interactive conception of the function of non-vocal communication. Here the verbal message and the non-vocal message are viewed as separate, but interact to produce a complete picture of the intended signal. McNeil (1992) suggests that gestures are integral to language. Both gestures and speech are part of a single system in which the two components are contributing equally in the process of communicating. It has also been suggested that gestures may even facilitate the process of recall by approximating a mental representation of a word that does not immediately come to mind (Krauss, 1998;
Rauscher, Krauss, & Chen, 1996).
The human non-vocal communicative behavioral repertoire itself varies across cultures. Cultures differ in their adoption of contact and proxemic (distance) behaviors.
For example, what one culture considers polite, another culture might consider offensive 6 or rude. North American, northern European, and Australian communicators employ greater interpersonal distances and less contact behavior when engaged in interactions than do communicators from many Latin and South American, southern and eastern
European, Middle Eastern, and Asian cultures (Barnlund, 1975; Little, 1968; Shuter,
1977). In the Middle East, moving away from your conversational partner is a sign of disrespect and hiding your breath is a sign of shame or deceit (DeVito & Hecht, 1990).
Jourard (1966) reported that the frequency of physical contacts between couples when in public varied across cultures: couples in San Juan, Puerto Rico, averaged 180 physical contacts per hour, couples in Paris averaged 110 per hour, and those in Florida only 2 per hour. In London, no contact behavior was noted between couples in a public circumstance. Through the processes of cultural transmission and social learning, the norms of individual societies are modeled and duplicated (and sometimes altered) by successive generations. This leads to general continuity within cultures, but differences between cultures. However, even in cultures that have adopted a greater interpersonal distance in communication situations, familiarity, kinship and context play a role in determining proxemic and haptic behavior (LaFrance & Mayo, 1979). In cultures that usually maintain larger interpersonal space during conversations, distance is minimized during interactions between k in or close relations (Argyle, 1969).
Cultural influences are not the only moderators of nonverbal behavior. Different rules govern behavior in different social situations. Behavior at a bachelor party would be very different than behavior at a church gathering. Even though both of these situations may involve a gathering of friends and the consumption of food and drink, behavior at the bachelor party would likely be much less subdued and controlled than 7 behavior at the church gathering. The social conventions about each of these situations
(contexts) influence what patterns of nonverbal behavior should be used. At the church gathering polite clapping might be used to indicate approval. At the bachelor party
stomping or more emphatic applause might be employed. Although there would be
individual differences in behavior, the cultural norms provide a fundamental model on
which to base this behavior.
In humans, then, we see many factors which might influence the form of non- vocal communication. These factors include not only the function of the signal to be transmitted, but also the cultural environment and the social context of the transmission.
The composition of the group (families, strangers, or co-workers) may also influence the behaviors exhibited in a given situation.
Communication in Non-human Animals
Like humans, animals have a multiplicity of signals and displays in their communication repertoire. Darwin (1898) was the first to describe the wide variety of expression in animal behavior ranging from anger (the change in the humming of a bee or the rattle of a porcupine’s quill) to impatience (pawing of the ground by a horse). Early ethologists suggest that some communicative signals in animals may have emerged in evolutionary history from movements that were goal-oriented but not produced to convey information (Andrew, 1963; Moynihan, 1967; Tinbergen, 1952). Once these “intention movements” were reliable predictors of behavioral events, they would become signals through a process of ritualization (Hauser, 1996). An example of this type of intention movement is the pre-flight movements of many birds. Prior to taking off, many birds crouch, raise their tail, or bob their head. Often these movements are repeated several 8 times prior to actual flight and may serve as a signal to flock members that the bird is about to fly. In pigeons, if pre-flight intention movements are performed prior to flight,
flockmates are not disturbed (Rogers & K aplan, 2000). However, if the pigeon takes
flight without performing these movements, it is a signal of alarm and the whole flock
will take off.
Intention movements may be used in other situations as well. A gull (Larus spp.)
poised to attack may approach an opponent with its neck stretched upward and head
pointed down. This posture is the same posture exhibited when the gull actually pecks its
opponent. Therefore, when the gull uses this display (without actually peck i ng), it is
effective as a threat gesture (Tinbergen, 1960, 1965). Similarly, a deer (Odocoileus spp.)
need only to lower its head and orient its horns at an opponent to indicate a threat
(Brown, 1974; Ozoga, 1972). Fox & Cohen (1976) suggest that the canid play bow (a
stereotypical display that often precedes playful interludes) may have evolved in this way
from a highly ritualized form of stretching that communicates a state of relaxation.
Through the use of these types of communicative signals social animals are able
to effectively transmit information among themselves with benefits at both the ultimate
and proximate levels. At the ultimate level, the function of communication is to increase
an individual’s or the group’s reproductive success. This is accomplished through the use
of communication in the defense of a territory, socialization of the young, or in the
facilitation of sexual interactions. In these situations, communication serves to enhance
survival and reproduction. At the proximate level, communication aids individuals in
meeting their everyday needs. Communication may provide information about location 9 of food or the approach of predators. It may also function to mediate social interactions
with other members of the group and provide information useful in later encounters.
The information obtained in these communicative exchanges is useful to the
recipient. It allows the recipient to choose a course of action appropriate to the situation.
In many cases, both individuals (sender and receiver) gain from sharing information. For
example, the alarm call of a vervet monk ey (Cercopithecus aethiops) (Cheney &
Seyfarth, 1990; Struhsaker, 1967), the raised hair of a cornered cat (Felis spp.) (Adamec,
Stark-Adamec, & Livingston, 1980), and the teeth-baring of a mad dog (Canis spp.) (Fox,
1969; Fox & Clark, 1971) are all examples of types of signal exchanges used to convey
information among non-human animals. The cornered cat, the mad dog and their
antagonists benefit by avoiding an agonistic encounter. In other cases (e.g., the vervet
monkey alarm call), the advantage to the receiver is obvious but it is unclear if the sender
benefits.
As many of the previous examples illustrate, all animal communication systems
are not alike. Animals may use photic (visual), mechanical (tactile and acoustic),
chemical (taste and olfaction), electromagnetic, or a combination of these channels to
convey information (Herman, 1980; Reynolds & Rommel, 1999; Sebeok, 1977). The
system must be based on the capabilities of the sender and receiver of the signal as well
as the environment in which the signals will be transmitted. For example, terrestrial
mammals with a well-developed tail and a keen sense of sight such as the wolf (Canis
lupus) may use tail position (photic channel) to convey internal states (submissiveness or tension) or to indicate a threat (Schenckel, 1947). In the foot-drumming k angaroo rat
(Dipodomys heermanni), a non-vocal acoustic signal is used to help distinguish 10 individuals and transmit information about willingness to interact (Shier & Yoerg, 1999).
In this case the primary stimulus is acoustic, but an animal witnessing the act may receive the same message via the photic channel. Another Dipodomys species, the banner-tailed k angaroo rat (D. spectabilis), uses airborne signals to communicate with distant neighbors while outside the burrow on windless nights. However, these banner-tailed k angaroo rats use substrate-borne vibration to communicate to near neighbors from inside the burrow on windy nights (Randall & Lewis, 1997).
The modality of signal transmission may also vary based upon the context of the
social interaction. Animals may use different k inds of communication to convey
information about different events. Bodily contact (mechanical/tactile) in pigtail
monkeys (Macaca nemestrina) helps to develop strong bonds and provide reassurance
when frightened, while vocalizations (mechanical/acoustic) or facial gestures (photic)
may be used in settling issues of dominance (Argyle, Alkema, & Gilmour, 1971).
Animals may use these different modes of communication singly, or in
conjunction with each other. Many mammals (e.g., primates, cats, and dogs) use facial
signals such as raised eyebrows or changes in muscular contractions of the mouth area
along with vocal signals to transmit information (Altmann, 1965a, 1965b; Fox, 1969;
Snowdon & Boe, 2003). The hands of primates or the forepaws of canids may be used in
combination with vocalizations as a means of solicitation. Chimps (Pan troglodytes)
often use a begging hand gesture paired with a vocalization to solicit food or attention
from another animal (de W aal, 1982) W olves also use gestures in conjunction with
vocalizations as solicitations for grooming, feeding, or affection from other wolves
(Peters, 1980). The gestures used include a paw raised in the direction of the receiver, or 11 a sharp, upward movement of the snout directed toward the underside of the receiver’s chin. The use of different modalities in the transmission of these signals is a reflection of the capabilities of the interacting organisms and the constraints of the physical and social environment (Hauser, 1996).
Visual signals are an important source of non-vocal information transfer in many animals. As in many of the above examples, visual signaling in animals may involve body posture, gesture, or facial displays. For example, wolves and other canids often use body posturing to convey information (Fox, 1971; Peters, 1980). In general, a dominant animal’s posture involves the head and tail erect, while a subordinate animal exhibits a crouched posture with tucked-down tail and flattened ears (Fox, 1971). V isual signaling may also involve movement of a particular portion of the body, as in the head bobbing
gestures of anole lizards (Anolis carolinensis) to attract mates or warn opponents
(DeCourcy & Jenssen, 1994; Lovern & Jenssen, 2003). Facial displays are exhibited by
many animals including primates and canids (van Hooff, 1967; Andrew, 1963; Chevalier-
Sk olnik off, 1982; Goodall, 1986). The silent bared-teeth grimace (mouth slightly open
with lips drawn back) is common in monkeys and apes when expressing fear or when
trying to appease another (de W aal & Luttrell, 1989). Canids often use facial displays
together with postural components to convey a message. During play in wolves, the
signaler may exhibit a “play face” in which the animal is panting with the mouth open
and the lips retracted horizontally (Peters, 1980). This facial display is commonly paired
with the canid play bow (Bek off, 1975) and is exhibited prior to and within a play
sequence. These stereotypical displays are thought to act as signals that the interaction is
not serious (Fox, 1971). 12
The eyes in particular, are frequently a focus of visual signals. The eyes may be used in affiliative and in agonistic contexts, as well as to direct another’s attention.
Gorillas (Gorilla gorilla) often use a stare as an invitation or in encouraging friendly interactions (Yamagiwa, 1992). However, a direct stare is a threat signal in most primate species, including humans. The mildest form of threat seen in intra-troop conflict in vervet monkeys is staring directly at the receiver for 3 to 5 seconds (Peters, 1980).
Gaze is another element in the complex social interactions of many primates and canids. It provides animals with a means for evaluating another’s interest and is a primary means of communication in many species (Cheney & Seyfarth, 1990; Tomasello & Call,
1997). Gaz e following (following the line of sight of another) also allows individuals to tak e advantage of the visual experience of others (Peters, 1980). Primates have the capacity to follow the gaze direction of conspecifics to outside obj ects. This ability has been documented in several species including chimpanzees, sooty mangabeys
(Cercocebus torquatus), and three macaque species (Macaca mulatta, M. nemestrina, &
M. arctoides) (Tomasello, Hare, & Fogelman, 2001). Gaze alternation (alternately directing gaze at a focus object then back to an audience) has also been recognized in the domesticated dog (Canus familiaris) in inter-species interactions with humans (Hare &
Tomasello, 1998; Mik ló si, Polgá rdi, Topál, & Csá nyi, 2000).
Tactile signaling is common in many animal species through the use of various
body parts. Grooming by hand in primates and by bill in birds is effective as a hygienic
regime, but is thought to also have a communicative component (Rogers & K aplan,
2000). Specifically, grooming may help in maintaining and establishing social bonds or
providing appeasement in tense situations. Grooming is associated with status in rhesus 13 monkeys (Macaca mulatta) and baboons (Papio ursinus) (Rogers & K aplan, 2000). In
these primate species, it is most often the dominant animals that are groomed by
subordinate ones. Elephants (Loxodonta & Elephus spp.) may also use this channel to communicate. The use of the trunk in rubbing another’s face is commonly seen in situations of affiliation or greeting (Langbauer, 2000; Rogers & K a plan, 2000). In canids, tactile communication may be accomplished through the use of the tongue.
Licking is used in reassurance as well as in bonding, conflict resolution, or as a signal of status (Fox, 1971).
Communication in Dolphins
Dolphins are both social and highly mobile (Dudzinski, 1998; Rossbach &
Herzing, 1999; Slooten, 1994; Smolker, Richards, Connor, & Pepper, 1992). Dolphins may use tactile (physical contact between animals), visual, auditory, acoustic behavioral
(sound emitted from a tail slap or jaw pop), or a combination of these means by which to relate signals (Dudzinski, 1998; Dudzinski, Thomas, & Douaze, 2002; Herman, 1980;
Pryor, 1990; Reynolds & Rommell, 1999). However, the meaning of the majority of these signals remains largely unknown. These tactile, auditory behavioral, and visual signals may be used in combination with vocal (whistles and pulsed sounds) signals to enhance or maximize a message (Dudzinski, 1998).
Dolphins have a large repertoire of vocalizations spanning frequencies from below 100 Hz to higher than 100 kHz (Tyack & Clark, 2000). Such sounds can be classified into three general categories: tonal whistles, pulsed sounds (short duration as used in echolocation) and less distinct burst-pulse sounds such as grunts, barks, and squawks (Richardson, Greene, Malme & Thomson, 1995). Whistles are generally 14 narrow-band sounds that range in frequency from about 2 to 25 kHz and have been observed in mother/calf reunions, courtship, alloparental care, other affiliative, and agonistic/aggressive situations (Caldwell & Caldwell, 1977; Dudzinski, 1996; Herzing,
2000; Smoker, Mann, & Smuts, 1993). The pulsed sounds, or clicks, are used primarily in echolocation but may occur in social contexts as well such as courtship, discipline, foraging/feeding and play (Evans, 1973; Herzing, 2000; Norris, 1969). The literature on burst-pulse sounds suggests that these sounds are produced in very different contexts.
They may occur in social situations (specifically sexual activity) or when closing in on
prey (Tyack & Clark, 2000). Herzing (2000) has suggested that this type of vocalization may have a tactile effect on the recipient, and that this effect could range from pleasure to pain depending on the intensity. However, the specific communicative functions of dolphin vocalizations have not been determined in any great detail.
Non-vocal Communication in Dolphins
Non-vocal communicative behavior in dolphins may take several forms. It may primarily be visual (such as a posture), tactile (such as a pectoral fin rubbed along the side of another dolphin), or it may be an acoustic non-vocal signal (such as a jaw clap or tail slap on the surface of the water).
However, the study of dolphin non-vocal behavior has been hindered in several ways. Dolphins (and cetaceans in general) lack the ability to vary their facial expressions. They also lack fingers and flexible limbs. In addition, most of the interactions that occur between dolphins take place underwater.
15
Visual Signals
A dolphin’s vision may play an important role in communication, allowing for conveyance of communicative signals. Visual displays such as leaping or position of body parts may help to communicate information about the environment or the behavioral state of the sender. For example, dusky dolphins (Lagenorhynchus obscurus) may leap to indicate prey has been located (Würsig & Würsig, 1980). A threat may be indicated by facing another animal with the head down and an open mouth (Caldwell &
Caldwell, 1972).
Identification of species, age, and individual by peers (as well as by observers) may be facilitated by visual inspection. Coloration and pigmentation patterns, dorsal fin shape, and natural scars are passive cues that dolphins may use when determining identification or potential threat (Dudzinski et al., 2002; MacLeod, 1998; Würsig,
Kieckhefer, & Jefferson, 1990). Coloration patterns in spotted dolphins are also
indicators of age. As the dolphin ages, the spots become more dense (Herzing, 1997;
Perrin, 1970). It is possible that the dolphins use these coloration patterns as well as
dorsal fin shape to identify each other. Scarring on an individual may be an indicator that this animal has been in many fights and may therefore be a potential threat (MacLeod,
1998; Pryor & Shallenberger, 1990).
Aggressive as well as affiliative interactions may involve visual signals at close
range (Tyack, 2000). These visual signals may be modified by the age of the signaler or
the angle of approach (Dudzinski, 1998). For example, a male dolphin will often assume
an S-shaped posture during an agonistic encounter (Caldwell & Caldwell, 1977; Defran
& Pryor, 1980; Tavolga, 1966). This is a posture in which the dolphin positions itself 16 with its head up, torso flexed and tail up. In Atlantic spotted dolphins, a display of the
“S-posture” by itself may be seen as a threat. However, if this posture is combined with an oblique angle of approach and presentation of the genital region, it is indicative of a less aggressive and more playful context (Dudzinski, 1998). Furthermore, age may also be a moderating factor in the meaning of this signal. Spotted dolphin sub-adults perform this S-posture in the context of aggressive activity. However, juveniles were observed displaying this posture in bouts of rough and tumble play with conspecifics.
Other visual signals that have been associated with aggression or threat include a
direct approach, facing another animal head-on, shaking of the head and opening and
closing of the jaws (Herman & Tavolga, 1980). In contrast, facing away may signal
appeasement or subordination (Bateson, 1965; Caldwell & Caldwell, 1972). Moreover,
Caldwell and Caldwell (1977) have suggested that a submissive posture in which the
mouth is closed and the lateral portion of the submissive dolphin’s body is presented
occurs in some agonistic interactions between animals. Additional visual signals that
have been correlated with submissive encounters in captive dolphins include looking
away (c.f. Pryor, 1990), flinching, and generally orienting the body away from the other
animal (Samuels & Gifford, 1997; Würsig et al., 1990).
Visual cues are not only received through the eye as in the previous examples, they may originate at the eye as well. This type of visual cue, social gaze, is well documented in primates and is thought to be an important component in primate communication (Cheney & Seyfarth, 1990; Okamoto et al., 2002; Tomasello & Call,
1997; Tomasello, Hare, & Fogleman, 2001). Pryor (1990) has suggested that gaze cues may be important for cetaceans as well. Recent evidence suggests that bottlenose 17 dolphins may also have the capacity to produce and comprehend gaze signals as well as a type of referential pointing. Xitco and colleagues (2001) reported on the emergence of previously undocumented behavior in this cetacean species. In the course of interacting with humans in an unrelated study, two bottlenose dolphins spontaneously produced behavior resembling gaze alternation coupled with referential pointing. In this situation, the gaze alternation/pointing behavior was directed at a human audience and only occurred when a human observer was present. However, intra-specific activity comparable to this pointing behavior was exhibited by several wild Indo-Pacific bottlenose dolphins in response to a dead conspecific (Dudzinski et al., 2003). Although monitoring behavior was not observed in the two occurrences reported by Dudzinski and colleagues, the positioning of the dolphins’ bodies toward the carcass was suggestive of referential pointing. While a communicative function for this behavior could not be confirmed in these intraspecific interactions in the wild, the evidence from captive animals may provide further support. The spontaneous emergence of this behavior in captive animals coupled with their ability to understand the pointing gesture of a human
(Herman et al., 1999) suggests a possible communicative capacity for such behaviors to
exist in wild populations.
