-
extracted text
-
USING VOCALIZATIONS TO DETERMINE
THE CAPACITY FOR FUTURE COGNITION
IN THE COMMON RAVEN (CORVUS CORAX)
by
Caitlyn E. Roehmholdt
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June 2019
©2019 by Caitlyn E. Roehmholdt. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Caitlyn E. Roehmholdt
has been approved for
The Evergreen State College
by
________________________
John Withey, PhD
Member of the Faculty
________________________
Date
ABSTRACT
Using vocalizations to determine the capacity for
future cognition in the common raven (Corvus corax)
Caitlyn E. Roehmholdt
Common ravens (Corvus corax) are known to be highly intelligent birds and have been
the subjects of many behavioral studies. However, not many natural experiments that
examine the ability of ravens to understand time and predict events have been conducted.
The population of wild ravens at Wolf Haven, a nonprofit sanctuary for abandoned and
displaced wolves, provides a unique opportunity to study both the behaviors of ravens
that have been exposed to a regular schedule of events (for example, scheduled feeding of
the wolves, regular sanctuary visitors, and non-food enrichment activities for the wolves)
and their ability to predict when food will be available. Here, recordings were taken of
raven vocalizations over 16 weeks on scheduled feeding days, fasting days, and two
instances where the feeding and fasting days were reversed. The recordings were
analyzed for presence of “haa” calls, a food associated call in ravens. The number of haa
calls observed on feeding days versus on fasting days was compared using a paired t-test
under the assumption that the anticipation of food would lead to more food related
vocalizations. The number of haa calls on days when feeding was expected to occur but
was not observed was also analyzed. Though there was not a statistically significant
difference between the number of calls observed on feeding days versus fasting days, this
study poses many questions and suggests possibilities for future research into the use of
audio recording to investigate the behavior and cognitive abilities of ravens.
Table of Contents
List of Figures ..................................................................................................................... v
List of Tables ..................................................................................................................... vi
Acknowledgements ............................................................................................................. 7
Chapter 1: Introduction ....................................................................................................... 1
Chapter 2: Literature Review .............................................................................................. 8
Avian Intelligence ........................................................................................................... 9
Corvids .......................................................................................................................... 11
Ravens and Myth ........................................................................................................... 12
Ravens and Wolves ....................................................................................................... 14
Raven Cognition ............................................................................................................ 16
Raven Vocalization ....................................................................................................... 18
New Frontiers ................................................................................................................ 19
Chapter 3: Methods ........................................................................................................... 21
Site Selection ................................................................................................................. 21
Observation Schedule and Recording ........................................................................... 22
Analysis ......................................................................................................................... 24
Chapter 4: Results ............................................................................................................. 27
Chapter 5: Conclusion and Discussion ............................................................................. 30
Bibliography ..................................................................................................................... 35
Appendices ........................................................................................................................ 43
List of Figures
Figure 1 Diagram of Wolf Haven study site ..................................................................... 23
Figure 2. Haa call spectrogram ........................................................................................ 24
Figure 3. Two examples of non-haa raven calls .............................................................. 24
v
List of Tables
Table 1. Average number of haa calls on feeding and fasting days ................................. 27
Table 2. Haa calls on observed feeding days .................................................................... 28
Table 3 Haa calls per hour on scheduled feeding days ..................................................... 28
Table 4 Summary of data used for ANOVA .................................................................... 29
vi
Acknowledgements
I would like to thank my advisor John Withey for his guidance, reassurance, sharing of
knowledge, and curiosity in and support of my work; my fiancé Aaron for his unending
support and belief in me and my abilities; my parents and family members for raising me
to believe that anything is possible; and my friends and cohort members for their much
needed emotional support. I would also like to thank the staff of Wolf Haven
International for allowing me to use their fantastic facility as my research site.
vii
CHAPTER 1: INTRODUCTION
The Corvidae family of birds is a highly intelligent group that includes crows,
ravens, rooks, jackdaws, jays, magpies, and nutcrackers. Members of this family group,
known collectively as corvids, have exhibited the ability to use tools, follow human gazes
and remember interactions with humans, communicate the presence of food and its status
(protected, available, etc.) to conspecifics, and live in highly social groups. Their
problem-solving ability is similar to that of great apes. Birds of any species have not
always been recognized for their intelligence, especially those abilities that indicate
higher level cognitive processes. Perhaps only the parrot, with its great ability to mimic,
has regularly been thought of as intelligent by the average layperson. However, as
advances in behavioral studies and neurological imaging allow for more in-depth study of
animal brains and abilities, the days of birds being regarded as unintelligent are slowly
becoming a relic of the past.