Tactile Signals
Touch is important in the communication system of dolphins (Caldwell &
Caldwell, 1977). The dolphin’s skin seems especially suited for this mode of information transfer as it is highly innervated and hence very sensitive (Palmer & Weddell, 1964).
Areas around the eye and blowhole are as sensitive to touch as human fingers and lips
(Connor & Peterson, 1994). Contact may be affiliative such as gentle nipping or 18 mouthing prior to sexual activity (Dudzinski, 1998; Herman & Tavolga, 1980), or rubbing or petting while pair-swimming (Dudzinski, 1998; Sakai, Hishii, Takeda, &
Kohshima, 2003). Contact may also be agonistic in nature such as ramming, biting, or tooth-raking.
In examining contact behavior and signal exchange in Atlantic spotted dolphins,
Dudzinski (1998) noted that affiliative contact was observed more than other types of physical contact. Behaviors seen most often were rubbing (body to body with movement), simple contact (body to body with no movement), petting (pectoral fin to pectoral fin with movement), and petting/rubbing (pectoral fin to body with movement).
Furthermore, the dolphins in this study were more likely to exchange rubs or pets with individuals of the same sex and age class, suggesting that they do have the ability to recognize the general age of pod-mates (Dudzinski, 1998).
Rubbing, both on other dolphins as well as on objects, is common. Social rubbing
(rubbing between dolphins) may be carried out using the whole body or just a pectoral fin. Dolphins often engage in rubs that are performed as two dolphins swim together
(Dudzinski, 1998). Head rubbing (pectoral fin of one animal rubbing melon of another) is seen in affiliative contexts such as greeting or in the process of courtship (Cummings,
Fish, & Thompson, 1972; Herman & Forestell, 1977; Marquette, 1978; Norris, Goodman,
Villa-Ramirez, & Hobbs, 1977; Payne, 1976; Saayman & Tayler, 1973). Dolphins may also rub their bodies on objects in their environment (Herman & Tavolga, 1980).
Atlantic spotted dolphins in the Bahamas engage in sand-rubbing (Dudzinski, 1998), an activity that involves rubbing their bodies along the sandy ocean floor. Captive bottlenose dolphins are often observed rubbing various parts of their body on objects in and affixed 19 to their environment, presumably for the stimulatory effect (Caldwell & Caldwell, 1972;
Defran & Pryor, 1980; Herman & Tavolga, 1980; McBride & Kritzler, 1951; Östman,
1990). Moreover, rubbing objects on the genital region is often seen even in the presence of available mates (Caldwell & Caldwell, 1972).
Some dolphin species may even have special adaptations to facilitate and enhance tactile stimulation. In the Commerson’s dolphin (Cephalorynchus commersonii), the leading edge of one or both pectoral fins is often characterized by saw-tooth serrations.
Johnson & Moewe (1999) reported that in males with only one serrated pectoral fin, 94% of the pectoral fin contact with other animals occurred through the use of the serrated fin.
They suggest that these fin serrations may enhance tactile stimulation during social contact.
Rubbing that occurs between animals may have a communicative function.
Often, a rub from one animal to another elicits the second animal to reciprocate
(Dudzinski, 1998; Herzing; 2000; Sakai et al., 2003). Alternatively, the dolphins may just be using other dolphins in the same way that they use the ocean floor or other objects--as a source of stimulation or for hygienic purposes. It is also possible that rubbing between animals could be a function of all of the above possibilities.
Tactile displays between dolphins may also be agonistic in nature. Aggressive behaviors (displayed in the context of establishing or maintaining dominance in captivity, protection of young, defending food items, or procuring sexual partners) include biting, hitting, tooth-raking, & ramming (Brown & Norris, 1956; Norris, 1967; Östman, 1990;
Slooten, 1994). These contact behaviors are often coupled with non-vocal auditory 20 signals such as jaw claps or tail slaps in agonistic encounters (Mann & Smuts, 1999;
Östman, 1990).
Tactile behaviors are not limited to intra-species interaction. Both agonistic and
affilitative behaviors were displayed in interspecific exchanges between a wild bottlenose
dolphin (Tursiops truncatus) and humans. In Belize, a solitary female dolphin displayed
a variety of behaviors in response to human activities. This dolphin was accustomed to
human activity and seemed to actively seek human interaction (Dudzinski, Frohoff, &
Crane, 1995). The behaviors most frequently observed were inquisitive in nature.
However, problematic behaviors were also observed when humans tried to exit the water.
The dolphin repeatedly positioned itself between the swimmers and access to land, often
pushing or bumping against them with her rostrum or body. Although some human
actions resulted in these aggressive behaviors in the dolphin, some human behaviors
elicited an opposite response. For example, gentle rubbing or slow swimming by a
human seemed to evoke a calm state in the dolphin (Dudzinski et al., 1995). Although
the true communicative nature of these behaviors remains unknown, the gentle rubbing
by a human may function similarly to the social rubs seen in affiliative contexts between
conspecifics.
Given the many possible functions of tactile behaviors, determining which
behaviors are communicative in nature can be difficult. According to Krebs and Davies
(1993), communication occurs when “… actors use specially designed signals or displays
to modify the behaviour of reactors” (p. 349). Therefore, a change in behavior would be
an expected outcome when a communicative behavior is produced. There may also be
levels of communicative behavior; some producing more obvious effects than others. 21
Perhaps it is only in these cases (where obvious behaviors result) that we can begin to tease apart communicative behavior from the other functions of tactile behavior.
Non-vocal Acoustic Signals
Acoustic non-vocal signals are another potential mode of non-vocal communication in dolphins. Dolphins produce non-vocal behaviors that result in a sound
being produced. Unlike vocalizations, these sounds are not produced through the vocal
pathway. They are a result of specific behaviors that produce an audible displacement of
water. These non-vocal acoustic behaviors have often been associated with situations of aggression, discipline, or attention (Pryor, 1990), and include tail slaps, breaches and jaw claps.
The jaw clap, first described by McBride and Hebb (1948), has both visual and acoustic properties. The jaws of the dolphin may open, exposing the teeth, then snap shut, producing a loud, percussive sound. This display is often coupled with a posture in which the animal arches its back while facing its opponent. If the receiver does not retreat, the displaying dolphin may charge, often resulting in an open-jawed chase
(Caldwell & Caldwell, 1977; Saayman & Tayler, 1973).
Tailslaps may be a warning signal from a dominant animal (Shane, Wells, &
Würsig,1986). Würsig & Würsig (1979) observed tailslapping when boats approached bottlenose dolphins in the south Atlantic. They suggested this might be an indicator that the animals were disturbed or annoyed.
In general, the social functions of these non-vocal acoustic signals seem to be associated with both affiliative and agonistic contexts. They manifest in situations of 22 affiliation and recruitment as well as in expressions of annoyance or disturbance
(Dudzinski et al., 2002).
CHAPTER II
THE PRESENT STUDY
The purpose of this study was to analyze non-vocal aspects of the communicative behavior of two species of dolphin, the Indo-Pacific bottlenose dolphin and the Atlantic spotted dolphin. This was accomplished by reviewing a subset of underwater video recordings of the dolphins collected in Japan (Indo-Pacific bottlenose dolphin) and in the
Bahamas (Atlantic spotted dolphin) over the past twelve years. This subset included data collected in 1993 and 1994 in the Bahamas, and 1997, 1998, and 2002 in Japan. The decision to use this subset of data was based on the availability of information regarding the identification of focal animals. Restricting analysis to identified dolphins prevents overestimation of a particular behavior due to the repeated production by a single animal.
This restriction also provides a way to track the behavior patterns of individual animals.
The video recordings were analyzed to ascertain the behaviors that occurred in specific social contexts (general social, foraging, play, inquisitive, and travel). Behavior was documented as dolphins interacted with peers and potential sequences of behaviors were explored. If a behavior or a sequence of behaviors is consistently associated with a particular context and elicits a similar response at each occurrence, it is possible that these behaviors communicate some type of information (Goodall, 1986; Struhsaker,
1967; Tschudin, Call, Dunbar, Harris, & van der Elst, 2001). Lag sequential analysis was used to ascertain if specific patterns of behavior existed in these five social contexts.
24
Sequential analysis has been used to determine associations between behaviors in
Hector’s dolphins (Cephalorhynchus hectori) (Slooten, 1994). The focus of the current project was to look for temporal patterns in behavior (A then B) and not just associations between behaviors (A associated with B either before or after). The current project added this temporal component while analyzing each species individually. The social context in which each dolphin species produced these behaviors was also investigated. The dolphins’ behavior was categorized in terms of sex, age (or age class), and context
(general social, foraging, play, inquisitive, and travel) to attempt to find answers to the following questions:
1. What types of behavioral sequences occur in specific social contexts? This
question was explored by attempting to identify behavioral patterns associated
with different social contexts in groups of dolphins.
2. Do different age classes use the same behaviors in the same context?
Comparisons of the behaviors of different age groups may help to determine the
ontogenetic course of non-vocal communication in dolphins.
3. Are the same sequences seen in the same contexts in the two different species?
This question addresses species differences in the non-vocal communication of
dolphins.
The ultimate goal of this project was to increase our understanding of non-vocal communication in the social lives of dolphins by examining communicative behaviors across age classes and species. A comparison of the communication systems within and between species may provide information about the importance of environmental or genetic influences.
CHAPTER III
METHODS
The Sample
Stenella frontalis
The Atlantic spotted dolphins represented in this study are a wild population found on the White Sand Ridge of the Little Bahamas Bank, which is approximately 64.5 km north of West End, Grand Bahama Island (Figure 1). This area ranges from 6 – 10 m in depth with a white sandy bottom, and generally good visibility to at least 30 m. The data used in this study were from a longitudinal study on dolphin communication with recordings made from 1992 to 1995, 1997, and in 2000 and 2001 (Dudzinski, 1996, 1998, unpublished data). For several months a year recordings were obtained in the course of
4-6 day boat trips. The data were collected using focal animal and all-occurrence sampling (Altmann, 1974).
26
Figure 1. Geographic location of video data acquisition for Stenella frontalis. (Adapted fr om Dudzinski, 1998)
Only video collected in 1993 and 1994 were used in this project due to availability of information regarding individual dolphin identification. Furthermore, video in which the number of animals in view affected the ability to accurately identify and follow specific animals was eliminated. Analysis was generally restricted to groups in which fewer than four animals were in view. This elimination may have an impact on the outcome of the analysis. However, it prevented inaccurate identification of the focal animal(s). For the period of consideration here (subset including 1993 and 1994), the entire identified population consisted of approximately 125 individuals with an overall equal male to female ratio (Dudzinski, 1996). For this project, 22 animals were identified in scored video clips from 1993 data and 21 animals from the 1994 data.
27
Tursiops aduncus
This study’s Indo-Pacific bottlenose dolphin group is a population resident to the area within 300 m of Mikura Island, Japan (Kogi, Hishii, Imamura, Iwatani, &
Dudzinski, in press) (Figure 2). Water depth at this location varies from 4 to 20 m within
300 m of shore. The seafloor is rocky and visibility is noticeably less than in the
Bahamas.
Figure 2. Geographic location of video data acquisition for Tursiops aduncus. (Courtesy of Mai Sakai).
Video and acoustic data were gathered using the same protocol as that described for the Bahamas. Similarly, the video data analyzed on these dolphins were a subset of
Dudzinski’s long-term, longitudinal study. The data used here included the recordings 28 obtained year-round in 1997 and 1998, and two months (May & June) in 2002. The entire identified population consisted of approximately 165 individuals with an overall equal male to female ratio (Kogi, Hishii, Imamura, Eiji, & Dudzinski, 2003). Again, video in which there were more than four dolphins in view was eliminated from the subset, resulting in a smaller subject set. For this project, 16 animals were identified in scored video clips from 1997 data, 17 animals from 1998 data, and 72 animals from the
2002 data.
Animal Variables
Sex
The sex of individual dolphins within both study groups was determined by visual inspection of the genital region. Females have one external uro-genital slit and two parallel mammary slits of ~2-3 cm. Males have an external genital, and posterior, shorter anal slit.
Sex determination for individuals within both study groups were provided by
MIK (Mikura Iruka Kenkyukai) for Indo-Pacific bottlenose dolphins and DCP (Dolphin
Communication Project at Mystic Aquarium & Institute for Exploration) for the Atlantic spotted dolphins. Both organizations conduct ongoing longitudinal research on these dolphin populations.
Age
Stenella frontalis. The age of an Atlantic spotted dolphin is relatively easy to determine as they are born spotless and gain spots with age. As a result, spot density increases with age. Table 1 is a guideline for age determination in this species 29
(Dudzinski, 1998, Herzing, 1997). Age determination for the individual spotted dolphins
observed in this project was made available by DCP.
Table 1 Spot Class Designations and Associated Color Pattern Changes in Atlantic Spotted Dolphins
Spot Class Age Grp Coloration Pattern Pattern Type 1 Neonate Gray and ivory Neonatal
2 Calf Dark gray dorsal, light gray ventral Two-tone
3 Juvenile Dark dorsal, light ventral, few spots Speckled
4 Sub-adult Entire body spotted Mottled
5 Adult Black mask, heavy spots, spots fused Fused and faded ventrally
Tursiops aduncus. The age of Indo-Pacific bottlenose dolphins is not as easily determined as that of the Atlantic spotted dolphin, but age estimations can be made. A calf (younger than 1 year of age) is approximately one-third to one-half the body length of an adult. A juvenile’s length is approximately three-fourths that of an adult animal
(Bel’kovich, Agafonov, Yefremenkova, & Kharitonov, 1991). Age estimations for
bottlenose dolphins involved in this project were provided by MIK.
30
Behavioral Context (Dudzinski, 1998)
The broad behavioral context (bbc) was determined at the time of data collection based on the activity of the majority of the group of dolphins present in the group being videotaped.
General social: This context involves physical contact among group members and generally slow movement but not in one particular direction (such as when traveling).
(Note: This context was labeled “social” by Dudzinski, 1996, 1998).
Foraging: Activity involved in procuring and eating food, such as chasing and ingesting fish, digging in the sand for fish, or any type of fish herding.
Play: Activity characterized by non-aggressive interplay (chasing) between dolphins or manipulation of objects (seaweed, fish) and/or fast and circular swimming.
Travel: One or more dolphins swimming steadily in the same general direction (not
varying more than 45° from each other).
Inquisitive: Apparent interest in the activities of swimmers in the water.
Behavior Codes
A total of 54 behavior codes in seven categories of behavior were used. Each
code was defined as either a ‘state’ or an ‘event’ type behavior. State behaviors were those behaviors that occurred over a longer period of time, whereas an event-type behavior was momentary (short duration). For example, swimming was defined as a state behavior (it has a longer duration), but a head jerk was defined as an event. The codes and descriptions of the behaviors they represent are presented in Appendix A
(adapted from Dudzinski, 1996). 31
Video Data Collection
A mobile video/acoustic array designed for underwater recording of dolphin interactions was used to capture dolphin behavior and sounds concurrently (Dudzinski,
Clark, & Würsig, 1995). This underwater array consisted of two omni-directional hydrophones cabled through a custom underwater housing. The hydrophones were connected to a Hi-8 video (Sony CCD-FX710) stereo camera.
Analysis
Continuous focal animal video was scored by viewing and simultaneously entering the behavioral data into an observational data computer program (The Observer® v5.0). Fifty-four state and event behaviors were sorted into seven categories. Video segments were examined and dolphin behavior was scored using these behaviors. After being scored with Observer® the data were tabulated and exported for further statistical
analysis.
A primary interest of this project was to determine if consistent sequences of
behavior occurred, indicating possible communicative exchanges. If sequences were
found to exist, they could signify that one dolphin is reacting in a specific way to the
behavior of another individual—suggesting that communication of some kind of
information is taking place. For this analysis, I chose to utilize lag sequential analysis
procedures. Sequential analysis is a procedure that focuses on identifying and
quantifying interactions and other relationships of particular behaviors in sequences. It provides a means of determining the probable effects one behavior might have on another based on their appearance in time. Sequential analysis can help to identify potential action chains exhibited by an animal in frequently occurring contexts as opposed to the 32 limited information that can be obtained from measures based solely on the frequency of behaviors. In the simplest case, sequential analysis can provide information in determining whether one behavior follows another more often than would be expected by chance. However, analysis is not limited to only the immediately preceding and succeeding events surrounding the behavior of interest. Conditional probabilities can be determined for events that occur at other time lags as well. A test that makes use of conditional probabilities to gauge the dependence of a later behavior (target behavior) on the occurrence of a preceding behavior (given behavior) is the Z-score binomial test
(Bakeman & Gottman, 1986). A Z-score is calculated using the conditional probability of the occurrence of the target behavior after the given behavior occurs:
p(t/g) – p(t) Z= p(t) [1-p(t)] [1-p(g)] Np(g)
Where:
N = total number of paired transitions p(t/g) = probability of target (t) occurring given the occurrence of the given (g) behavior p(t) = probability of the target behavior or p(t) = f(g)/N p(g) = probability of the given behavior or p(g) = f(t)/N
Bakeman and Gottman (1986) indicate that as N increases beyond 25 and N * p(t/g)[1-p(t/g)] > 9, the binomial distribution approximates the normal distribution.
Given these conditions, if Z > |1.96|, then the observed occurrence is significantly different from the null hypothesis (that the two behaviors are not sequentially related) at 33 the p< .05 level. However, the problem of type I error—of claiming that sequences are significant when they are not—is a problem when performing sequential analyses, especially as the number of behavioral codes used and pair-wise comparisons made increases (Bakeman & Gottman, 1986). As this project explored many pair-wise comparisons, a smaller p value (p<.001) was chosen to reduce this effect.
Sequences that originated or culminated in specific behaviors were chosen to reduce the ambiguity involved in determining the beginning or end of a potential behavior chain. These ‘target events’ were: depart, join, and contact between animals
(target events will be italicized to avoid confusion with other behaviors). These
particular events were chosen because they involve interaction between animals and
provide an opportunity to examine possible precursors to specific social behaviors.
Although this method does not completely eliminate ambiguity, the use of target events
reduces the uncertainty of where a sequence may start or end by providing a specific
target likely to elicit some sort of communication between interacting animals.
Transitions between behaviors at lags one, two, and three behaviors prior to or after the target event were examined (lags -1, -2, -3 or +1, +2, +3). For the target event depart, negative lags were examined to examine the behaviors that led up to the departure. For the target event join, positive lags were examined to examine the behaviors that occurred after the joining of animals. For the target event contact,
behaviors both before and after were examined. I chose to use state-lag analyses rather
than time-lag analyses because the coding scheme employed included both state and
event-type behaviors. The use of a time-lag analysis (which considers all behaviors that
occur within a specific time) might be misleading as it may overestimate the importance 34 of state behaviors at the expense of the shorter (but critical) event behaviors. In contrast, state-lag analyses consider all behaviors occurring within a designated number of behaviors (state or event) from the event of interest.
Transitions at lags -1, -2, and -3 or +1, +2, and +3 (depending upon the target event) were calculated and transformed into Z-scores using a macro written for Excel.
These Z-scores were examined for significance at the p<.001 level. This analysis provided information regarding what behaviors occurred significantly more often than would be expected by chance at positions one, two, and three behaviors prior to or after a target event. However, this does not provide information about the actual sequence of events that occurred. To examine the intervening events, the significant behaviors found in this preliminary analysis were used as a ‘skeleton’ from which to examine the transitions between lags. For example, when examining lags prior to a target behavior, the intervening lags (from lag -1 to -2, and -2 to -3) are also examined (Figure 3). Once significant behaviors are determined for lag -1, lag -2, and lag -3 (here behaviors A, B, and C), a determination must be made as to whether the transition from A to B is significant (resulting in a 3 event sequence: B – A – target event). This must also be repeated to determine if the transition from C to B is significant. If the intervening transitions are significant, (C to B, and B to A) then a significant sequence has been determined: C – B – A – target event. A similar procedure was employed to examine positive lags (+1, +2, and +3) from the target events. 35
Figure 3. Diagram illustrating ‘skeleton’ transitions (solid lines) and intervening transitions (dotted lines) between behaviors C, B, A, and target event.
Reliability
To assess reliability, the video segments were divided into four groups based on length in seconds and inter-rater reliability was determined using Cohen’s kappa (Cohen,
1960). Ten percent of the videos from each group were scored by a second rater and the overall value of kappa was .89 indicating a high level of agreement between raters.
CHAPTER IV
RESULTS
A total of 18,069.5 seconds of video data was used in the analysis for this project.
The video data were parsed into video clips called ‘observations’ for ease of scoring and digital storage. Observations varied in length from 3.26 to 813.16 seconds. The number of seconds by site, year, and broad behavioral context are given in Table 2. Each behavioral context was represented each year with the exceptions of inquisitive in 1993 and foraging in 1997.
Table 2 Number of Seconds by Site, Year, and Broad Behavioral Context
General Travel Play Forage Inquisitive Total Social Bahamas 1993 2058.31 844.73 1649.92 750.67 0 5303.63 1994 894.68 811.06 671.74 71.17 738.96 3187.61 Bahamas 2952.99 1655.79 2321.66 821.84 738.96 8491.24 Total
Japan 1997 309.17 107.89 306.37 0 536.55 1259.98 1998 679.82 134.89 37.1 11.13 298.83 1161.77 2002 1353.25 1593.75 2344.23 409.99 1455.28 7156.50 Japan 2342.24 1836.53 2687.7 421.12 2290.66 9578.25 Total
37
The number of target events examined varied by species (Table 3). Contact
behaviors are grouped in Table 3. This category is comprised of the following contact
behaviors cdb, cdf, cfl, cge, cla, cme, cmo, cpc, cpd, cro, and cve (as described in
Appendix A).
Table 3 Number of Target Events for Atlantic Spotted Dolphins in the Bahamas and Indo-pacific Bottlenose Dolphins in Japan by Broad Behavioral Context
General Travel Forage Play Inquisitive TOTALS Social Bahamas contact 43 15 1 29 1 89 join 25 16 7 19 4 71 depart 17 11 2 18 5 53 TOTALS 85 42 10 66 10 213
Japan contact 24 7 5 12 8 56 join 9 8 2 17 8 44 depart 6 5 1 6 11 29 TOTALS 39 20 8 35 27 129
The number of animals identified each year and the proportion of animals represented by each age group are shown in Table 4. All age groups were represented each year with the exception of calves. The proportion of animals in other age classes varies between sites and collection years as shown.
38
Table 4 Proportion of Animals in Each Age Group by Year
Adult Sub-adult Juvenile Calf Bahamas 1993 (22) .41 .05 .45 .09 Bahamas 1994 (21) .29 .14 .48 .10 Japan 1997 (16) .38 .31 .31 0 Japan 1998 (17) .59 .35 .06 0 Japan 2002 (72) .35 .60 .04 .01
The total number of animals identified for that year is given in parentheses.
This study sought answers to three basic questions in an attempt to determine if
sequences of non-vocal behaviors might have a communicative function. The first
question addressed was whether or not sequences of behavior occur in the communicative exchanges examined and if so, whether they are context specific? Second, if sequences of behavior do occur, are animals of different ages using the same sequences of behavior or different ones in the same social contexts? Finally, do animals from the two study groups (Tursiops aduncus and Stenella frontalis) use the same sequences of behavior in
similar social contexts?
Sequences of Behavior
To address the first question, lag sequential analyses were performed on specific
target events as described in the methods section. The lag analyses revealed one
sequence of more than two behaviors (that was repeated more than one time) at the
p<.001 level of significance. This three-behavior sequence was: join – pair swim –
contact pectoral fin (Figure 4). 39
joi psm cpc
Figure 4. Three-behavior sequence depicting pattern that occurred more than once.
This pattern was observed once in the bottlenose dolphins (Japan) when all contexts were grouped and analyzed together. In this interaction, a sub-adult female was interacting with an unidentified calf. This pattern was also observed in the spotted dolphins (Bahamas) in three different behavioral contexts. In the general social context, an adult female was interacting with an unidentified juvenile. In travel, two juvenile females were interacting with each other. In inquisitive, one of the same juvenile females from the travel context was interacting with the adult female seen involved in this behavior sequence in the general social context. In essence, four different animals from the spotted dolphin group were involved in producing this pattern of behavior on three different occasions. Logically, it is not surprising to see pair swim following join. The pair swim behavior, by definition, occurs when two animals are swimming together.
However, the significant association between join and a contact behavior occurring in both species as well as across different social contexts warranted additional investigation.
The lack of confirmed sequences of behavior paired with the significant occurrence of contact after joining led to an analysis of simple associations between behaviors
(described later in further detail). 40
Due to the lack of confirmed patterns, further analysis of behavioral sequences
(addressing questions 2 and 3) was not possible (however, these questions were
addressed with analyses performed to ascertain simple associations between behaviors).
Further analyses of behavioral sequences were performed after separating younger from
older dolphins. Young animals (calves and juveniles) and older animals (sub-adults and
adults) were analyzed separately in each species to see if the outcome would be affected.
The separate analysis of the two groups of animals did not affect the outcome. No significant sequences were found to occur in either the younger or older group. Table 5 provides information regarding the number of dolphins in each of these analyses.
Table 5 Number of Dolphins Included in the Sequential Analyses by Site and Age Group
All dolphins Adults & Sub-adults Calves & Juveniles Bahamas 43 19 24 Japan 105 95 10
Behavioral Associations
Due to the lack of identified behavioral sequences (of greater than two behaviors),
a decision was made to examine the behaviors simply associated with each target event as
opposed to looking for a structured pattern. Tables 6, 9, and 12 list the behaviors that
were significantly associated (Z>3.291) with each of the target events (depart, join, and
contact). Scree plots of the behaviors significantly associated with each target event (by
study group and behavioral context) were constructed (Appendix B). These plots are 41
graphic representations of the percentage of the significantly associated behavior that occurs within each of the broad behavioral contexts (bbcs). This frequency was calculated by dividing the total number of times the significantly associated behavior occurred within a particular bbc by the number of times it occurred within the three lags before or after the target event. Plots were constructed only for target events with more than one significantly associated behavior. Appendix C provides information regarding those target events that had only one significantly associated behavior as well as all data
used in calculating these frequencies.
Associations - Depart
For the target event depart, negative lags of -1, -2, and -3 were examined to determine which behaviors consistently occur prior to the departure of an animal (Table
6). For this target behavior only the behaviors that occurred prior to the target (negative
lags) were examined due to the potential absence of at least one of the subjects after
departing.
42
Table 6 List of Significant Behaviors (Z> 3.291) Before (lags -1, -2, or -3) the Target Event depart
Before depart Bahamas Japan fik (s) bbs (s) cch (i) cme (s) cla (p) cpc (i,s,p) cme (t) cks (t) cpc (p,t) xxu (i) cro (f) xxo (i) xxo (p) gsm (i,t) gsm (p) hjk (i) jop (i) joi (i,t) joi (s,p) psm (i,t) psm (f) rst (p) srb (t) sps (t) vps (p) tjk (t) vps (s) osc (f,s,p) The broad behavioral context is noted in parentheses: f=forage, i=inquisitive, s=general social,
p=play, and t=travel. Note: some behaviors occurred in multiple contexts.
Thirteen behaviors were significantly associated with one spotted dolphin departing from another. These significantly occurring behaviors included various swim behaviors (gsm, psm, and vps) as well as four contact behaviors (cla, cme, cpc, and cro).
However, three of the five behavioral contexts were represented in the interactions involving the four contact behaviors. Contact that occurred prior to the departure of an
animal occurred only in the contexts play, forage, and travel. In four of the 14 instances
(28.57%) of contact occurring prior to an animal departing, the initiator of the contact
was also the departing animal (Figure 5). However, the difference between the species
use of this behavioral combination was not significant (Z=1.27; p > .05).
43
16 14 14
12
10 total # occurrances
# initiator is departer 8 7
6 44 4
2
0 Bahamas Japan
Figure 5. Number of total occurrences of contact-depart and the number of occurrences in which the dolphin is both the initiator of contact and the ‘departer’ in both study groups.
The percentage of total animals in each age class that was both the initiator of contact and the one to depart was also determined (Figure 6).
14
12 12 12
10
8 Total # # of initiators/departers 6
4 3 3 2 2 11 0 0 Adult Subadult Juvenile Calf
Figure 6. Number of total occurrences of contact-depart in the Bahamas study group and the number of occurrences in which the dolphin is both the initiator of contact and the ‘departer’ for each age class. 44
In the bottlenose dolphins, 15 behaviors were significantly associated with the departure of an animal. These significantly occurring behaviors included the expected swim behaviors (gsm, psm, sps and vps) as well as two contact behaviors (cme and cpc).
In the bottlenose dolphins, three of the five behavioral contexts were represented in the interactions involving contact behavior. The behavior cpc (contact pectoral) occurred in all three of these contexts (inquisitive, general social, and play) while cme (contact melon) occurred only in the context general social. In four out of seven cases (57.14%), the bottlenose dolphin that initiated the contact was the dolphin to depart (Figure 5). The percentage of total animals in each age class that was both the animal that initiated contact as well as the one to depart was also determined (Figure 7).
50 47
45
40
35 31 30 Total # 25 # of initiators/departers 20
15
10 6 5 2 1 0 1 1 0 Adult Subadult Juvenile Calf
Figure 7. Number of total occurrences of contact-d epart in the Japan study group and the number of occurrences in which the dolphin is both the initiator of contact and the ‘departer’ for each age class.
The behavior crossover (xxo) was significantly associated with depart in the spotted dolphins. In the three instances (performed by one adult and two juveniles) 45 where xxo occurred prior to depart, the animal that crossed over another was also the animal that departed.
The behaviors crossover (xxo) and crossunder (xxu) were also significantly associated with depart in the bottlenose dolphins. In the three instances where xxo occurred prior to depart, the animal that crossed over another was also the animal that departed (two sub-adults and one juvenile). In the two instances of an animal crossing under another before departing, the “crosser” was also the departing animal only once and was an adult.
Overall, contact behavior is associated with the departure of an animal in both species. In the spotted dolphins 26.42% of all depart events were preceded (within three lags) by a contact behavior. In the bottlenose dolphins 24.14% of all depart events were preceded by a contact behavior (for a breakdown by social context, see Table 7 and
Figures 8 & 9).
Table 7 Percentage of depart Occurrences that were Preceded by a Contact Behavior
Bahamas Japan Inquisitive 0.00% 9.09% Forage 50.00% 100.00% General social 35.29% 50.00% Play 22.22% 33.33% Travel 27.27% 0.00% This percentage was calculated by dividing the number of depart behaviors that was followed by a
contact behavior by the total number of depart behaviors for that context.
46
20 18 18 17
16
14
12 11 # of departs 10 # preceded by contact 8 6 6 5 4 4 3 2 2 1 0 0 inquisitive forage gen. social play travel
Figure 8. Number of depart occurrences by behavioral context that were preceded by a contact behavior in the Bahamas study group.
12 11
10
8
66 # of departs 6 5 # preceded by contact
4 3
2 2 11 1
0 0 inquisitive forage gen. social play travel
Figure 9. Number of depart occurrences by behavioral context that were preceded by a contact behavior in the Japan study group.
A two-way contingency table analysis was conducted to evaluate whether a contact behavior occurring prior to an animal departing occurred differently in five 47
behavioral contexts. The two variables were site with two levels (Bahamas for spotted
dolphins, and Japan for bottlenose dolphins) and broad behavioral context (bbc) with five
levels (inquisitive, forage, travel, general social, and play). Site and bbc were found to be
significantly related (χ2=48.737; df 4; p<.001).
Subsequent to the contingency table analysis, follow-up pair-wise comparisons were conducted to evaluate where the significant differences occurred (Table 8). The
Holm’s sequential Bonferroni method was used to control for Type I error at the .05 level across all comparisons. Significant pair-wise differences were found between forage and travel, inquisitive and travel, general social and travel, and play and travel.
Table 8 Results for the Pair-wise Comparisons (for contact-depart) using Holm’s Sequential Bonferroni Method
Comparison Pearson p-value Required p-value for chi-square significance forage vs. travel 41.377 .000 .005* inquisitive vs. travel 36.000 .000 .0055* gen. social vs. travel 28.691 .000 .0062* play vs. travel 27.110 .000 .0071* inquisitive vs. gen. social 5.904 .015 .0083 inquisitive vs. play 5.486 .019 .0100 inquisitive vs. forage 4.376 .036 .0125 forage vs. gen. social 1.446 .229 .0167 forage vs. play .785 .376 .025 gen. social vs. play .019 .890 .05 (* indicates significance)
48
Associations - Join
For the target event join only the significantly occurring behaviors after the target were examined (positive lags of +1, +2, and +3). In many cases the animals that joined were not both visible on the video tape prior to the act of coming together.
Table 9 List of Significant (Z> 3.291) Behaviors After (lags +1, +2, or +3) the Target Event join
After join Bahamas Japan btg (s) cme (s.p) cch (i) cpc (f,s) cfl (s) xxu (p,t) cme (i) xxo (s) cpc (t) dep (i,s,t) cro (t) hjk (p) xxu (p) psm (i,f,p,t) xxo (s.t) sps (t) dep (s,p) vps (i,s) gsm (f,s,p,t) hjk (t) psm (f,p) rzz (p) tjk (p)
The broad behavioral context is noted in parentheses: f=forage, i=inquisitive, s=general social,
p=play, and t=travel. Note: some behaviors occurred in multiple contexts
Fourteen behaviors were significantly associated with one spotted dolphin joining another (Table 9). As with the target behavior depart, some behaviors were associated with join in both species. These behaviors included cme, cpd, xxu, xxo, dep, hjk, and psm. With the target event depart, none of the significantly occurring behaviors were seen in the same bbcs. However, with the target event join, four behaviors manifested in 49 the same bbc in both study groups. The behavior xxu was seen in play in both groups, and xxo and dep were seen in the general social context. The behavior psm was seen in two bbcs in both species (forage and play).
Significantly associated behaviors also included the expected swim behaviors
(psm or gsm) as well as five contact behaviors (cfl, cme, cpc, cro, and rzz). The only context in which contact between animals was not significantly associated with join was forage.
In 13 of the 17 instances (76.47%) of contact following join in the spotted dolphins, the dolphin that initiated the contact was also the dolphin that joined another
(initiated the join behavior) (Figure 10). The difference in use of this behavioral combination between species was not significant (Z=1.82; p> .05). However, spotted dolphins were more likely to touch another individual after joining (76.47%) than before departing (28.57%) (Z=2.58; p<.01). The Indo-Pacific bottlenose dolphins were equally likely to touch another after joining or before departing (44.44% vs. 57.14%).
18 17
16
14 13
12
10 9 total # occurrances
8 # joiner is initiator
6 4 4
2
0 Bahamas Japan
Figure 10. Number of total occurrences of join-contact and the number of occurrences in which the dolphin is both the initiator of contact and the ‘joiner’.
50
The percentage of total animals in each age class that was both the dolphin that joined
another as well as the initiator of contact was also determined (Figure 11).
14
12 12 12
10
8 Total # # of joiner/initiators 6
4 4 4 3 3
2 1 1
0 Adult Subadult Juvenile Calf
Figure 11 . Total number of identified dolphins in th e Bahamas study group by age class and the number of dolphins that were both the joiner and initiator of contact.
Nine behaviors were significantly associated with the target behavior join in the
bottlenose dolphins in this study (Table 9). The expected swim behaviors (psm, vps, and
sps) were also significantly associated with the target behavior in this species. Contact
between animals after joining was observed as well. However, only two contact
behaviors (cme and cpc) were significantly associated with the target event join in this species. The behavioral contexts travel and inquisitive were not represented by contact behavior after this target event.
In four of the nine instances (44.44%) of contact following join in the bottlenose dolphins, the dolphin that initiated the contact was also the dolphin that joined another
(i.e., initiated the join behavior) (Figure 10). The percentage of total animals in each age 51
class that was both the dolphin that joined another as well as the initiator of contact was
also determined (Figure 12).
50 47
45
40
35 31 30 Total # 25 # of joiners/initiators 20
15
10 6 4 5 0 001 0 Adult Subadult Juvenile Calf
Figure 12. Total number of identified dolphins in the Japan study group by age class and the number of dolphins that were both the joiner and initiator of contact.
The behaviors crossover (xxo) and crossunder (xxu) were also significantly associated with join in the spotted dolphins. In the two instances where xxo occurred after animals joined, the dolphin that crossed over another was also the animal that joined another (one juvenile and one calf). In the five instances of a dolphin crossing under another after joining (xxu), the “crosser” was also the dolphin initiating the join behavior on three occasions. The age breakdown of these three animals was one adult, one sub- adult, and one juvenile.