All birds have a brain that is very different from the mammalian brain. For many
years, it was assumed that due to this difference in structure birds were incapable of
producing complex, flexible, intelligent behavior (Emery, 2006). For instance, whereas
mammals have a neocortex, birds have a structure called a pallium; however, the pallium
has been found to have analogous functions to the neocortex, proving that the brain’s
structure and anatomy does not necessarily determine the intelligence and abilities of a
species (Nomura & Izawa, 2017).
1
In addition to assumptions about structural differences, it was long thought that
the intelligence of an animal corresponded to the size and structure of an animal’s brain,
but recent studies have begun to change that notion: it is the ratio of the size of the
animal’s brain to its body (called encephalization) that seems to correlate with
intelligence and cognitive ability, not just physical brain size or the presence of a humanlike neocortex (Emery, 2006). Emery and Clayton (2004) theorize that the brains of
corvids evolved such similar processes and capabilities to the brains of apes despite the
structural differences due to the two groups facing many of the same socioecological
challenges, using the understanding of complex social dynamics and the ability to cache
and retrieve food as examples. Indeed, although the brains of corvids and parrots are
large in comparison to their body size and have a similar encephalization to that of
monkeys and apes, they are still small, with the average raven’s brain weighing in at
around 15 grams (Iwaniuk & Nelson, 2002). Despite this, the corvid brain is packed with
neurons; ravens not only have a large amount of neurons, their neurons are very tightly
packed, which allows their brains to contain as many or more neurons than that of the
much larger brains of monkeys (Fitch et al., 2016).
These neuron-dense brains also enable corvids to use and manufacture tools,
something once thought to be an ability exclusive to primates but now understood to be
present in many animal species (Anderson, 2002). The most prolific and best studied
corvid tool maker is the New Caledonian crow (Corvus moneduloides), which is the only
non-human species known to manufacture tools in the wild other than chimpanzees
(Emery & Clayton, 2009). Although only the two aforementioned non-human species
2
have been found to manufacture tools, many more species have been found to use tools,
and even species that do not use tools in the wild can be prolific tool users under novel
conditions in a lab setting (Emery & Clayton, 2009).
Although ravens do not manufacture tools like their New Caledonian relatives,
they have other abilities that place them in league with primates. Captive ravens exhibit
bartering abilities exceeding those of both the great apes, as well as an average 4-year-old
human (Kabadayi and Osvath 2017). Recent experiments have shown that ravens exhibit
flexible behavior and even imagination: when presented with a novel situation requiring
problem solving to find the only solution—in this case, a piece of meat hanging from a
long string attached to a perch—the hand-raised ravens participating in the study were
able to solve the puzzle by repeatedly pulling on the string, holding it with their foot, and
pulling again, some of them solving the problem on their first attempt (Emery & Clayton,
2004).
High sociality and communication ability are also raven traits. Through
experimentation, Schwab and colleagues (2008) found that raven sibling pairs tended to
sit together both more frequently and for a longer duration than non-sibling pairs,
pointing to the highly social and family-oriented nature of the species. Furthermore, the
study showed that social learning in ravens is influenced by the interrelationships of the
birds, not just by the birds being in a similar age range. Fraser and Bugnyar (2010)
confirmed that raven siblings have more meaningful relationships than non-siblings,
furthering the comparison between the sociality of ravens and that of great apes.