The behavior crossover (xxo) and crossunder (xxu) were also significantly associated with join in the bottlenose dolphins. In the two instances where xxo occurred 52
after dolphins joined, the animal that crossed over another was not the dolphin that joined
either time. In the five instances of a dolphin crossing under another after joining, the
“crosser” was also the joining animal only twice and both times this individual was a sub-
adult.
Overall, contact behavior is associated with the joining of dolphins in both species. In the spotted dolphins 23.94% of all join behaviors were followed (within three lags) by a contact behavior. In the bottlenose dolphins 20.45% of all join behaviors were followed by a contact behavior (for a breakdown by social context, see Table 10 and
Figures 13 & 14).
Table 10 Percentage of join Occurrences that were Followed by a Contact Behavior
Bahamas Japan Inquisitive 25.00% 0.00% Forage 0.00% 50.00% General Social 32.00% 33.33% Play 15.79% 23.53% Travel 31.25% 12.50% This percentage was calculated by dividing the number of join behaviors that was
followed by a contact behavior by the total number of join behaviors for that context.
53
30
25 25
20 19
16 # of joins 15 # followed by contact
10 8 7 5 5 4 3 1 0 0 inquisitive forage gen. social play travel
Figure 13. Total number of join events in the Bahamas study group by bbc and the number of times that contact followed those join events.
18 17
16
14
12
10 9 # of joins 8 8 8 # followed by contact
6 4 4 3 2 2 1 1 0 0 inquisitive forage gen. social play travel
Figure 14. Total number of join events in the Japan study group by bbc and the number of times that contact followed those join events.
54
A two-way contingency table analysis was conducted to evaluate whether a contact behavior occurring after dolphins join occurred differently in the five behavioral contexts. The two variables were site with two levels (Bahamas for spotted dolphins, and
Japan for bottlenose dolphins) and broad behavioral context (bbc) with five levels
(inquisitive, forage, travel, general social, and play). Site and bbc were found to be significantly related (χ2=83.261; df 4; p<.001).
Subsequent to the contingency table analysis, follow-up pair-wise comparisons were conducted to evaluate where the significant differences occurred (Table 11). The
Holm’s sequential Bonferroni method was used to control for Type I error at the .05 level across all comparisons.
Table 11 Results for the Pair-wise Comparisons (for join-contact) using Holm’s Sequential Bonferroni Method
Comparison Pearson p-value Required p-value for chi-square significance forage vs. inquisitive 75.00 .000 .005* forage vs. travel 52.561 .000 .0055* forage vs. gen. social 34.106 .000 .0062* forage vs. play 24.324 .000 .0071* inquisitive vs. play 23.78 .000 .0083* inquisitive vs. gen. social 20.04 .000 .0100* inquisitive vs. travel 9.101 .003 .0125* play vs. travel 7.885 .005 .0167* gen. social vs. travel 4.846 .028 .025 gen. social vs. play .850 .356 .05 (* indicates significance)
55
Associations - Contact
The analysis of the target event contact included examination of both negative
(-1, -2, and -3) as well as positive (+1, +2, and +3) lag behaviors (Table 12). Behaviors leading up to, as well as behaviors occurring after a contact event might assist in
determining whether the contact behavior itself has a communicative effect. A change in
behavior is an indication that communication may have occurred; therefore significantly
occurring behaviors that are produced prior to the contact might in some way be eliciting
the contact. Similarly, behaviors that are produced after the contact event might be a
result of the contact occurring. For ease of presentation, all contact behaviors are
grouped here as the target event, with significantly associated behaviors listed. However,
more detailed information regarding which contact behaviors were associated with which
other behaviors may be found in Appendices B and C.
56
Table 12 List of Significant (Z> 3.291) Behaviors that Occurred before (lags -1, -2, or -3) and After (lags +1, +2, or +3) the Target Event contact
Before contact After contact Bahamas Japan Bahamas Japan cch (p) bbs (s) cla (p) bbs (p) cla (s) cme (s) cme (t) cme (s) cme (s,p) cpc (i,s,p) cpc (i,s,t) cpc (i,s,p) cmo (i) xxu (s,t) cro (s,p,t) xxu (f,t) cpc (s,p,t) flx (s) cve (s) xxo (i,s) cro (s,p) hsc (p) xxu (p) dep (s,p) xxu (i,s) joi (f,s,p) dep (f,t) gsm (i,s) xxo (s,t) psm (i,s) gsm (p,t) hjk (t) gsm (s,t) pjk (p) hjk (t) joi (f) joi (i,s,p,t) rst (t) jop (p) psm (f,i,s,p,t) psm (i,s,t) sps (i) joi (t) pjk (p) rzz (s) vps (s) psm (i,f,p) sps (i) rst (f,p) rzz (s) tjk (f) sps (s) rst (t) ssm (s) srb (s) tjk (p) sps (s) vps (p) snk (s) vps (s,p) The broad behavioral context is noted in parentheses: f=forage, i=inquisitive,
s=general social, p=play, and t=travel. Note: some behaviors occurred in multiple contexts.
Seventeen behaviors were significantly associated with contact in the negative lag
positions (lags -1, -2, or -3) in the spotted dolphins. These behaviors include the
expected swim behaviors (gsm, psm, sps, and vps) as well as the previously discussed
join (joi), crossover (xxo), and crossunder (xxu). In addition, six contact behaviors (cla,
cme, cmo, cpc, cro, rzz) are present in this list. Contact behaviors were often seen
following one another. For example, a pectoral rub by one dolphin might elicit a pectoral rub from another individual. Eighteen behaviors were significantly associated with 57
contact in the positive lag positions (lags +1, +2, and +3). Twelve of these behaviors
were those that were also significantly associated with the negative lag positions.
Twelve behaviors were significantly associated with contact in the negative lag
positions in the bottlenose dolphins. As with the spotted dolphins, these behaviors
include the expected swim behaviors (psm, sps, and vps) as well as join (joi) and
crossunder (xxu). Two contact behaviors (cme, cpc) are significantly associated with the
negative lag positions in this species. Thirteen behaviors were significantly associated with contact in the positive lag positions, eight of which were also deemed significant in the negative lag positions.
Adventitious Finding: Oscillating Swim
While watching and scoring the video segments for this project, a previously
undocumented behavior was exhibited by individuals of both study groups. This
behavior, henceforward called an ‘oscillating swim’ is best described as a rolling of the
head from one side to the other (often in a figure eight pattern) as the body follows while projecting forward. It is often performed at a higher rate of speed than a normal swim and is frequently accompanied by abrupt changes in direction. The behavior was seen in single animals 185 times, and in pair-swimming animals (where one or both animals were exhibiting this behavior) 15 times.
Performance of the oscillating swim was similarly distributed in dolphins across social contexts (χ2 = 4.43; df = 4; p>.05) in both sites. However, there was a difference
in the sexes of dolphins performing this behavior between species (χ2 = 3.98; df = 1; p<.05) (Figure 15). 58
60 54 51 50
40
29 total # 30 28 26 # performing osc
20 14 10 10
2
0 female male female male
Bahamas Japan
Figure 15. Total number of identified dolphins and number observed performing oscillating swim by sex and site.
In the bottlenose dolphin, 52% of all the females and 51% of all males identified and observed in this project were seen performing the oscillating swim. For the Atlantic spotted dolphin, 34% of all females and 14% of all males performed this behavior. There was a species-specific difference in the ages of the dolphins performing this behavior
(χ2 = 24.241; df = 3; p <.001). Only one bottlenose calf was observed in this project and
that calf was not seen exhibiting this behavior. However, all other age classes are
represented for both study species (Figures 16 & 17).
59
14
12 12 12
10
8 total # # performing osc 6 5
4 3 3
22 2 1
0 Adult Subadult Juvenile Calf
Figure 16. Total number of identified dolphins and number of dolphins observed performing the oscillating swim in the Bahamas study group by age.
50 47
45
40
35 31 30 30
total # 25 # performing osc
20
15 12
10 6 4 5 1 0 0 Adult Subadult Juvenile Calf
Figure 17. Total number of identified dolphins and number of dolphins observed performing the oscillating swim in the Japan study group by age.
60
Subsequent to the contingency table analysis, follow-up pair-wise comparisons were conducted to evaluate where the significant differences occurred (Table 13). The
Holm’s sequential Bonferroni method was used to control for Type I error at the .05 level across all comparisons. Significant pair-wise differences were found between juveniles and sub-adults, calves and sub-adults, and juveniles and adults.
Table 13 Results for the Age Pair-wise Comparisons using Holm’s Sequential Bonferroni Method
Comparison Pearson p-value Required p-value for chi-square significance juvenile vs. sub-adult 18.358 .000 .008* calf vs. sub-adult 11.648 .001 .01* juvenile vs. adult 8.801 .003 .0125* calf vs. adult 5.63 .018 .0167 calf vs. juvenile .545 .460 .025 sub-adult vs. adult .540 .462 .05 (* indicates significance)
CHAPTER V
DISCUSSION
Sequences of Behavior
The search for sequences (incorporating greater than two behaviors) of communicative behavior(s) in these two dolphin species did not yield the expected results. Several factors could have contributed to this lack of confirmed sequenced patterns of behavior. A methodological explanation would point to the removal of groups of more than four animals from the analysis.
Perhaps dolphins use a more rigid (and hence patterned) response system when in a large group. This alludes to the levels of communication discussed earlier. Levels of subtlety (subtle versus overt behavior) might be utilized as well as levels of flexibility in the signal itself (flexible versus rigid) when dolphins are in a large group setting. If a dolphin is trying to convey information to a large number of individuals who are not in direct proximity to the sender of the signal, subtle behavior might go unnoticed.
Therefore, the use of more overt behavior and possibly more overt and recognizable patterns or sequences might be seen in larger groups. The more subtle behavior may function well within a small group setting; however understated behavior is also likely to go unnoticed by human researchers. Similarly, if dolphins are in a larger group they may not use the same signals to convey information that they would use with close associates.
A more rigid behavior pattern might be employed in the large group as a result of 62
diminished familiarity with group members. With close associates, flexibility could be
more easily tolerated due to the greater amount of time spent with the sender of the signal
(and hence the possibility of an increased familiarity with the sender’s repertoire). In
either case, by eliminating the larger groups of dolphins this information is unavailable to the current study.
An alternative explanation for the absence of longer behavioral sequences is that dolphins may not use sequences to communicate at all. Perhaps a single behavior is all that is required to successfully relay information from one individual to another. Longer sequences (incorporating three or four behaviors) are undoubtedly more complex, requiring increased cognitive functioning. Although presumably cognitively capable (see
Herman, 1980, 2002; Reiss & Marino, 2001), Morgan’s canon would warn against assigning this higher-order cognitive explanation when a lower-order explanation can be found for not detecting consistent sequences of behavior.
The removal of young animals from the analyses was done to reduce the possible effect of behavioral flexibility often seen in young animals (see Bekoff & Byers, 1998;
Fagen, 1981; Vandenberg, 1978). However, separate analyses for sub-adult and adult animals versus juveniles and calves still did not reveal any consistent sequences. This could be another indication that dolphins might not be using sequences of behavior to communicate or at least, that consistency is not a factor in information exchange.
Despite the fact that no significant sequences of behavior were found, it is unlikely that these dolphins’ communicative behavior is completely random. In order for communication to be effective, both the sender and receiver of the signal must have some
expectation as to the meaning of the signal transmitted (Smith, 1990). However, there 63
are situations in which it would be more beneficial to have a more specific
communicative repertoire (e.g., group herding of prey). Although both of these species
have been observed engaging in this activity (Acevedo-Gutiérrez, 1999; Fertl & Würsig,
1995; Würsig & Würsig, 1979), the subset of data utilized here did not contain group
herding of prey. The lack of any occurrences of group herding behavior or other behavior
requiring a high level of coordination in the observations could be one explanation for the
absence of sequential behavior.
It is more likely that the individuals represented in this project have a flexible
repertoire of communicative behaviors. According to Hauser (1996), the informational
content inherent in signals can be manipulated by a sender and differentially acted upon by a receiver. This behavioral flexibility is especially important in groups where the context of the signal is important in determining meaning. Individuals living in a social group must continually anticipate the course of developing events, flexibly adjusting their behavior based on both current stimuli and prior experience. So too, must individuals adjust their expectations based on the idiosyncrasies of signal use that might be characteristic of their associates. This may help to explain why the patterns of behavior seen were not consistent, but significant associations between behaviors were uncovered.
Overall, behavioral flexibility is advantageous. This advantage is evident not only in the ability to adjust behavior according to the context of a signal and peculiarities of the sender, but in the ability to use a single signal in multiple contexts (Smith, 1990).
The use of a single signal in multiple contexts facilitates communication while reducing cognitive costs. Ultimately, flexibility is advantageous because it can facilitate dealing 64
with ontogenetic changes in behavior as well changes that may occur in the environment,
thus increasing the overall fitness of the individual (Smith, 1985).
Behavioral Associations
Associations – Depart
Overall, there were a variety of behaviors that were significantly associated with
the target event depart in each study group. Some of these behaviors manifested in both
study groups (xxo, gsm, joi, psm, and vps). None of the common behaviors were
significantly associated with the target event in the same broad behavioral context in both
species. This could be due to actual differences in the way the dolphins use the behaviors,
environment constraints, or a combination of both. For example, the behavior srb (sand
rubbing) was significantly associated with depart in the spotted dolphin group.
However, due to the characteristics of the ocean floor in the area around Mikura island
(volcanic, generally boulder-strewn), it is highly unlikely that this behavior would be
exhibited within the bottlenose dolphin study group. The behavior bbs (bubble stream)
was significantly associated with depart in the bottlenose dolphin group. The production of bubbles is often accompanied by the production of whistles (Pryor, 1980). This study did not examine vocal behaviors, so confirmation of the simultaneous occurrence of these two behaviors is unavailable. However, the dolphins in the Japan study group may use a more vocal repertoire of communicative signals due to the diminished underwater visibility (K.M. Dudzinski, personal communication, 2004). The production of bubbles could act as a visual signal that a whistle is being produced. 65
It is not surprising to see the three swim behaviors (gsm, psm, and vps) associated with animals departing in both groups. In general, animals would be swimming in some sort of a pair or group situation for departing to be possible.
In both the Atlantic spotted and the Indo-Pacific bottlenose dolphins studied, contact behaviors as well as crossing behaviors (i.e., xxo or xxu) were found to be significantly associated with the target behavior depart. The four contact behaviors in the spotted dolphins were evidenced in three of the five behavioral contexts (forage, travel and play). In the bottlenose dolphins only two contact behaviors were significantly associated with depart and were observed in three of the five potential behavioral contexts (general social, inquisitive, and play).
The incidence of contact occurring prior to the departure of an animal in these contexts could be a signal that the animal is about to depart. This was examined by determining how many times the animal that departed was also the same one that initiated the contact. In the spotted dolphins, 28.57% (four out of 14) of the cases involving contact before departure also involved the same animal initiating both behaviors. In the bottlenose dolphin, 57.14% (four out of seven) cases were represented by a dolphin being both contact initiator and departer. There did not appear to be an age specificity for this behavior. In the spotted dolphins, two adults, one sub-adult, and one juvenile were the initiators, while in the bottlenose dolphins the initiators were one adult, two sub-adults, and one juvenile. However, with so few cases it is difficult to determine with certainty if age plays a factor in tactile contact with a peer before departing. It may be more important for dolphins in Japan (Indo-Pacific bottlenose dolphins) to indicate to another that it is about to depart because water visibility may deter them from maintaining visual 66
contact after departure. In contrast, underwater visibility in the Bahamas is better than
that in Japan providing the opportunity for prolonged visual contact after departure.
The behavior xxo was significantly associated with depart in both species (xxu
was only significantly associated with the bottlenose dolphins). The crossover behavior
was examined as to its potential communicative properties by examining which animal
produced this behavior in relation to the dolphin that departed. If the same dolphin
initiates both behaviors, the crossover behavior could be a signal that the dolphin is about to depart. In each species, three instances of xxo occurred prior to depart. In all six
cases (three from each species), the dolphin that crossed under the other was also the one
to depart. However, the production of this behavior may only be a function of the actual
act of departing. When one dolphin leaves another, it may change the direction in which it is swimming. The intended path of the departing animal may be perpendicular to the current direction of swimming, in which case it may have to cross over or cross under another animal to reach that goal. Therefore crossing behaviors could also have a locomotory function.
Associations – Join
Overall, there were a variety of behaviors that were significantly associated with the target behavior join in each study group. These behaviors included the expected swim behaviors (gsm, psm, sps, and vps) as well as shorter duration behaviors such as hjk
(seen in both study groups) and tjk (seen only in the spotted dolphin group). These behaviors could be potentially communicative. Maintenance of a pair swim behavior may be a signal indicating a willingness to maintain close proximity, merely by not bolting away when joined by another individual. Overstrom (1983) suggested that a head 67 jerk could be a step in an escalating agonistic encounter. However, no such encounter ensued in the dolphin group in which this behavior occurred.
Contact behaviors were significantly associated with the behavior join in both species. In the spotted dolphins, the only behavioral context not represented by a contact behavior occurring after join was forage. In the bottlenose dolphins, the behavioral context travel and inquisitive were not represented. In addition, crossover and crossunder were also seen to occur significantly after dolphins joined. However, for the reasons stated above these crossing behaviors may have a locomotory function as well as a potentially communicative function.
The occurrence of contact after dolphins joined was further examined for its role in potentially communicative exchanges. This was accomplished by determining how many times the dolphin that joined was also the individual that initiated contact. In the spotted dolphins, 76.47% (13 out of 17) of the cases of contact occurring after dolphins joined, the animal that joined was also the individual that initiated the contact. Ages of the dolphins that were initiator of both join and contact were varied (four adults, one sub- adult, four juveniles, one calf), but directly mirrored the demographics of the population
(12 adults, three sub-adults, 12 juveniles, three calves). This results in an equal distribution of this behavioral association across age classes in the spotted dolphins. For the bottlenose dolphins, 44.44% (four out of nine) of the cases of join-contact, the dolphin that joined was also the one that initiated the contact. In all four of these cases, the pairs were comprised of two sub-adults. Therefore, the initiator animal was a sub- adult (four different sub-adults). However, it is important to note that the age group ‘sub- adult’ had the most members (47) in the population as a whole. 68
The high percentage of occurrence of this behavioral combination coupled with
the uniform distribution across age classes suggests that the production of a contact
behavior after joining may be a regular part of the behavioral communicative repertoire
of the spotted dolphins. Contact after joining has been documented in other species (e.g.,
bottlenose dolphin, Tursiops truncatus, Saayman & Tayler, 1972; Würsig & Würsig,
1979; and for killer whales, Orcinus orca, Jacobsen, 1986) and it has been suggested that it may function as a form of greeting, especially when involving the pectoral fin
(Dudzinski, 1998).
The data from the bottlenose dolphin population belie the scenario outlined for the
spotted dolphins. In this study group, the dolphins were equally likely to make contact
before departing as after joining. This could indicate that overall, contact is involved
throughout their behavioral repertoire and does not necessarily play a more important role
in departing than in joining. It may not have a specific communicative function such as a
greeting behavior, but may function more in establishing and maintaining social bonds
through the population (Moehlman, 1987; Seyfarth, 1980), therefore who initiates the
behavior is not always important. The elimination of large groups of dolphins may have
contributed to the decreased occurrence of contact after joining seen in the Indo-Pacific
bottlenose dolphins.
Associations – Contact
In the Atlantic spotted dolphin study group, an analysis of significantly occurring
behaviors at both positive and negative lag positions for all contact target behaviors
revealed associations with a variety of behaviors. These behaviors include the expected
swim behaviors (gsm, psm, sps, and vps) seen in with the other target events as well as 69
some behaviors not associated with the target events depart and join such as rst, jop, and
snk.
The behavior rst (rest horizontal) is a behavior where the dolphin is not actively
swimming through the water. This type of generally motionless stance may be
functioning simply as a break from swimming. However, it could also facilitate contact
by providing a stationary target. On a more subtle level, if the function of the behavior is
truly one of resting, it could passively convey a signal that the dolphin is tired, or
potentially even sick or injured.
The behavior jop (jaw open) is generally regarded as an agonistic behavior
(Overstrom, 1983; Samuels & Gifford, 1997). However, it occurred in the play context, so could have been part of a rough and tumble play encounter between dolphins. Many of the behaviors seen in agonistic encounters are also seen in play bouts (Saayman,
Tayler, & Bower, 1973). Mock fighting and rough and tumble-type play assists young animals in learning social skills and the proper use of particular signals within their social structure (Bekoff & Byers, 1998; Fagen, 1981).
The behavior snk (sink) could have potentially communicative aspects as well. It has been suggested that this behavior could be a signal advertising the reproductive state of a female (K.M. Dudzinski, personal communication, 2003). In this study, the animal performing this behavior was a juvenile female (in the general social behavioral context), so although not yet sexually mature, she could have been ‘practicing’ this behavior for future use. Alternatively, this behavior may have been used to move from the top of the water column to the bottom. 70
In the Indo-Pacific bottlenose dolphin study group, several behaviors were
significantly associated with contact that were not associated with depart or join. These
behaviors were hsc (head scan) and flx (flex). Head scanning in dolphins is usually a visual action associated with the production of echolocation clicks. Although thought to
be used primarily in object detection and navigation (McBride, 1956; Norris, 1969), it has
been suggested that dolphins may also direct these clicks at another dolphin for a
stimulatory effect (Dudzinski et al., 2002). Echolocation can have a communicative
function in both of the above situations. A dolphin might “eavesdrop” on another
dolphin’s click train, thereby receiving information from an unintentional signal (Xitco &
Roitblat, 1996). Certainly in the second situation, a tactile signal is being conveyed.
Functionally, this intended signal could be affiliative, particularly since it is often seen
directed at the genital region.
The behavior flx (flex) was also only seen in this study group associated with the
target event contact. The flexing of the torso could simply be a part of the locomotory
function of swimming. However, it is also seen as a component of the “S-posture”
exhibited in aggressive encounters between adults. The “S-posture” is also seen during
juvenile play and is often accompanied by contact (Dudzinski et al., 2003). In this study,
the dolphin that produced this flex behavior was a subadult female interacting with a calf.
Immediately after flexing, the subadult female made contact with her rostrum to the calf’s
fluke. Here the signal might have been closer in meaning to the “S-posture” used in
playful encounters.
Other contact behaviors were also significantly associated with this target
behavior in both species. Contact of one dolphin with another may have several 71
communicative functions. It may function as the solicitation of reciprocated touch for
grooming or in the establishment, maintenance, or advertisement of social bonds (Norris,
Würsig, Wells, & Würsig, 1994; Östman, 1990; Pryor, 1990). The part of the body touched could also contribute to the signal’s meaning (Sakai et al., 2003).
If contact truly is a solicitation of touch (for whatever purpose) it should be reciprocated in a successful conveyance of the signal. However, this does assume the animal solicited is willing to reciprocate. To address this issue, the data were analyzed to determine how often the contact was reciprocated. In the spotted dolphins, there were seven cases of reciprocated contact with seven different animals participating (one adult, one sub-adult, five juveniles). All of these cases occurred in the general social or play
behavioral contexts. In the bottlenose dolphins, there were only three cases of
reciprocation with four different animals participating (three sub-adults, one calf).
However, these cases also all occurred in the general social or play behavioral contexts.
The reciprocation of contact behavior in these two contexts represents 31.25% of all
interaction events that included contact in the general social context and 20.00% in the
play context in the spotted dolphins. In the bottlenose dolphins, 16.67% of interaction
events that included contact were reciprocated in the general social context and 14.29%
in the play context.
These percentages do not strongly support the hypothesis that contact may
possess a communicative function solely as a solicitor of reciprocated touch. However,
other factors must be weighed in this analysis. This analysis looked only at the six (three
negative and three positive) lags around a contact behavior. It is possible that the
reciprocation of the contact did not occur immediately, but occurred later in the 72 behavioral chain of events. Another factor to be considered is that dolphins frequently move in and out of the camera’s view. The reciprocation of touch could have occurred out of the observer’s field of view. Another potential alternative is the willingness of a receiver of the contact signal to reciprocate. It is possible that touch was used as a communicative signal, but the receiver of the signal chose to ignore it (or the human observer was unable to perceive it).
Both study species engaged in contact-contact behaviors in the general social and play behavioral contexts. Although the general social and play contexts contained the most contact events in both species, these contexts also represent activity in which more socialization occurred. In the foraging context the focus is on procuring food. Travel has a focus on the movement from one area to another. In the inquisitive context, the focus of the dolphins’ attention is turned away from each other and is on something else in their environment (i.e., humans). This suggests that contact exchanges and socialization are associated with each other, supporting the hypothesis that contact could be a solicitation for something (here reciprocated contact) from the receiver of the signal.
The literature on cetacean contact suggests that dolphins use this type of signal as a means to maintain social bonds as well as in other situations such as aggression or sex
(Caldwell & Caldwell, 1977; Dudzinski, 1998; Evans & Bastian, 1969; Pryor, 1986;
Sakai et al., 2003). However, these situations do not always require a behavioral response from the receiver and hence make determining communicative function for the observer more difficult. Despite the overall paucity of support for contact as a request for reciprocal behavior arising from this study, the consistency of occurrence in the more social of the behavioral contexts suggests otherwise. 73
The spotted dolphins in this study are utilizing a wider variety of contact
behaviors. This was consistently seen when contact occurred prior to an animal
departing, after animals joined, and when contact occurred after another contact behavior.
Perhaps for these dolphins, contact is serving more of a communicative function. The environment in which this study group lives may facilitate a more visual and tactile
communication system. The Indo-Pacific bottlenose dolphins in this study live in an
environment where a more vocal communicative repertoire may be beneficial due to the
reduced underwater visibility.
Adventitious Finding: Oscillating Swim
The oscillating swim is a form of swimming that is structurally different from a
normal, forward projecting movement by dolphins through the water. The movements
involved are more exaggerated than normal swimming. Observations of the oscillating
swim also indicate that it is a purposeful motion and not one that just happens to occur
while the dolphin is swimming. Therefore, it is likely that this oscillating swim uses
more energy than its normal counterpart. A normal swim as a means of locomotion, has
the lowest energetic cost (Tucker, 1975), where energetic cost is measured as the rate of
oxygen consumption per unit of time while performing a ‘steady-state’ (as opposed to
‘sprint-type’) locomotive motion (McArdle, Katch, & Katch, 1991). Steady-state
locomotion refers to a relatively uniform type of movement that is fueled by aerobic
means via the process of oxidative phosphorylation, so the rate of oxygen consumption
becomes a good indicator of the rate of energy production (Steudel, 2000).
Ease of movement through the water is dependent upon the shape of the object
and its position in the water. Due to the high density of water (relative to air), a fusiform 74
shape is the most energetically beneficial (Tucker, 1975). This torpedo-like shape allows
water to flow over and around the body thereby reducing drag (Williams et al., 1992).
The oscillating swim actually increases drag by positioning parts of the body at an angle
that catch or push the water (reducing the hydrodynamic flow), thus decreasing the
efficiency of the swim. Since this type of movement is less energy efficient both in the
energy required to produce it and in the energy required to maintain it while swimming through the water, it would need a beneficial purpose for the dolphin to offset its energetic cost.
Due to the higher energy cost associated with the oscillating swim, it is unlikely that this type of swim behavior is an indicator of aggression. Performing this behavior uses up energy sources that could be used to fight or flee in an aggressive situation therefore depletion of these stores would be a disadvantage. Moreover, no aggressive threats (such as jaw pops or open mouth displays) were observed during the performance of this behavior. It is also not likely that this motion is performed solely for a stimulatory effect (the same motion in humans results in a dizzy feeling that some find pleasurable).
The anatomy of the dolphin ear does not support this alternative. Although the cetacean middle and inner ear structures are very similar to those of other mammals (Ketten,
2000), the semicircular canals are attenuated (Gao & Zhou, 1995; Ketten, 1992).
Humans and cats without functional semicircular canals do not exhibit motion sickness
(Graybiel, 1964). Therefore, the reduction of this structure could reduce the effect of rotational motion in cetaceans, thereby eliminating the sensation associated with this type of movement. 75
I suggest that the oscillating swim has a communicative function, perhaps akin to the ‘pointing’ behavior described by Xitco and colleagues (2001) (also see Dudzinski et al., 2003). The movement of the animal around the object of interest (in all cases observed herein, this object is a human) may stimulate something akin to the orienting reflex in humans (Cowan, 1995; Pavlov, 1927). Cowan noted that the kinds of stimuli that trigger this reflex fall into two categories: stimuli that are significant for the organism, and stimuli that are novel. This orienting focuses the organism so it can devote
attention to the stimulus if warranted. By performing this type of swim behavior that is not the ‘norm’, the performing animal is calling attention to itself, therefore by association, also calling attention to the object(s) to which it is attending. The oscillating swim, performed while circling an object, could be a way of indicating something novel or different in the environment of which perhaps others should be mindful.
CHAPTER VI
CONCLUSION
It is difficult to establish a firm boundary between stimuli produced by animals that are communicative and those that are not (Hauser & Konishi, 1999). Many cases fall directly on the border, possessing some of the characteristics that are required for communication while lacking others. For example, telling a man to jump off a bridge is an act of communication, but pushing him off the bridge is not (Cherry, 1966). Others could argue that in this case, the point was certainly made, albeit in a very different manner.
The study of non-human animal communication is even more difficult. We cannot know for certain what the internal affective state of the animal is. We cannot absolutely attribute motivation, nor infer intentionality. However, we can look at the behavior from an external viewpoint and attempt to ascribe behavioral changes to a potential exchange of some kind of information. This study provided information on what behaviors were significantly associated with specific, potentially communicative situations.
Every perceivable behavior that an organism exhibits can convey information to conspecifics. This study revealed a variety of behaviors that were significantly associated with situations that might precipitate the conveyance of information. These situations included the departing or joining of dolphins as well as contact between dolphins. The decision to more closely examine certain behaviors that were significantly 77
associated with these target events was based on previous research into the possible functions of contact between dolphins (Dudzinski, 1996, 1998; Saayman & Tayler, 1972) and the occurrence of certain behaviors across both study groups, different age classes, and different broad behavioral contexts.
Overall, contact behaviors were significantly associated with joining and departing in both the spotted dolphins and the bottlenose dolphins. However, the spotted
dolphins were more likely to use contact after joining than before departing, whereas the
bottlenose dolphins were equally likely to use contact in both situations. Contact behaviors were also significantly associated with other contact behaviors in both species.
Both species exhibited documented reciprocal exchanges of contact in only two
behavioral contexts (general social and play). The wider variety of contact behaviors
exhibited by the spotted dolphins may be an indication that contact is serving a more communicative function in this population of animals. The environment (good visibility) is optimal for a communicative repertoire that emphasizes visual and tactile signals in this study group. The Indo-Pacific bottlenose study group produces three times as many sounds as those dolphins in the Atlantic spotted dolphin study group (Dudzinski, 1999).
Therefore, the bottlenose dolphins may rely more on an auditory based signal system due to the reduced underwater visibility.
Limitations and Future Study
Several limitations have been identified in this study. The choice of behaviors to
score may not have been those that these dolphins use in their repertoire of
communicative behaviors. This project may have focused on behaviors extraneous to the
process of communication. Similarly, perhaps the meaningful communicative behaviors 78
employed by these dolphins are behaviors not observed either due to camera angle or
subtlety in their use by dolphins. A slight change in position, swimming speed, or glance may have a profound effect on the information transferred from one individual to another, but for human observers these behaviors may go unnoticed.
This study was also limited to the behavior of individuals captured on film.
However, animals may be reacting to signals from others not in the camera’s field of view. Similarly, animals out of view of the camera may be reacting to those signals sent by animals within the purview of the camera’s lens. Both of these scenarios could result in an incomplete sequence of behavior being documented.
Throughout all video/audio recordings of both species in this study, humans were present in the water. This human presence was necessary to collect the data, but may have provided a basis for behavior alteration. Although the populations studied were habituated to human swimmers, we cannot truly assess the effect on behavior that a human presence may have evoked in these animals. However, we have no way to observe and record dolphin behavior in these two wild groups without humans being present.