3
Ravens are also known for being scavengers in the habit of caching their food, a
behavior that requires them to remember where they place their caches for extended
periods of time for the purpose of later retrieval. Ravens are able to distinguish whether
other ravens have observed their caching, and retrieve food from caches observed by
other ravens sooner than from unobserved caches (Bugnyar and Heinrich 2005). A
similar earlier study determined that raven that cache and ravens that observe employ
“countertactics” to gain the upper hand in securing food (Bugnyar & Kotrschal, 2002). It
has also been shown that ravens observe other ravens both inside and outside of their
social group more than other species (Scheid et al., 2007), which is possibly linked to
their tendency for social learning. This type of behavior implies that ravens can take their
observations, commit them to memory, and use them to plan for future events.
Ravens have also been shown to engage in, and possibly understand, social
behavior and cues from humans. Bugnyar, Stöwe, and Heinrich (2004) conducted an
experiment, the first of its kind on birds, to determine whether ravens were able to follow
a human experimenter’s gaze around a barrier. They found that ravens of different ages
succeeded at this task during around 80% of the trials. However, the experimenters do
caution that while the results of the experiment do point to the possibility that ravens are
able to understand the visual activities of others, there is still a question as to whether or
not ravens understand or attribute any mental significance to the gaze of others (Bugnyar
et al., 2004).
4
The robust social lives of ravens are also tied to their memories, both for other
ravens and for humans. It has been found that ravens can remember which human
experimenter in a food exchange experiment trial gives “fair” exchanges and which is
“unfair” after single events and act accordingly in future trials, preferring to trade with
the “fair” experimenter for up to a month after the observation of the experiment (Muller
et al., 2017). Their memories are not just related to the humans that they may interact
with: Ravens are able to remember the individual calls of past group mates for up to 3
years (Taylor, 2014).
Vocalizations aren’t just used for group and individual identification. As a highly
social species, ravens use vocalizations to communicate with each other. One of the most
commonly communicated topics amongst ravens is food: whether it is available or
unavailable, its location, and even the quality of the food (Bugnyar et al., 2001). Food
calls also carry information about the caller: it has been found that ravens can distinguish
between individuals based upon the characteristics of their food calls (Szipl et al., 2015).
Ravens are also known to call conspecifics to feeding sites, with resident ravens tending
to produce food calls more frequently than vagrant birds, supporting the idea that raven
social groups are strongly based on population dynamics that change over time as
individual animals enter and leave the group (Szipl & Bugnyar, 2014).
The specific call type that ravens use to “talk about” food is referred to in the
literature as either the “haa” call or the food yell or scream (Bugnyar et al., 2001). These
calls, which are speculated to have developed from the begging calls of young ravens,
5
have been found to be practiced and modified throughout a raven’s early years and may
even be used by raven parents to locate individual young (Boeckle et al., 2012). Haa
calls are used all throughout a raven’s life; Although the call varies slightly based on a
caller’s age, the core characteristics are the same (Boeckle et al., 2012; Boeckle et al.,
2018) and the call can be identified relatively easily using audio spectrogram analysis.
All of these traits combine to paint a picture of an extremely intelligent and social
animal not too unlike humans. However, the study of foresight, future cognition, and
planning in animals in general is still in the early stages due to the difficulties inherent in
studying the minds of species we cannot directly communicate with to ask about their
experiences. Many opportunities exist for studies in this realm, especially as our
understanding of animal neuroscience continues to advance. For example, Zentall (2005)
and associates have shown through experimentation the different ways that pigeons and
mice may experience both episodic and semantic memory, and how they can be trained to
report their experience by pressing different colored keys. Studies with corvids have
helped to illuminate their intelligence: Corvids have demonstrated a superior ability to
understand the relationships between items, being able to make pairwise comparisons and
correctly order a given set of objects (Emery & Clayton, 2004). Crows are able to
recognize specific, individual human faces of those humans that have been deemed
dangerous for at least 3 years after the initial encounter (Marzulff et al., 2010). Ravens
recruit others from roosts outside of their own to feeding bonanza, many times from
miles away (Heinrich, 1994).
6
These studies and many more beg the question: If corvids can learn things such as
facial recognition and tool use, couldn’t they learn the schedule of a guaranteed food
source appearing?
The goal of this study is to answer the aforementioned question by attempting to
determine whether wild ravens can predict and anticipate scheduled feeding events at a
wolf sanctuary. At the time of this writing, to the author’s knowledge no natural
experiments that examine the ability of ravens to understand time and predict events have
been conducted, although experimental studies examining the reaction of wild ravens to
the feeding of game park animals at various game parks around the world have been
conducted (Bugnyar & Kotrschal, 2001; Bugnyar & Kotrschal, 2002; Bugnyar et al.,
2001; Loretto et al., 2016).
7
CHAPTER 2: LITERATURE REVIEW
Long before animals were considered to have thoughts, feelings, and emotions,
humans were interested in the thoughts of their own kind. Since the time of the ancient
Greek philosophers, human thought and the human condition in general has been
pondered at great length, with human beings held in high esteem and seen as being of
higher status and, in fact, as higher beings not related to animals at all (Adamson &
Edwards, 2018). This idea that animals are unintelligent, “soulless” automatons is an idea
that was solidified by Rene Descartes’s assertion that animals are no more than mere
machines (Gould, 2004). However, during the past century, attitudes towards animals
have changed and more and more studies of animal cognition have taken place. Although
scientists and many non-scientists alike now recognize that humans are animals and are
content to group mankind in with the rest of the primates, this concept is far from being
internalized in the majority of the population (Boddice, 2011).
The human brain is similar to the brain of other mammals, and is similar enough
that other animal brains are used as models for uncovering the mysteries of the brain; for
example, dye can be injected into the brain of a mouse to trace the neural networks and
reactions of the brain to stimuli (Allen, 2006). The concept that different areas of the
brain are linked to different abilities is not a new one. In the mid eighteen hundreds, the
pseudoscience of phrenology was extremely popular. Thought at the time to be fact based
science, phrenology was based on the belief that the physical characteristics of the
skull—lumps, bumps, protrusions, and the like—were determined by the shape of the
brain underneath, and subsequently that areas of the brain that were assumed “larger” due
8
to the shape of the skull imparted increased aptitude or enhanced abilities in the assigned
corresponding “character trait” (Parssmen, 1974). Although phrenology has long been
debunked and relegated to the realm of pseudoscience, it has helped steer the scientific
community into the more modern field of cognitive neuroscience (Voros & Markic,
2014). For example, it is currently understood that different regions of the brain do serve
different functions: there is a region for processing sensory stimuli, one for managing and
coordinating movement, one for memory, decision making, language processing, and so
on (Stegmann, 2013).
Beginning in the early 1900s, the study of what we would refer to today as animal
cognition began in earnest. The terms “animal intelligence” and “animal mind” were used
by Thorndike and Washburn, respectively, around the turn of the century to theorize on
the psychological underpinnings of behavior in animals, whereas in the 1920s and 30s the
more well-known Pavlov and Skinner were examining the behavior of animals through
the lens of conditioning and reinforcement (Church, 2001). According to Church, “At
least four major intellectual movements in the 20th century have provided challenges to
the field of animal cognition: cognitive psychology, animal rights, ethology, and
molecular biology” (2001).
AVIAN INTELLIGENCE
While there have been numerous studies and paradigm shifts in the general
populous since the time of Descartes that have begun to undo the stereotype of the
9
unintelligent animal, until quite recently birds were thought to be some of the most
unintelligent in the animal kingdom. Phrases like “bird brain” commonly used by the
general public to refer to someone who is unintelligent and small minded support this
stereotype as a cultural mainstay. However, the truth is that many birds are very
intelligent.
Avian intelligence, though informally recognized through myth and religion for
thousands of years, has not been formally studied until the late twentieth century. It was
Darwin’s theory of evolution that first gave scientific respectability to the idea that
humans are not separate from or somehow above other animals, and it was Darwin’s
theory that subsequently introduced the study of animal behavior into the realm of
mainstream science (Mackintosh, 1994). However, studying the behavior of animals is
not the same as recognizing them as cognizant, sentient beings. Donald R. Griffin, a
pioneer of cognitive ethology and one of the first scientists to convincingly prove the
existence of echolocation in bats, caused a paradigm shift when he suggested that animals
have consciousness in his book “The Question of Animal Awareness,” first published in
1977. In it, he suggested that although it was unlikely that animals think exactly in the
same way that humans do, it would benefit the scientific community to consider the
possibility that animals have their own cognitive abilities: essentially, the idea that
animals think like animals (Griffin, 1977).