Future study may concentrate on dolphins in a captive or semi-captive environment with continuous video recording and no humans present in the water. An optimum situation might include several cameras set at different angles to cover as much of the living space as possible. This would solve the problem of a human presence in the water and provide several angles of viewing as well as provide for the opportunity for an experimental design. APPENDIX A BEHAVIOR CODES AND DESCRIPTIONS
Category 1 - General Behaviors: BEHAVIOR NAME CODE TYPE DESCRIPTION barrell roll brl Event roll body 360 degrees while not projecting forward
bottom grubbing btg State inverted vertically; dolphin rostrum near or at sea floor, dolphin rotating peduncle first and head follows
corkscrew ckw State barrel roll while in forward motion
feeding fed State dolphin is eating
bury flk in sand fik State dolphin actively buries flukes in the sandy ocean floor
fluke stand fks State dolphin vertical with flukes touching sand
flex flx Event head and tail move concave and convex alternately
pick up object piu Event dolphin picks up object with mouth, fluke, dorsal or pec fin
rest horizontal rst State dolphin is motionless horizontal in water column or at surface
sand rubbing srb State rubbing all part of the body in the sand
sink snk State slowly move deeper in the water column, any position
sommersault som Event forward flip underwater
suspend vertical sus State dolphin is suspended vertically in the water column
s position spo Event dolphin exhibits classic s-posture
solo swim ssm State animal is swimming alone 80
Category 1 - General Behaviors (continued): BEHAVIOR NAME CODE TYPE DESCRIPTION
group swim gsm State animal is swimming with more than one other
oscillating swim wsm State animal is swimming while moving back and forth, flexing vertically and horizontally, often involves changes in direction
out of view oov State animal is out of view
circle swim csm State animal swims in a circle around an object
head up swim hdu State swim with head up
head down swim hdd State swim with head down
circle chase cch State dolphins circling each other while swimming fast
chase chs State one or more dolphins swiftly following other(s)
cross in back of xxb Event animal crosses in back of another
cross in front of xxf Event animal crosses in front of another
crossover xxo Event animal crosses over another
crossunder xxu Event animal crosses under another
departing dep Event animal departs company of other(s)
ventral pr swim vps State animal is swimming with another while in the ventral up position
side pair swim sps State animal is swimming with another while in the side up position
joining joi Event animal joins another animal
pair swim psm State animal is swimming with another animal within 2 meters
81
Category 2 - Contact Behaviors: BEHAVIOR NAME CODE TYPE DESCRIPTION
full body bod Event dolphin rolls over another, with full body contact
reciprocal nuzzle rzz State dolphins rubbing rostrums against each others' bodies
contact dorsal fin cdf State animal makes contact with another dolphin using its dorsal fin
contact dorsal back cdb State animal makes contact with another using the dorsal side of its body
contact fluke cfl State animal makes contact with another using its fluke
contact genitals cge State animal makes contact with another using its genital region
contact lateral cla State animal makes contact with another using the lateral portion of its body
contact melon cme State animal makes contact with another using the melon
contact mouth cmo State animal makes contact with another using its mouth
contact pec fin cpc State animal makes contact with another using its pec fin
contact peduncle cpd State animal makes contact with another using its peduncle
contact rostrum cro State animal makes contact with another using its rostrum
contact ventral cve State animal makes contact with another using its ventral side
Category 3 - Bubble Behaviors: BEHAVIOR NAME CODE TYPE DESCRIPTION
bubble bbl Event single bubble from blowhole
bubble stream bbs State several small bubbles produced in a stream
82
Category 4 - Head Movements: BEHAVIOR NAME CODE TYPE DESCRIPTION
head jerk hje Event single quick deliberate movement of head
head scanning hsc Event moving head laterally side to side
Category 5 - Jaw Movements: BEHAVIOR NAME CODE TYPE DESCRIPTION
jaw clap jcp Event dolphin opens and closes jaws
jaws open jop Event jaws are open
jaws open/close joc Event jaws are opening and closing
Category 6 - Pectoral Fin Movements: BEHAVIOR NAME CODE TYPE DESCRIPTION
pec jerk pjk Event movement up and down of pectoral fin, flaring pecs
Category 7 - Tail Movements: BEHAVIOR NAME CODE TYPE DESCRIPTION
tail jerk tjk Event tail movement vertically or laterally
APPENDIX B SCREE PLOTS BY SITE AND BROAD BEHAVIORAL CONTEXT FOR ALL TARGET EVENTS
Bahamas
50.0% 45.0% 40.0% 35.0%
30.0% % .0 25.0% 0 k 2 fi 20.0% 8% 10. 15.0% i jo 10.0% 5.0% 0.0%
Plot 1 General Social, depart
50.0% 45.0% % .3 40.0% 3 e 3 35.0% cm 30.0% 25.0%
20.0% 0% 10. 15.0% rb s 7% 10.0% 4. pc c 5.0% 0.0%
Plot 2 Travel, depart
84
50.0% 45.0% 40.0%
35.0% % 0 5. 2 30.0% o cr 25.0%
20.0% % .9 15.0% m 8 ps 10.0% 5.0% 0.0%
Plot 3 Forage, depart
50.0% 45.0% 40.0% 35.0%
30.0% .0% % 19 7 25.0% m . % 3 gs . % la 16 4 8 20.0% c s 1 % 11. .2 vp o 0 9% 15.0% xx i 1 8. jo c 10.0% cp 5.0% 0.0%
Plot 4 Play, depart
50.0% 45.0% 40.0%
35.0% 0% % 5. .0 30.0% 2 25 h p cc jo 25.0% 20.0% 15.0% 10.0% 5.0% 0.0%
Plot 5 Inquisitive, depart
85
50.0% 45.0% % % .3 .3 % 40.0% 3 3 3 33 l xo cf x g 31. 35.0% bt 30.0% % 2 25.0% 7. 1 % sm .3 20.0% g 3 p 1 15.0% de 10.0% 5.0% 0.0%
Plot 6 Social, join
50.0% 45.0% % 3 40.0% 3. 3 o 35.0% cr 30.0%
% 25.0% .7 6 1 0% k 20.0% hj 14. % c .0 cp % 15.0% .1 xo 10 x m 7 10.0% gs 5.0% 0.0%
Plot 7 Travel, join
50.0% 45.0% .3% 40.0% 33 m 35.0% gs 30.0% 0% 9. 25.0% 1 m 20.0% ps 15.0% 10.0% 5.0% 0.0%
Plot 8 Forage, join
86
50.0% 45.0% 40.0%
35.0% % .0 5 30.0% z 2 rz 25.0% 9% 4. % % .3 .0 % 20.0% p 1 4 2 1% .8 1 de jk m 1 10 t s 11. 15.0% g u m xx ps 10.0% 5.0% 0.0%
Plot 9 Play, join
50.0% 45.0% % % .3 .3 3 3 40.0% , 3 , 3 e h m 35.0% cc c 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0%
Plot 10 Inquisitive, join
50.0% 45.0%
40.0% .3%
xo 33 35.0% x 30.0% 25.0% 20.0% .1% 15.0% 9 ps s 10.0% 8% % .4 i 3. 1 jo 5.0% sm s 0.0%
Plot 11 Social, CFL (negative lags) 87
50.0% 45.0% 40.0% % .6 35.0% 8 2 la 30.0% c % 25.0% .7 % 3 16 . % c 3 4 20.0% cp 1 . u 11 15.0% xx m ps 10.0% 5.0% 0.0%
Plot 12 Social, CPC (negative lags)
50.0% 45.0% 40.0% 35.0% 30.0%
% 25.0% 4 20.0% z 17. .3% rz tg 13 15.0% b % .7 7 10.0% o cr 5.0% 0.0%
Plot 13 Social, CRO (negative lags)
50.0% 45.0% 40.0% 35.0% 30.0% % .7 25.0% 6 1 20.0% e cm 1% 15.0% 9. s 10.0% sp 5.0% 0.0%
Plot 14 Social, CVE (negative lags)
88
50.0% 45.0% 40.0% % .6 8 35.0% 2 ro c 30.0% 25.0% 20.0% % .4 15.0% 7 sm 10.0% g 5.0% 0.0%
Plot 15 Social, RZZ (positive lags)
50.0% 45.0% % 40.0% .3 33 ve 35.0% c 30.0% 25.0% 20.0% % .1 15.0% s 9 p s 10.0% 5.0% 0.0%
Plot 16 Social, CME (positive lags)
50.0% 45.0% % .3 40.0% 3 k 3 35.0% sn 30.0% 25.0%
20.0% 1% 1. s 1 15.0% vp 10.0% 5.0% 0.0%
Plot 17 Social CVE (positive lags)
89
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0%
15.0% 3% 1% z 8. z ro 7. 10.0% r c 5.0% 0.0%
Plot 18 Social, CRO (positive lags)
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% 15.0% % 9 % % 10.0% 2. .9 .8 m m 2 2 ps pc 5.0% gs c 0.0%
Plot 19 Travel, CME (negative lags)
50.0% 45.0% 3% 40.0% 3. o 3 35.0% xx
30.0% % 9 % .6 20. 8 25.0% c 1 cp m gs 20.0% 9% 1. 1 i 15.0% jo % .6 10.0% m 5 ps 5.0% 0.0%
Plot 20 Travel, CPC (negative lags)
90
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% 15.0% % 9% % 8 10.0% .7 2. . 3 m 2 joi c 5.0% gs cp 0.0%
Plot 21 Travel, CRO (negative lags)
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% 15.0% 7% 7. p % de 8% .8 10.0% . 2 i 3 m 5.0% jo gs 0.0%
Plot 22 Travel, CME (positive lags)
50.0% 45.0% % .3 % .3 40.0% 3 3 e 3 3 m ro 35.0% c c % 30.0% .9 0 2 % 25.0% pc .7 c 6 1 k 20.0% hj 15.0% .7% t 6 10.0% rs 5.0% 0.0%
Plot 23 Travel, CPC (positive lags)
91
50.0% 45.0% % .3 3 40.0% 3 h 35.0% cc 30.0% 25.0% 20.0% 15.0% 10.0% 1% % c 3. .3 i 2 5.0% cp jo 0.0%
Plot 24 Play, CLA (negative lags)
50.0% 45.0% % 40.0% 3 33. z 35.0% rz 30.0% 25.0% 3% % 4. .3 20.0% s 1 4 k 1 vp tj 15.0% 10.0% 5.0% 0.0%
Plot 25 Play, CPC (negative lags)
50.0% 45.0% .3% 40.0% 3 e 3 m 35.0% c 30.0% 25.0% % 3 20.0% 13. o 15.0% cr 10.0% 5.0% 0.0%
Plot 26 Play, CRO (negative lags)
92
50.0% 45.0% 40.0% 35.0% 30.0%
% 25.0% 7 6. a 1 20.0% cl 15.0%
10.0% % .3 i 2 5.0% jo 0.0%
Plot 27 Play, RZZ (negative lags)
50.0% 45.0% 40.0% 35.0% % 2 30.0% s 22. p 25.0% v 20.0%
15.0% % 9 . % c 4 1 10.0% 3. cp p 5.0% de 0.0%
Plot 28 Play, RZZ (positive lags)
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% % .0 15.0% % m 8 7 s 6. g ro 10.0% c 5.0% 0.0%
Plot 29 Play, CME (positive lags)
93
50.0% 45.0% 40.0% 35.0% 30.0% 2% 22. s 25.0% vp .7% 16 % 20.0% la 1 c 1. 1 15.0% u xx 10.0% 5.0% 0.0%
Plot 30 Play, CPC (positive lags)
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0%
% 15.0% 7 p 6. % 10.0% jo .0 2 sm 5.0% p 0.0%
Plot 31 Play, CLA (positive lags)
50.0% 45.0% 40.0% 35.0% 30.0% % 25.0% 7 6. 1 a 20.0% cl 15.0% % .0 10.0% 4 sm 5.0% g 0.0%
Plot 32 Play, CMO (positive lags)
94
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0%
15.0% % .1 7% 6. .3% i 7 u 4 10.0% jo xx sm p 5.0% 0.0%
Plot 33 Inquisitive, CME (negative lags)
50.0% 45.0% 40.0% 35.0%
% 30.0% .0 20 25.0% e cm 20.0% % 7 15.0% 8. m 10.0% ps 5.0% 0.0%
Plot 34 Inquisitive, CPC (negative lags)
50.0% 45.0% % 3 40.0% 3. c 3 35.0% cp 30.0% 25.0% % 20.0% .5 10 15.0% m ps 10.0% 5.0% 0.0%
Plot 35 Inquisitive, CME (positive lags)
95
50.0% 45.0% 40.0% 35.0% 30.0% 25.0%
20.0% .1% 11 p 15.0% de % 4 10.0% 2. m 5.0% ps 0.0%
Plot 36 Forage, CRO (positive lags)
Japan
50.0% 45.0% 40.0% 35.0% 30.0% 25.0%
20.0% % .1 1 15.0% s 1 % vp .7 7% 6 5. e .0% % 10.0% c .2 cm cp s 3 b 2 b sc 5.0% o 0.0%
Plot 37 Social, depart
96
50.0% 45.0% 40.0%
35.0% % % 0 .0 . 25 30.0% 25 m ks c gs 25.0% 20.0% 15.0% % % .3 % .2 % 8 10.0% k 6 4 7 tj s . 1. sp i 2 m 5.0% jo ps 0.0%
Plot 38 Travel, depart
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% 15.0% % .3 10.0% c 5 .1% % .5 cp t 3 5.0% rs c 1 os 0.0%
Plot 39 Play, depart
50.0% 45.0% % 40.0% 3 33. u 35.0% xx 30.0% 25.0% % 20.0% .5
xo 12 15.0% x % 3 % 6. .3 % % m .0 % 4 10.0% c 5 .6 gs i 5 3 3. cp jo k m 5.0% hj ps 0.0%
Plot 40 Inquisitive, depart
97
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% % % .5 .5 20.0% 2 2 , 1 1 e s, m 7% % 15.0% c vp 7. .9 % , 6 6 xo p, , 5. 10.0% x de c cp 5.0% 0.0%
Plot 41 Social, join
50.0% 45.0% 40.0% 35.0% 30.0% % 25.0% 6 7. % .3 u 1 4 20.0% xx p 1 5% e 0. d 15.0% s 1 % sp 3 10.0% 4. m 5.0% ps 0.0%
Plot 42 Travel, join
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% % % 15.0% .3 .7 7 7 m pc 10.0% c ps 5.0% 0.0%
Plot 43 Forage, join
98
50.0% 45.0% 40.0%
35.0% % 0 25. 30.0% u xx 25.0% % 20.0% .5 12 me 15.0% c 5% 7. m % ps .0 10.0% 4 k 5.0% hj 0.0%
Plot 44 Play, join
50.0% 45.0% 40.0% 35.0% 30.0% % 25.0% 7 s 16. 20.0% p v % .5 15.0% % p 9 .8 de m 5 10.0% s p 5.0% 0.0%
Plot 45 Inquisitive, join
50.0% 45.0% 40.0%
35.0% % 0 30.0% xu 25. x 25.0% 3% 4. % 20.0% 1 .1 i 1 jo s 1 % 15.0% p .7 v 6 me 10.0% c 5.0% 0.0%
Plot 46 Social, CME (negative lags)
99
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 3% 4. 20.0% 1 x fl 15.0% % % 10.0% .2 0 3 c s 3. 5.0% cp bb 0.0%
Plot 47 Social, CRO (negative lags)
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0%
15.0% 7% 6. 3% % e 4. 9 10.0% m m c 2. ps pc 5.0% c 0.0%
Plot 48 Social, CPC (negative lags)
50.0% 45.0% 40.0% 35.0% 30.0% % .2 25.0% 7 p 1 20.0% de
15.0% % .6 10.0% c 5 cp 5.0% 0.0%
Plot 49 Social, CPC (negative lags)
100
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% % 15.0% .1 % .3 p 7 6 6% 4% e e 5. 10.0% d c 3. cm cp sm 5.0% p 0.0%
Plot 50 Social, CME (positive lags)
50.0% 45.0% 40.0% 35.0% 30.0%
25.0% % .3 14 20.0% m 1% gs 1. 15.0% o 1 xx 10.0% 5.0% 0.0%
Plot 51 Social, CRO (positive lags)
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% % 15.0% 1 % 7. .3 xu t 6 10.0% x rs 5.0% 0.0%
Plot 52 Travel, CPC (negative lags)
101
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% 15.0% % 0 % 10.0% 4. .8 m k 3 ps hj 5.0% 0.0%
Plot 53 Travel, CPC (positive lags)
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% 15.0%
10.0% 4% % 2. .6 c i 1 5.0% hs jo 0.0%
Plot 54 Play, CME (negative lags)
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% % .3 20.0% 14 jk p 15.0% % .3 10.0% c 5 cp 5.0% 0.0%
Plot 55 Play, CPC (negative lags)
102
50.0% 45.0% 40.0% 35.0% 30.0% 25.0%
% 20.0% 3 0. p 1 % 15.0% 6 de 5. .9% 10.0% pc s 4 % c b 8 b . i 1 5.0% jo 0.0%
Plot 56 Play, CPC (positive lags)
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% % .1 15.0% 9 jk p % 10.0% .4 1 sm 5.0% p 0.0%
Plot 57 Play, CME (positive lags)
50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0%
15.0% % % .3 .8 5 4 10.0% pc ps c s 5.0% 0.0%
Plot 58 Inquisitive, CPC (negative lags)
103
50.0% 45.0% 40.0% % 35.0% .7 26 m 30.0% gs 25.0% 20.0% % 7 15.0% 8. 1% xo x 6. .2% 10.0% ps 4 s pc c 5.0% 0.0%
Plot 59 Inquisitive, CPC (positive lags)
75.0% 7% 6. 70.0% 6 u 65.0% xx 60.0% 55.0% 50.0% 45.0% 40.0% 0% 35.0% 5. 2 30.0% tjk 25.0% 5% 20.0% 8. 15.0% m ps 10.0% 5.0% 0.0%
Plot 60 Forage, CPC (positive lags)
APPENDIX C
DATA USED IN DETERMINING PERCENTAGE OF OCCURRENCE OF SAB TO
TARGET EVENTS WITHIN PARTICULAR BBCS
The following table provides information regarding the number of total occurrences of
the sab within a particular bbc, the number of occurrences of that sab within three lags of
the listed target event, and the associated percentage of the occurrence of the sab within
the given bbc. Where plots were constructed, associated plot numbers are given (plots in
Appendix B). For the target event contact, positive or negative lag analyses are indicated
by a (+) or (-) respectively (i.e., cme (+) indicates the target was cme and the data given
were for analysis of positive lags).