The physical structure of the avian brain has been one of the major causes of
dismissing the possibility of intelligence in birds. Bird brains do not have an elaborate
10
cerebral cortex, the part of the brain responsible for higher level cognitive abilities in
mammals. Due to this, any attempts to “train” birds—usually pigeons or parrots—using
standard psychological techniques that failed were blamed on the fact that birds seemed
to lack an area of the brain critical to higher thought processes and learning abilities
(Pepperberg, 2010). This idea was challenged by neuroscientist Stanley Cobb, who in the
1960s discovered that although avian brains are vastly different from mammalian brains,
they are complicated systems containing structures that are analogous to those in the
mammalian brain and are capable of complex cognitive feats (Cohen & Stemmer, 2007).
Birds belonging to the corvid (Corvidae) and parrot (Psittacinae) families have
forebrains that are comparatively large and relatively the same size as that of the great
apes, live in complex social groups sometimes including extended family ties, and have
young that have a long developmental period before becoming independent from their
parents (Emery, 2006).
The corvid family—crows, ravens, jays, nutcrackers, and magpies—is regarded as
one of the most intelligent bird families, perhaps rivalling the intelligence and problemsolving ability of the great apes (Kabadayi et al., 2017).
CORVIDS
Birds in the corvid family have been shown to have many abilities indicative of
higher intelligence such as problem solving, familial/social associations, long juvenile
period, tool use, and learning and idea transmission. New Caledonian crows are able to
11
not only use but to craft tools to complete novel tasks (Hunt & Gray, 2004), and
American crows and magpies can distinguish between human faces and can even
distinguish between the faces of friendly humans and foes (Marzluff et al., 2010; Lee et
al., 2011). Further examples of corvid intelligence is seen in the ability of Clark’s
nutcrackers to remember the location of their seed caches over 280 days after caching
(Balda & Kamil, 1992), and the ability of scrub jays, who can preferentially cache food in
“private” locations as opposed to locations that other birds have seen them use, thereby
deterring thieves (Grodzinski & Clayton, 2010).
The common raven (Corvus corax) is the largest bird in the passerine group.
Ravens can measure over 2 feet long and have an average weight of one and a half to
three and a half pounds (Boarman & Heinrich, 1999). They are larger than their more
commonly seen crow ‘cousins’, and have a more robust beak, a chevron shaped tail (as
opposed to a crow’s fan shaped tail) and distinctive long throat feathers called hackles;
ravens are also more solitary than crows, found in pairs much more than in the large
extended social groups that crows are known for (Boarman & Heinrich, 1999). Ravens
are the largest of all passerine birds. Passerines, sometimes referred to as “songbirds” or
“perching birds,” are defined by two key features: three toes facing forward and two toes
facing back, and their ability to sing (Unwin, 2011). Ravens form monogamous life-long
breeding pairs, are omnivorous although they seem to prefer carrion, and are known for
being very vocal and having a large call repertoire (Driver, 1999).
RAVENS AND MYTH
12
Ravens have always shared an interesting relationship with humans. Sometimes
seen as gods or helpful tricksters and sometimes reviled, ravens have been a part of
humans’ lives since time immemorial. Tales of ravens and their exploits and abilities
have been told since ancient times in many cultures around the globe. Native American
mythology varies from tribe to tribe, but many tribes have stories featuring Raven (which
is capitalized as a formal name in these tales). In the northwestern North American and
Siberian lore, Raven primary plays the role of the trickster. In this trickster role, Raven
often acts both altruistically and selfishly, making him paradoxical (Carroll, 1984).
Heinrich hypothesizes that tales of ravens aren’t just interesting stories, but that the myths
and legends surrounding the bird can give clues towards understanding their distribution
and even their food sharing and foraging behaviors (1989).
In other cultures, ravens have held various places in mythology from being a
messenger of the sun (Greek) or a gatherer of news for a god (Norse), predicting disasters
at sea (Alaskan Native) and epidemics (Dene), and even locating game animals for
human hunters (Blackfoot, Cheyenne, Koyukon, Dene, and Apache tribes; Driver, 1999).
Ravens are also thought to have brought light to the world and even to have created the
world in the beliefs of some North American native peoples (Plec, 2012). Ravens (and
other corvids) have also been discovered with some frequency in Iron Age and Roman
Britain burial sites, though the meaning of their inclusion in these burials is still up for
debate (Serjeantson & Morris, 2011). Similarly, the skeletal remains of ravens have also
13
been found in Paleoindian burial sites over 10,000 years old with evidence suggesting
that humans buried the remains intentionally (Driver, 1999).
Humans have had close relationships with ravens since time immemorial. In fact,
there have been studies confirming that ravens associate with human hunters. Ravens are
scavenging birds and have been seen at the gut piles left from carcasses of animals killed
by human hunters (White, 2005). This seems counterintuitive as gunshots signal danger
and disturbance, both of which are typically avoided by animals—especially scavenger
animals that have a history of being shot at (White, 2005). White hypothesizes that due
to the long lifespan of ravens it’s unlikely that they have evolved an affinity for gunshot
sounds in so few generations, and instead suggests that ravens have learned to associate
gunshot sounds with food as a foraging strategy (2005). This behavior is hypothesized
to have replaced the relationship between wolves and ravens in an environment that has
become distinctly lacking in wolves due to human involvement.
RAVENS AND WOLVES
Ravens have been known to associate closely with wolves. It is thought that this
association with wolves is a survival strategy on the raven’s end, as ravens primarily feed
on carrion which can be difficult to locate in winter; associating with wolves almost
guarantees them a year-round food source (Vucetich et al., 2004). However, this is not
the only time during which ravens can be seen with wolves. It has been found that ravens
associate with wolves not only when they are feeding on kills but when they rest, travel,
14
and hunt prey, and that a similar association does not occur between ravens and other
animals (Stahler et al., 2002).
Ravens have been observed to be cautious of new food sources, going so far as to
avoid approaching a food source until other animals begin feeding (Boarman & Heinrich,
1999). However, when ravens associating with wolves were studied for this innate
reaction it was found that ravens did not exhibit this fear response towards carcasses
when wolves were present (Stahler et al., 2002), possibly indicating that ravens connect
wolves feeding with the safety to eat carrion. It has also been observed that ravens are
attracted to the distress calls of dying animals, suggesting that auditory cues may be one
way that they are able to locate wolf kill sites to scavenge from (Frame, 2010).
Wolves and ravens make a great hunting team. If a flock of ravens happens upon
a carcass of an animal that has died of natural causes or has frozen to death in the winter,
their beaks are not strong enough to access the meat; if a predator such as a wolf kills an
animal, ravens can take advantage of the situation and scavenge some of the meat for
themselves with greater ease (Plec, 2012). Between eating and caching, an individual
raven can consume or stash away up to 2 kg of food per day from large prey animal
carcasses, and it is estimated that for every one wolf that participates in a kill, between 2
to 4 kg of meat are lost to scavenging ravens (Vucetich et al., 2004). Although it is easy
to imagine that ravens may associate with wolves when food is involved, ravens also
associate with wolves for other reasons not yet entirely understood by humans. Heinrich
also observed that ravens at Yellowstone preferentially associated with wolves, even
15
when there was no meat in the immediate vicinity (1999). Although wolves hunt for
themselves and their packmates and not for ravens, wolves have been found to be
remarkably tolerant of the birds. Ravens have been observed to exhibit exceptionally
brave, play-like behaviors around wolves, diving at wolves and even nipping at their tails,
even after being snapped at (Heinrich, 1989).
Even the population dynamics of the two species are intertwined: it was found
that when grey wolves were reintroduced to Yellowstone National Park in the late 1990s,
the ravens in the park began to depend less on anthropogenic food sources, began to
associate preferentially with wolves, and experienced population stabilization (Walker et
al., 2018). Associating with wolves benefits ravens in other ways as well: the number of
eggs laid by ravens annually has been found to correlate with the availability of carrion
(Newton et al., 1982), with more carrion leading to more eggs. In this way, wolves
positively impact the yearly growth of raven populations.