Study Plot # BBC Target SAB Total # # of SAB w/i % Group of SAB 3 lags of target B 1 g. soc depart fik 10 2 20.0%
B 1 g. soc depart joi 74 8 10.8%
B 2 travel depart cme 3 1 33.3%
B 2 travel depart srb 10 1 10.0%
B 2 travel depart cpc 43 2 4.7%
B 3 forage depart cro 4 1 25.0%
B 3 forage depart psm 45 4 8.9%
B 4 play depart gsm 21 4 19.0%
B 4 play depart cla 6 1 16.7% 45
Study Plot # BBC Target SAB Total # # of SAB w/i % Group of SAB 3 lags of target
B 4 play depart vps 7 1 14.3%
B 4 play depart xxo 17 2 11.8%
B 4 play depart joi 59 6 10.2%
B 4 play depart cpc 45 4 8.9%
B 5 inq depart cch 8 2 25.0%
B 5 inq depart jop 4 1 25.0%
B 6 g. soc join cfl 6 2 33.3%
B 6 g. soc join xxo 3 1 33.3%
B 6 g. soc join btg 16 5 31.3%
B 6 g. soc join gsm 64 11 17.2%
B 6 g. soc join dep 60 8 13.3%
B 7 travel join cro 3 1 33.3%
B 7 travel join hjk 6 1 16.7%
B 7 travel join cpc 43 6 14.0%
B 7 travel join xxo 10 1 10.0%
B 7 travel join gsm 42 3 7.1%
B 8 forage join gsm 3 1 33.3%
B 8 forage join psm 42 8 19.0%
B 9 play join rzz 8 2 25.0%
46
Study Plot # BBC Target SAB Total # # of SAB w/i % Group of SAB 3 lags of target
B 9 play join dep 67 10 14.9%
B 9 play join tjk 7 1 14.3%
B 9 play join gsm 25 3 12.0%
B 9 play join xxu 9 1 11.1%
B 9 play join psm 186 20 10.8%
B 10 inq join cch 6 2 33.3%
B 10 inq joi cme 3 1 33.3%
B 11 g. soc cfl (-) xxo 3 1 33.3%
B 11 g. soc cfl (-) sps 11 1 9.1%
B 11 g. soc cfl (-) joi 53 2 3.8%
B 11 g. soc cfl (-) ssm 147 2 1.4%
B 12 g. soc cpc (-) cla 7 2 28.6%
B 12 g. soc cpc (-) cpc 78 13 16.7%
B 12 g. soc cpc (-) xxu 15 2 13.3%
B 12 g. soc cpc (-) psm 202 23 11.4%
B 13 g. soc cro (-) rzz 23 4 17.4%
B 13 g. soc cro (-) btg 15 2 13.3%
B 13 g. soc cro (-) cro 13 1 7.7%
B 14 g. soc cve (-) cme 6 1 16.7%
47
Study Plot # BBC Target SAB Total # # of SAB w/i 3 % Group of SAB lags of target
B 14 g. soc cve (-) sps 11 1 9.1%
B none g. soc cdb (-) gsm 48 4 8.3%
B none g. soc cme (-) psm 146 2 1.4%
B none g. soc cla (-) rzz 23 2 8.7%
B none g. soc cmo (-) fik 7 1 14.3%
B 15 g. soc rzz (+) cro 14 4 28.6%
B 15 g. soc rzz (+) gsm 54 4 7.4%
B 16 g. soc cme (+) cve 3 1 33.3%
B 16 g. soc cme (+) sps 11 1 9.1%
B 17 g. soc cve (+) snk 3 1 33.3%
B 17 g. soc cve (+) vps 9 1 11.1%
B 18 g. soc cro (+) rzz 24 2 8.3%
B 18 g. soc cro (+) cro 14 1 7.1%
B none g. soc cfl (+) srb 27 1 3.7%
B none g. soc cla (+) cpc 58 2 3.4%
B none g. soc cpc (+) cpc 74 13 17.6%
B 19 travel cme (-) psm 34 1 2.9%
B 19 travel cme (-) gsm 35 1 2.9%
B 19 travel cme (-) cpc 36 1 2.8%
48
Study Plot # BBC Target SAB Total # # of SAB w/i % Group of SAB 3 lags of target
B 20 travel cpc (-) xxo 3 1 33.3%
B 20 travel cpc (-) cpc 43 9 20.9%
B 20 travel cpc (-) gsm 43 8 18.6%
B 20 travel cpc (-) joi 42 5 11.9%
B 20 travel cpc (-) psm 71 4 5.6%
B 21 travel cro (-) joi 27 1 3.7%
B 21 travel cro (-) gsm 35 1 2.9%
B 21 travel cro (-) cpc 36 1 2.8%
B 22 travel cme (+) dep 13 1 7.7%
B 22 travel cme (+) joi 26 1 3.8%
B 22 travel cme (+) gsm 36 1 2.8%
B 23 travel cpc (+) cme 3 1 33.3%
B 23 travel cpc (+) cro 3 1 33.3%
B 23 travel cpc (+) cpc 43 9 20.9%
B 23 travel cpc (+) hjk 6 1 16.7%
B 23 travel cpc (+) rst 15 1 6.7%
B none travel cro (+) gsm 36 1 2.8%
B 24 play cla (-) cch 3 1 33.3%
B 24 play cla (-) cpc 32 1 3.1%
49
Study Plot # BBC Target SAB Total # # of SAB w/i % Group of SAB 3 lags of target
B 24 play cla (-) joi 44 1 2.3%
B 25 play cpc (-) rzz 3 1 33.3%
B 25 play cpc (-) vps 7 1 14.3%
B 25 play cpc (-) tjk 7 1 14.3%
B 26 play cro (-) cme 3 1 33.3%
B 26 play cro (-) cro 15 2 13.3%
B 27 play rzz (-) cla 6 1 16.7%
B 27 play rzz (-) joi 44 1 2.3%
B none play cme (-) joi 44 1 2.3%
B 28 play rzz (+) vps 9 2 22.2%
B 28 play rzz (+) cpc 41 2 4.9%
B 28 play rzz (+) dep 64 2 3.1%
B 29 play cme (+) gsm 25 2 8.0%
B 29 play cme (+) cro 15 1 6.7%
B 30 play cpc (+) vps 9 2 22.2%
B 30 play cpc (+) cla 6 1 16.7%
B 30 play cpc (+) xxu 9 1 11.1%
B 31 play cla (+) jop 15 1 6.7%
B 31 play cla (+) psm 151 3 2.0%
50
Study Plot # BBC Target SAB Total # # of SAB w/i % Group of SAB 3 lags of target
B 32 play cmo(+) cla 6 1 16.7%
B 32 play cmo(+) gsm 25 1 4.0%
B none play cro (+) cro 15 2 13.3%
B 33 inq cme (-) joi 14 1 7.1%
B 33 inq cme (-) xxu 15 1 6.7%
B 33 inq cme (-) psm 23 1 4.3%
B 34 inq cpc (-) cme 5 1 20.0%
B 34 inq cpc (-) psm 23 2 8.7%
B 35 inq cme (+) psm 19 2 10.5%
B 35 inq cme (+) cpc 3 1 33.3%
B 36 for cro (+) dep 9 1 11.1%
B 36 for cro (+) psm 42 1 2.4%
J 37 g. soc depart vps 9 1 11.1%
J 37 g. soc depart cme 15 1 6.7%
J 37 g. soc depart cpc 35 2 5.7%
J 37 g. soc depart bbs 33 1 3.0%
J 37 g. soc depart osc 89 2 2.2%
J 38 travel depart cks 4 1 25.0%
J 38 travel depart gsm 4 1 25.0%
51
Study Plot # BBC Target SAB Total # # of SAB w/i % Group of SAB 3 lags of target
J 38 travel depart tjk 16 1 6.3%
J 38 travel depart sps 24 1 4.2%
J 38 travel depart joi 37 1 2.7%
J 38 travel depart psm 271 5 1.8%
J none forage depart osc 12 1 8.3%
J 39 play depart cpc 38 2 5.3%
J 39 play depart rst 32 1 3.1%
J 39 play depart osc 196 3 1.5%
J 40 inq depart xxu 6 2 33.3%
J 40 inq depart xxo 8 1 12.5%
J 40 inq depart gsm 16 1 6.3%
J 40 inq depart cpc 19 1 5.3%
J 40 inq depart joi 20 1 5.0%
J 40 inq depart hjk 28 1 3.6%
J 40 inq depart psm 147 5 3.4%
J 41 g. soc join cme 16 2 12.5%
J 41 g. soc join vps 8 1 12.5%
J 41 g. soc join xxo 13 1 7.7%
J 41 g. soc join dep 29 2 6.9%
52
Study Plot # BBC Target SAB Total # # of SAB w/i % Group of SAB 3 lags of target
J 41 g. soc join cpc 54 3 5.6%
J 42 travel join xxu 17 3 17.6%
J 42 travel join dep 14 2 14.3%
J 42 travel join sps 19 2 10.5%
J 42 travel join psm 301 13 4.3%
J 43 forage join cpc 13 1 7.7%
J 43 forage join psm 82 6 7.3%
J 44 play join xxu 8 2 25.0%
J 44 play join cme 8 1 12.5%
J 44 play join psm 400 30 7.5%
J 44 play join hjk 50 2 4.0%
J 45 inq join vps 12 2 16.7%
J 45 inq join dep 42 4 9.5%
J 45 inq join psm 243 14 5.8%
J 46 g. soc cme (-) xxu 4 1 25.0%
J 46 g. soc cme (-) joi 14 2 14.3%
J 46 g. soc cme (-) vps 9 1 11.1%
J 46 g. soc cme (-) cme 15 1 6.7%
J 47 g. soc cro (-) flx 7 1 14.3%
53
Study Plot # BBC Target SAB Total # # of SAB w/i % Group of SAB 3 lags of target
J 47 g. soc cro (-) cpc 31 1 3.2%
J 47 g. soc cro (-) bbs 7 1 3.0%
J 48 g. soc cpc (-) cme 15 1 6.7%
J 48 g. soc cpc (-) psm 300 13 4.3%
J 48 g. soc cpc (-) cpc 35 1 2.9%
J 49 g. soc cpc (+) dep 29 5 17.2%
J 49 g. soc cpc (+) cpc 54 3 5.6%
J 50 g. soc cme (+) dep 14 1 7.1%
J 50 g. soc cme (+) cme 16 1 6.3%
J 50 g. soc cme (+) cpc 18 1 5.6%
J 50 g. soc cme (+) psm 206 7 3.4%
J 51 g. soc cro (+) gsm 7 1 14.3%
J 51 g. soc cro (+) xxo 9 1 11.1%
J 52 travel cpc (-) xxu 14 1 7.1%
J 52 travel cpc (-) rst 16 1 6.3%
J 53 travel cpc (+) psm 301 12 4.0%
J 53 travel cpc (+) hjk 26 1 3.8%
J 54 play cme (-) hsc 42 1 2.4%
J 54 play cme (-) joi 64 1 1.6%
54
Study Plot # BBC Target SAB Total # # of SAB w/i % Group of SAB 3 lags of target
J 55 play cpc (-) pjk 7 1 14.3%
J 55 play cpc (-) cpc 38 2 5.3%
J 56 play cpc (+) dep 29 3 10.3%
J 56 play cpc (+) cpc 36 2 5.6%
J 56 play cpc (+) bbs 61 3 4.9%
J 56 play cpc (+) joi 56 1 1.8%
J 57 play cme (+) pjk 11 1 9.1%
J 57 play cme (+) psm 361 5 1.4%
J 58 inq cpc (-) cpc 19 1 5.3%
J 59 inq cpc (+) gsm 15 4 26.7%
J 59 inq cpc (+) xxo 23 2 8.7%
J 59 inq cpc (+) sps 33 2 6.1%
J 59 inq cpc (+) cpc 24 1 4.2%
J none inq cmo (+) psm 30 2 6.7%
J 60 forage cpc (+) xxu 3 2 66.7%
J 60 forage cpc (+) tjk 8 2 25.0%
J 60 forage cpc (+) psm 82 7 8.5%
J none forage cpc (-) joi 10 1 10.0%
REFERENCES
Acevedo-Gutiérrez, A. (1999). Aerial behavior is not a social facilitator in bottlenose
dolphins hunting in small groups. Journal of Mammalogy, 80(3), 768-776.
Acredolo, L. & Goodwyn, S. (1988). Symbolic gesturing in normal infants. Child
Development, 59, 450-466.
Adamec, R.E., Stark-Adamec, C., & Livingston, K.E. (1980). The development of
predatory aggression and defense in the domestic cat (Felis catus): Development
of aggression and defense in the first 164 days of life. Behavioral and Neural
Biology, 30(4), 410-434.
Altmann, J. (1974). Observational Study of behavior: Sampling methods. Behaviour,
49(3,4), 227-267.
Altmann, S.A. (1965a). Sociobiology of rhesus monkeys. II: Stochastics of social
communication. Journal of Theoretical Biology, 8, 490-522.
Altmann, S.A. (1965b). The structure of primate communication. In S.A. Altmann (Ed.).
Social communication among primates (pp. 325-362). Chicago: Chicago
University Press.
Altmann, S.A. (1967). The structure of primate communication. In S.A. Altmann (Ed.),
Social communication among primates. Chicago: University of Chicago Press.
Andrew, R.J. (1963). The displays of the primates. In J. Buethner-Janusch (Ed.),
Genetic and evolutionary biology of the primates. New York: Academic Press.
Argyle, M. (1969). Social interaction. New York: Atherton. 116
Argyle, M. (1975). Bodily communication. New York: International Universities Press.
Argyle, M., Alkema, F. & Gilmour, R. (1971). The communication of friendly and
hostile attitudes by verbal and non-verbal signals. European Journal of Social
Psychology, 1, 385-402.
Barnlund, D.C. (1975). Communicative styles in two cultures; Japan and the United
States. In A. Kendon, R.M. Harris, & M.R. Key (Eds.), Organization of behavior
in face-to-face interaction (pp. 427-456). The Hague: Mouton.
Bakeman, R., & Gottman, J.M. (1986). Observing interaction: An introduction to
sequential analysis. Cambridge, U.K.: Cambridge University Press.
Bates, E. (1976). Language and context. New York: Academic Press.
Bateson, G. (1965). Porpoise community research: Final report. National Institutes of
Health Contract No. N 60530-C-1098.
Bekoff, M. (1975). The communication of play intention: Are play signals functional?
Semiotica, 15(3), 231-240.
Bekoff, M., & Byers, J. A. (1998). Animal play: Evolutionary, comparative, and
ecological perspectives. Cambridge, UK: Cambridge University Press.
Bel’kovich, A.V., Agafonov, A.V., Yefremenkova, L.B., & Kharitonov, S.P. (1991).
Herd structure, hunting and play, bottlenose dolphins in the Black Sea. In K.
Pryor & K.S. Norris (Eds.), Dolphin societies, discoveries and puzzles (pp 17-77).
Berkeley: University of California Press.
Birdwhistell, R. (1955). Background to kinesics. ETC., 13, 10-18.
117
Brown, B.A. (1974). Social organization in male groups of white-tailed deer. In V. Geist
& F. Walther (Eds.), The Behaviour of ungulates and its relation to management
(pp. 436-446). Morges, Switzerland: IUCN Publications.
Brown, D.H. & Norris, K.S. (1956). Observations of captive and wild cetaceans. Journal
of Mammalogy, 37. 311-326.
Burgoon, J.K., Buller, D.B. & Woodall, W.G. (1996). Nonverbal communication: The
unspoken dialog. New York: McGraw-Hill.
Caldwell, D.K. & Caldwell, M.C. (1972). The world of the bottlenose dolphin.
Philadelphia: Lippencott.
Caldwell, D.K. & Caldwell, M.C. (1977). Cetaceans. In T. Sebeok (Ed.), How animals
communicate (pp. 794-808). Bloomington, IN: Indiana University Press.
Cheney, D.L. & Seyfarth, R.M. (1990). How monkeys see the world: Inside the mind of
another species. Chicago: University of Chicago Press.
Cherry, C. (1966). On human communication (2nd ed.). Cambridge, MA: MIT Press.
Chevalier-Skolnikoff, S. (1982). A cognitive analysis of facial behavior in Old World
monkeys, apes, and human beings. In C. T. Snowdon, C. H. Brown & M. R.
Petersen (Eds.), Primate communication. Cambridge, MA: Cambridge University
Press.
Cohen, J.A. (1960). Coefficient of agreement for nominal scales. Educational &
Psychological Measurement, 20, 37-46.
Connor, R.C. & Peterson, D.M. (1994). The lives of whales and dolphins. New York:
Henry Holt and Company.
118
Cowan, N (1995). Attention and memory: An integrated framework. New York: Oxford
University Press.
Cullen, J.M. (1972). Some principles of animal communication. In Robert A. Hinde
(Ed.), Non-verbal communication (pp 101- 125). Cambridge University Press:
Cambridge.
Cummings, W.C., Fish, J.F., & Thompson, P.O. (1972). Sound production and other
behavior of southern right whales, Eubalaena glacialis. Transcripts of the San
Diego Society of Natural History, 17, 1-13.
Darwin, C. (1898). The expression of emotions in man and animals. New York: D.
Appleton Company.
DeCourcy, K.R. & Jenssen, T.A. (1994). Structure and use of male territorial headbob
signals by the lizard Anolis carolinensis. Animal Behaviour, 47. 251-262.
Defran, R.H. & Pryor, K. (1980). The behavior and training of cetaceans in captivity. In
L.M. Herman (Ed.), Cetacean behavior: Mechanisms and functions (pp 319-362).
New York: John Wiley & Sons.
DeVito, J.A. & Hecht, M.L. (1990). The nonverbal communication reader. Prospect
Heights, Illinois: Waveland Press.
Dittmann, A.T. (1971). Book review of kinesics and context by Ray L. Birdwhistell.
Psychiatry, 34, 334-342.
Dudzinski, K.M. (1996). Communication and behavior in the Atlantic spotted dolphin
(Stenella frontalis): Relationships between vocal and behavioral activities. PhD
dissertation, Texas A&M University. 215 pp.
119
Dudzinski, K.M. (1998). Contact behavior and signal exchange in Atlantic spotted
dolphins (Stenella frontalis). Aquatic Mammals. 24(3), 129-142.
Dudzinski, K.M. (1999). [Longitudinal study of Stenella frontalis and Tursiops aduncus
behavior]. Unpublished raw data.
Dudzinski, K.M., Clark, C.W., & Würsig, B. (1995). A mobile video/acoustic system for
simultaneous underwater recording of dolphin interactions. Aquatic Mammals,
21(3), 187-193.
Dudzinski, K.M., Frohoff, T.G., & Crane, N.L. (1995). Behavior of a lone female
bottlenose dolphin (Tursiops truncatus) with humans off the coast of Belize.
Aquatic Mammals, 21(2), 149-153.
Dudzinski, K.M., Sakai, M., Masaki, K., Kogi, K., Hishii, T., & Kurimoto, M. (2003).
Behavioral observations of adult and sub-adult dolphins towards two dead
bottlenose dolphins (one female and one male). Aquatic Mammals, 29(1), 108-
116.
Dudzinski, K.M., Thomas, J.A., & Douaze, E. (2002). Communication In W.F. Perrin,
B. Würsig, & J.G.M. Thewissen, (Eds.), Encyclopedia of marine mammals (pp.
248-268). New York: Academic Press.
Dusenbery, D.B. (1992). Sensory ecology: How organisms acquire and respond to
information. New York: WH Freeman.
Ekman, P. & Friesen, W.V. (1969). The repertoire of nonverbal behavior: Categories,
origins, usage, and coding. Semiotica, 1, 49-98.
Evans, W.E. (1973). Echolocation by marine delphinids and one species of fresh water
dolphin. Journal of the Acoustical Society of America, 54(1). 191-199.
120
Evans, W.E. & Bastian, J. (1969). Marine mammal communication: Social and
ecological factors. In Anderson, H.T. (Ed.) The Biology of Marine Mammals. (pp
425-475). New York: Academic Press.
Fagen, R. (1981). Animal play behavior. New York: Oxford University Press.
Fehr, B.J. and Exline, Ralph V. (1987) Social visual interaction: A conceptual and
literature review. In Aron W. Siegman & Stanley Feldstein, (Eds.), Nonverbal
behavior and communication (pp 225-326). New Jersey: Lawrence Erlbaum &
Associates.
Fertl, D. & Würsig, B. (1995). Coordinated feeding by Atlantic spotted dolphins
(Stenella frontalis) in the Gulf of Mexico. Aquatic Mammals, 21, 3-5.
Fox, M.W. (1969). The anatomy of aggression and its ritualization in Canidae: A
developmental and comparative study. Behaviour, 35, 242-258.
Fox, M.W. (1971). Behaviour of wolves, dogs, and related canids. New York: Harper &
Row.
Fox, M.W. & Clark, A.L. (1971). The development and temporal sequencing of agonistic
behavior in the coyote (Canis latrans). Z. Tierpsychol., 28, 262-278.
Fox, M.W. & Cohen, J.A. (1976). Vocalizations in wild canids and possible effects of
domestication. Behavioural Processes, 1(1), 77-92.
Gao, G. & Zhou, K. (1995). Fiber analysis of the vestibular nerve of small cetaceans. In
R.A. Kastelein, J.A. Thomas, & P.E. Natchtigall (Eds.), Sensory systems of
aquatic mammals (pp 447-453). Woerden: DeSpil.
Goodall, J. (1986). The chimpanzees of gombe: Patterns of behavior. Cambridge:
Belknap Press of Harvard University Press.
121
Graybiel, A. (1964). Vestibular sickness and some of its implications for space flight. In
W.S. Fields & R.R. Alford (Eds.), Neurological aspects of auditory and
vestibular disorders (pp. 248-270). Springfield: Charles C. Thomas.
Hare, B., & Tomasello, M. (1998). Domestic dogs (Canis familiaris) use human and
conspecific social cues to locate hidden food. Journal of Comparative
Psychology, 113, 1-5.
Hauser, M.D. (1996). The evolution of communication. Cambridge MA: MIT Press.
Hauser, M.D. & Konishi, M. (1999). The design of animal communication. Cambridge,
MA: MIT Press.
Herman, L.M. (1980). Cetacean behavior: Mechanisms and functions. New York:
Wiley, Inc.
Herman, L.M. (2002). Exploring the cognitive world of the bottlenosed dolphin. In M.
Bekoff & C. Allen (Eds.), The cognitive animal: Empirical and theoretical
perspectives on animal cognition. Cambridge, MA: MIT Press.
Herman, L.M., Abichandani, S.L., Elhajj, A.N., Herman, E.Y.K., Sanchez, J.L., & Pack,
A.A. (1999). Dolphins (Tursiops truncatus) comprehend the referential character
of the human pointing gesture. Journal of Comparative Psychology, 113, 347-
364.
Herman, L. M., & Forestell, P. H. (1977, December). The Hawaiian humpback whale:
behaviors. Paper presented at the 2nd Conference of the Biology of Marine
Mammals, San Diego, CA.