RAVEN COGNITION
Historical accounts, myths, and observations like these have led towards the
formal scientific study of ravens as intelligent animals capable of many complex mental
abilities. Memory, and the ability to act based upon memory, is one such complex mental
ability. Episodic memory, the ability to remember events that an individual personally
experienced, allows for contextual information for memory retrieval; this provides for a
more rich and complete retrieval of memory (Kelley et al., 2018). The ability of scrub
jays in the wild to locate their caches accurately even after months was stated to be the
16
first conclusive evidence of behavior indicating episodic memory in animals other than
humans by Clayton and Dickinson (as cited in Salwiczek et al., 2010). Ravens, like their
scrub jay cousins, have also shown to have strong memory-related abilities.
Like scrub jays, ravens cache food; they are also similar to scrub jays in that they
both raid other ravens’ caches and anticipate that their own caches may be raided if they
are not careful to cache out of view of any potential competitors (Bugnyar & Kotrschal,
2002). However, the ability of ravens to remember is not restricted to food related events.
Kabadayi and Osvath found that ravens who were trained to exchange tokens for food
from the researchers would, when they knew trading would happen the next day, collect
and store tokens as soon as the opportunity to take the tokens was presented to them (as
cited in Boeckle & Clayton, 2017). When ravens participate in experiments with human
experimenters, they remember which experimenters are fair and which are unfair and can
remember this for up to a month; they then also show preference to those experimenters
that they remember for their fairness (Muller et al., 2017).
Ravens also form strong bonds and memories for other birds in their social circle.
They are able to remember former group members by their calls after spending up to
three years separated from them (Boeckle & Bugnyar, 2012). Ravens also have strong
relationships with their siblings - stronger than their relationships with other unrelated
birds (Fraser & Bugnyar, 2010). Additionally, they receive more benefit from social
learning experiences with the other bird or birds involved are their siblings (Schwab et
al., 2008).
17
RAVEN VOCALIZATION
As social birds, ravens communicate in a variety of ways. Perhaps the most
noticeable way they communicate is through their vocalizations.
Avian vocalization is fairly unique amongst animal communication in that birds,
unlike the great majority of animals, are able to learn from the vocalizations of others.
Although there have been anecdotes about other animals—humpback whales, porpoises,
orca, harbor seals, bats, and elephants—being able to imitate vocal sounds, the best and
most well documented examples come from birds (Nottebohm & Liu, 2010).
Birds are very creative and able in their vocal abilities. They are able to learn the
calls and songs of their own species in the wild, and under laboratory conditions many
species of songbird are also able to learn the songs of “tutors” of a different species
(Nottebohm & Liu, 2010). Some species of birds, such as parrots and mynahs, are
extremely skilled at mimicking and making a variety of sounds and are also able to form
associations between a particular auditory stimulus and a specific response in a seemingly
unconstrained manner (Seyfarth & Cheney, 2010). A high level of complexity and
sophistication in avian communication has been found from field research in recent
decades, a complexity that may rival the communication abilities of non-human primates
(Suzuki, 2016).
18
Ravens, with their large brains, are especially skilled at communication. As
reviewed in Reber et al. (2016), ravens are able to distinguish between the calls of ravens
that are familiar and those that are unfamiliar, or even ravens that were former associates;
they can also recognize in group or out group membership of other ravens based on
vocalizations. Societal changes can also be detected by ravens through vocalizations.
When recorded raven calls are played back over loudspeaker to the same group of ravens
that the recording was taken from, the ravens are able to recognize the calls as coming
from a bird that they are familiar with and pay attention to the social rank of the birds
they are hearing; when researchers altered the recording to make dominant individuals
sound as if they had become submissive, the listening ravens would stop what they were
doing and show signs of distress (De Waal, 2016). This shows the importance of
vocalizations in developing and maintaining the social ties between ravens.