122
Herman, L.M. & Tavolga, W.N. (1980). The communication systems of cetaceans. In
L.M. Herman (Ed), Cetacean behavior: Mechanisms and functions (pp 149-209).
New York: Wiley, Inc.
Herzing, D.L. (1997). The life-history of free-ranging Atlantic spotted dolphins (Stenella
frontalis): Age classes, color phases, and female reproduction. Marine Mammal
Science, 13, 576-595.
Herzing, D.L. (2000). Acoustics and social behavior. In W.W.L. Au, A.N. Popper, &
R.R. Fay (Eds.), Hearing by whales and dolphins (pp 225-272). New York:
Springer-Verlag.
Herzing, D.L., Moewe, K., & Brunnick, B.J. (2003). Interspecies interactions between
Atlantic spotted dolphins, Stenella frontalis, and bottlenose dolphins, Tursiops
truncatus, on Great Bahama Bank, Bahamas. Aquatic Mammals, 29(3), 335-341.
Hinde, Robert A. (1972). Non-verbal communication. Cambridge: Cambridge
University Press.
Iverson, J.M. & Goldin-Meadow, S. (1998). Why people gesture when they speak.
Nature, 396, 228.
Jacobsen, J.K. (1986). The behavior of Orcinus orca in the Johnstone Strait, British
Columbia. In B.C. Krikevold & J.S. Lockard (Eds.), Behavioral biology of killer
whales (pp. 135-186). New York: Allan R. Liss.
Johnson, C.M. & Moewe, K. (1999). Pectoral fin preference during contact in
Commerson’s dolphins (Cephalorhynchus commersonii). Aquatic Mammals,
25(2), 73-77.
123
Jolly, S. (2000). Understanding body language: Birdwhistell’s theory of kinesics.
Corporate Communications, 5(3), 133-139.
Jourard, S.M. (1966). An exploratory study of body-accessibility. British Journal of
Social and Clinical Psychology, 5, 221-231.
Ketten, D. R. (1992). The marine mammal ear: Specializations for aquatic audition and
echolocation. In D. Webster, R. Fay, & A.N. Popper (Eds.), The Evolutionary
biology of hearing (pp 717-754). New York: Springer-Verlag.
Ketten, Darlene R. (2000). Cetacean ears. In Whitlow W.L. Au, Arthur N. Popper &
Richard R. Fay (Eds.), Hearing by whales and dolphins (pp 43-108). New York:
Springer-Verlag.
Kimura, D. (1993). Neuromotor mechanisms in human communication. Oxford, U.K.:
Oxford University Press.
Kogi K., Hishii T., Imamura A., Iwatani, T., & Dudzinski K.M. (in press). Preliminary
demographic parameters and reproductive rates of Indo-Pacific
bottlenose dolphins (Tursiops aduncus) around Mikura Island, Japan. Marine
Mammal Science.
Kogi K., Hishii T., Imamura A., Eiji O., & Dudzinski K.M. (2003). Population
demographics, including composition, reproductive rates, population size
estimates, and preliminary behavioral descriptions, for an eight-year period in a
longitudinal study of Indo-Pacific bottlenose dolphins (Tursiops aduncus) around
Mikura Island, Japan. Manuscript submitted for publication.
Krauss, R.M. (1998). Why do we gesture when we speak? Current Directions in
Psychological Science, 7(2), 54-60.
124
Krebs, J.R. & Davies, N.B. (1993). Introduction to behavioural ecology. Oxford, U.K.:
Blackwell Scientific.
Langbauer Jr., W.R. (2000). Elephant communication. Zoo Biology, 19, 425-445.
LaFrance, M. & Mayo, C. (1979). A review of nonverbal behaviors of women and men.
Western Journal of Speech Communication, 43, 96-107.
Lehner, P.N. (1996). Handbook of ethological methods. Cambridge, U.K.: Cambridge
University Press.
Little, K.B. (1968). Cultural variations in social schemata. Journal of Personality and
Social Psychology, 10, 1-7.
Lovern, M.B. & Jenssen, T.A. (2003). Form emergence and fixation of head bobbing
displays in the green Anole lizard (Anolis carolinensis): A reptilian model of
signal ontogeny. Journal of Comparative Psychology, 117(2), 133-141.
MacLeod, C.D. (1998). Intraspecific scarring in odontocete cetaceans: an indicator of
male ‘quality’ in aggressive social interactions? Journal of Zoology London, 244,
71-77.
Mann, J. & Smuts, B. (1999). Behavioral development in wild bottlenose dolphin
newborns (Tursiops spp.). Behaviour, 136, 529-566.
Marler, P. (1977). The evolution of communication. In T.A. Sebeok, (Ed.), How animals
communicate (pp. 45-70). Bloomington, IN: Indiana University Press.
Marquette, W. M. (1978). Bowhead whale. In D. Haley (Ed.), Marine mammals (pp. 70-
81). Seattle: Pacific Search.
McArdle, W.D., Katch, F.I., and Katch, V.I. (1991). Exercise physiology: Energy,
nutrition, and human performance (3rd ed.). Philadelphia: Lea and Febiger.
125
McBride, A.F. (1956). Evidence for echolocation by cetaceans. Deep-sea Research, 3,
153.
McBride, A. F., & Hebb, D. O. (1948). Behavior of the captive bottle-nose dolphin,
Tursiops truncatus. Journal of Comparative and Physiological Psychology, 41,
111-123.
McBride, A.F. & Kritzler, H. (1951). Observations on pregnancy, parturition, and post-
natal behavior in the bottlenose dolphin. Journal of Mammalogy, 32, 251-266.
McNeill, D. (1992). Hand and mind. Chicago: University of Chicago Press.
Miklósi, A, Polgárdi, R. Topál, J. & Csányi, V. (2000). Intentional behaviour in dog-
human communication: an experimental analysis of “showing” behaviour in the
dog. Animal Cognition, 3, 159-166.
Moehlman, P.D. (1987). Social organization in jackals. American Scientist, 75, 366-375.
Moynihan, M. (1967). Comparative aspects of communication in New World primates.
In D. Morris (Ed.), Primate ethology. London: Weidenfeld and Nicolson.
Noldus Information Technology (2003). The Observer (Version 5.0) [Computer
software]. Wageningen, The Netherlands: NIT b.v.
Norris, K.S. (1967). Aggressive behavior in cetacean. In C.D. Clemente & D.B. Lindsley
(Eds.). Aggression and defense (pp. 225-241). Berkeley, CA: University of
California Press.
Norris, K.S. (1969). The echolocation of marine mammals. In Harald T. Anderson
(Ed.), The biology of marine mammals (pp. 391-424). New York: Academic
Press.
126
Norris, K. S., Goodman, R. M., Villa-Ramirez, B., & Hobbs, L. (1977). Behavior of
California gray whale, Eschrichitus robustus, in southern Baja California,
Mexico. Fisheries Bulletin, 75, 159-171.
Norris, K.S., Würsig, B., Wells, R.S., & Würsig, M. (1994). The Hawaiian spinner
dolphin. Berkeley: University of California Press.
Okamoto, S., Tomonaga, M., Ishii, K., Kawai, N., Tanaka, M., Matsuzawa, T. (2002).
An infant chimpanzee (Pan troglodytes) follows human gaze. Animal Cognition,
5, 107-114.
Östman, J. (1990). Changes in aggressive and sexual behavior between two male
bottlenose dolphins (Tursiops truncatus) in a captive colony. In K.W. Pryor &
K.S. Norris (Eds.), Dolphin societies: Discoveries and puzzles (pp. 305-318).
Berkeley, CA: University of California Press.
Otte, D. (1974). Effects and functions in the evolution of signaling systems. Annual
Review of Ecological Systems, 5, 385-417.
Overstrom, N. (1983). Association between burst-pulse sounds and aggressive behavior
in captive Atlantic bottlenosed dolphins (Tursiops truncatus). Zoo Biology, 2, 93-
103.
Ozoga, J.J. (1972). Aggressive behavior of white-tailed deer at winter cuttings. Journal
of Wildlife Management, 36(3), 861-868.
Palmer, E. & Weddell, G. (1964). The relationship between structure, innervation, and
function of the skin of the bottlenose dolphin, Tursiops truncatus. Proc. Zool.
Soc. Lond., 143, 553-568.
127
Pavlov, I.P. (1927). Conditioned reflexes, (G.V. Anrep, Trans.). London: Oxford
University Press.
Payne, R. S. (1976). At home with right whales. National Geographic, 149, 322-339.
Perrin, W.F. (1970). Color pattern of the eastern Pacific spotted porpoise Stenella
graffmani Lonnberg (Cetacea, Delphinidae). Zoology, 54, 135-142.
Peters, R. (1980). Mammalian communication: A behavioral analysis of meaning.
Monterey: Brooks/Cole Publishing Company.
Pryor, K.W. (1986). Non-acoustic communicative behavior of the great whales: Origins,
comparisons, and implications for management. Rep. Int. Whal. Comm. 8, 89-96.
Pryor, K.W. (1990). Non-acoustic communication in small cetaceans: Glance, touch,
position, gesture, and bubbles. In J.A. Thomas & R. Kastelein (Eds.), Sensory
abilities of cetaceans (pp 537-544). New York: Plenum Press.
Pryor, K.W., & Shallenberger, I.K. (1990). Social structure in spotted dolphins (Stenella
attenuata) in the tuna purse seine fishery in the Eastern Tropical Pacific. In K.W.
Pryor & K.S. Norris (Eds.), Dolphin societies: Discoveries and puzzles (pp.161-
198). Berkeley, CA: University of California Press.
Randall, J.A. & Lewis, E.R. (1997). Seismic communication between the burrows of
kangaroo rats, Dipodomys spectabilis. Journal of Comparative Physiology, 181,
525-531.
Rauscher, F. H., Krauss, R. M., & Chen, Y. (1996). Gesture, speech, and lexical access:
The role of lexical movements in speech production. Psychological Science, 7(4),
226-231.
128
Reynolds, J.E. & Rommel, R. (1999). The biology of marine mammals. Washington,
D.C.: Smithsonian Institution Press.
Richardson, W.J., Greene Jr., C.R., Malme, C.I. & Thomson, D.H. (1995). Marine
mammals and noise. San Diego: Academic Press.
Rogers, L.J., & Kaplan, G. (2000). Songs, roars, and rituals: Communication in birds,
mammals, and other animals. Cambridge, MA: Harvard University Press.
Rossbach, K.A. & Herzing, D.L. (1999). Inshore and offshore bottlenose dolphin
(Tursiops truncatus) communities distinguished by association patterns near
Grand Bahama Island, Bahamas. Canadian Journal of Zoology, 77(4), 581-592.
Saayman, G. S., & Tayler, C. K. (1972). The social organization and behavior of dolphins
(Tursiops aduncus) and baboons (Papio ursinus): Some comparisons and
assessments. Ann. Cape Prov. Mus. (Nat. Hist.), 9, 11-49.
Saayman, G. S., & Tayler, C. K. (1973). Some behaviour patterns of the southern right
whale Eubalaena australis. Zeitschrift fur Saugetierkunde, 38(3), 172-183.
Saayman, G. S., & Tayler, C. K., & Bower, D. (1973). Diurnal activity cycles in captive
and free-ranging Indian Ocean dolphins. Behaviour, 44, 212-233.
Sakai, M., Hishii, T., Takeda, S, & Kohshima, S. (2003). Quantitative analyses of flipper
rubbing behavior in wild Indo-Pacific bottlenose dolphins (Tursiops aduncus).
Oral presentation presented at the 15th Biennial conference on the Biology of
Marine Mammals. Greensboro, North Carolina.
Samuels, A. & Gifford, T. (1997). A quantitative assessment of dominance relations
among bottlenose dolphins. Marine Mammal Science, 13, 70-99.
Schenckel, R. (1947). Expression studies of wolves. Behaviour, 1, 81-129.
129
Scherer, K.R., & Ekman, P. (1982). Handbook of methods in nonverbal behavior
research. Cambridge: Cambridge University Press.
Sebeok, T. (1977). How Animals Communicate. Bloomington, Indiana: Indiana
University Press.
Seyfarth, R.M. (1980). The distribution of grooming and related behaviour among adult
female vervet monkeys. Animal Behavior, 28, 798-813.
Shane, S.H., Wells, R.S., & Würsig, B. (1986). Ecology, behavior and social organization
of the bottlenose dolphin: A review. Marine Mammal Science, 2(1), 34-63.
Shier, D.M. & Yoerg, S.I. (1999). What footdrumming signals in kangaroo rats
(Diposomys heermanni). Journal of Comparative Psychology, 113, 66-73.
Shuter, R. (1977). A field study of nonverbal communication in Germany, Italy and the
United States. Communication Monographs, 44, 298-305.
Siegman, A.W. & Feldstein, S. (1987). Nonverbal behavior and communication.
Hillsdale, NJ: Lawrence Erlbaum Associates.
Slooten, E. (1994). Behavior of Hector’s dolphin: Classifying behavior by sequence
analysis. Journal of Mammalogy. 75(4), 956-964.
Smith, W.J. (1977). The Behavior of communicating: An ethological approach.
Cambridge, MA: Harvard University Press.
Smith, W.J. (1985). Consistency and change in communication. In G. Zivin (Ed.), The
development of expressive behavior: Biology-environment interactions (pp.51-76)
New York: Academic Press.
130
Smith, W.J. (1990). Communication and expectations: A social process and the cognitive
operations it depends upon and influences. In M. Bekoff & D. Jamieson (Eds.),
Interpretation and explanation in the study of animal behavior: Volume I
interpretation, intentionality & communication (pp. 234-253). Boulder, CO:
Westview Press.
Smolker, R.A., Mann, J., & Smuts, B.B. (1993). Use of signature whistles during
separations and reunions by wild bottlenose dolphin mothers and infants.
Behavioral Ecology & Sociobiology, 33, 393-402.
Smolker, R.A, Richards, A.F., Connor, R.C., & Pepper, J.W. (1992). Sex differences in
patterns of association among Indian Ocean bottlenose dolphins. Behaviour, 123
(1-2), 38-69.
Snowdon, C.T. & Boe, C.Y. (2003). Social communication about unpalatable foods in
tamarins (Saguinus oedipus). Journal of Comparative Psychology, 117(2), 142-
148.
Steudel, K. (2000). The physiology and energetics of movement: Effects on individuals
and groups. In Sue Boinski & Paul A. Garber (Eds.), On the move: How and why
animals travel in groups (pp 9-23). Chicago: University of Chicago Press.
Struhsaker, T. (1967). Auditory communication among vervet monkeys (Cercopithecus
aethiops). In S.A. Altmann (Ed), Social communication among primates (pp.
281-324). Chicago: University of Chicago Press.
Tavolga, M.C. (1966). Behavior of the bottenose dolphin (Tursiops truncatus): Social
interactions in a captive colony. In K.S. Norris (Ed.), Whales, dolphins, and
porpoises. Berkeley, CA: University of California Press.
131
Tinbergen, N. (1952). Derived activities: Their causation, biological significance, origin
and emancipation during evolution. Quarterly Review of Biology, 27, 1-32
Tinbergen, N. (1960). The herring gull’s world. New York: Basic Books.
Tinbergen, N. (1965). The study of instinct. Oxford: Clarendon Press.
Tomasello, M., & Call, J. (1997). Primate cognition. New York: Oxford University
Press.
Tomasello, M., Hare, B., & Fogleman, T. (2001). The ontogeny of gaze following in
chimpanzees, Pan troglodytes, and rhesus macaques, Macaca mulatta. Animal
Behaviour, 61, 335-343.
Tschudin, A., Call, J., Dunbar, R.I.M., Harris, G., & van der Elst, C. (2001).
Comprehension of signs by dolphins (Tursiops truncatus). Journal of
Comparative Psychology, 115(1), 100-105.
Tucker, V.A. (1975). The energetic cost of moving about. American Scientist, 63, 413-
419.
Tyack, P.L. (2000). Functional aspects of cetacean communication. In J. Mann, R.C.
Connor, P.L. Tyack & H. Whitehead (Eds.), Cetacean societies: Field studies of
dolphins and whales (pp 270-307). Chicago: University of Chicago Press.
Tyack, P.L. & Clark, C.W. (2000). Communication and acoustic behavior of dolphins
and whales. In W.W.L. Au, A.N. Popper, & R.R. Fay (Eds.), Hearing by whales
and dolphins (pp 156-224). New York: Springer-Verlag.
Vandenberg, B. (1978). Play and development from an ethological perspective. American
Psychologist, 33, 724-738.
132 van Hooff, J.A.R.A.M. (1967). The facial displays of the catarrhine monkeys and apes.
In D. Morris (Ed.), Primate ethology (pp.7-68). London; Weidenfeld and
Nicolson.
Vauclair, J. (1996). Animal cognition. Cambridge, MA: Harvard University Press.
de Waal, F.B.M. (1982). Chimpanzee politics. London: Jonathan Cape.
de Waal, F.B.M. & Luttrell, L.M. (1989). Toward a comparative socioecology of the
genus Macaca: Different dominance styles in rhesus and stumptail monkeys.
American Journal of Primatology, 19, 83-109.
Williams, T.M., Friedl, W.A., Fong, M.L., Yamada, R.M., Dedivy, P. and Haun, J.
(1992). Travel at low energetic cost by swimming and wave-riding bottlenose
dolphins. Nature, 355, 821- 823.
Würsig, B., Kieckhefer, T.R. & Jefferson, T.A. (1990). Visual displays for
communication in cetaceans. In J.A. Thomas & R. Kastelein (Eds.), Sensory
abilities of cetaceans (pp 545-559). New York: Plenum Press.
Würsig, B. & Würsig, M. (1979). Behavior and ecology of bottlenose porpoises,
Tursiops truncatus, in the South Atlantic. Fishery Bulletin, 77, 399-442.
Würsig, B. & Würsig, M. (1980). Behavior and ecology of the dusky dolphin,
Lagenorhynchus obscurus, in the South Atlantic. Fishery Bulletin, 77, 871-890.
Xitco, M.J., Gory, J. D., & Kuczaj, S.A. II. (2001). Spontaneous pointing by bottlenose
dolphins (Tursiops truncatus). Animal Cognition, 4, 115-123.
Xitco, M.J., & Roitblat, H.L. (1996). Object recognition through eavesdropping: passive
echolocation in bottlenose dolphins. Animal Learning and Behavior, 24, 355-365.
133
Yamagiwa, J. (1992). Functional analysis of social staring behaviour in an all-male group
of mountain gorillas. Primates, 33, 523-524.