Ravens also yell at food sources both to communicate their status in the group and
to attract others to the food (Heinrich & Marzluff, 1991). Social status is also an
important factor in group feeding behavior. At a food source, ravens are more likely to
respond to the food-related haa calls of other ravens that are familiar to them, especially
those of their same social status, effectively choosing their dining groups based on
vocally transmitted information (Szipl et al., 2015).
NEW FRONTIERS
The body of work on avian cognition grows larger as research in new areas
expands to include more facets of animal intelligence yet to be explored. Like the
19
aforementioned study by Kabadayi and Osvath, this study explores another hypothesis
related to a raven’s ability to plan for the future and behave accordingly: the hypothesis
that wild ravens can anticipate the occurrence of scheduled feedings in a sanctuary setting
independently of behavioral cues from the captive wolves that reside there.
20
CHAPTER 3: METHODS
Although wild ravens are typically territorial animals living together in pair bonds
upon reaching breeding age, they do form bonds with other ravens that they remember
for years (Braun & Bugnyar, 2012) and tend to gather at sources of food whether led
there by vocal or behavioral cues from conspecifics, especially as unpaired young adults
(Bugnyar & Kotrschal, 2001). However, in situations where food is readily available on a
regular basis such as at a garbage dump, camp site, or game park, ravens may form more
permanent flocks (Loretto, 2016). The latter is the case at Wolf Haven International
(“Wolf Haven”). Wolf Haven is a not for profit animal sanctuary located in the city of
Tenino in Washington State, taking in displaced and formerly captive grey wolves (Canis
lupus) and becoming a permanent home for them. They also participate in species
recovery for the endangered Mexican grey wolf (Canis lupus baileyi) and red wolf (Canis
rufus). At any given time, Wolf Haven is home to approximately 60 wolves, though
numbers fluctuate with births, deaths, and new arrivals from outside. Wolf Haven is also
home to large population of wild ravens that live on the sanctuary premises year-round.
SITE SELECTION
Wolf Haven was selected as the site for this study due to both its large resident
raven population and the manner in which the animal care workers feed the resident
wolves. In the wild, wolves are only successful at bringing down prey about 10% of the
time (Mech et al., 2015); To mimic the natural feeding habits of wolves as much as is
possible in captivity, the wolves are not fed every day. Instead, the wolves are fed 3 to 4
times per week with Saturday always being a scheduled feeding day and Sunday always
21
being a scheduled fasting day with very few, very infrequent exceptions (Erik Wilber of
Wolf Haven International, personal communication, August 1, 2018).
On feeding days, wolves are fed a diet of chicken, fish, and beef, as well as deer
carcasses as they are available. Food is typically given to the wolves around noon unless
there is an emergency at the sanctuary. On fasting days wolves are typically supplied
with enrichment items, which are usually not food items; however, on hot days they may
be supplied with “popsicles” made of frozen tuna juice or blood.
OBSERVATION SCHEDULE AND RECORDING
No regular sanctuary activities were disrupted by data collection. Observations
took place every Saturday and Sunday from September 15, 2018 to February 3, 2019,
beginning at 11:00 a.m. and lasting for at least one hour but for no more than two hours.
The first two days of recording—September 15 and 16—no haa calls were recorded and
no feeding was observed. Scheduling adjustments were made with Wolf Haven staff to
ensure that, barring any unforeseen emergency circumstances, feeding would be observed
on all subsequent observation days. On two days—December 22 and January 12—
technical issues with the recording equipment caused recordings to be only 46 and 39
minutes in duration, respectively. Recording on January 13 was also cut short due to poor
weather. This observation window was selected in order to allow for observation of
ravens prior to the scheduled wolf feeding time, which under normal circumstances
occurs around noon on feeding days. A total of just over 58 hours of recording was
completed for this study.
22
Audio recordings were taken using a ZOOM H4N Pro handheld recording device.
Observations took place at the same stationary location near the center of the public
visitor route (a loop of 6 wolf enclosures). The recording device was attached to a small
“tabletop” tripod with a shock mount, a sound buffer was placed over the microphone,
and the device was then placed directly onto the ground. The recorder was oriented
towards the south for every recording session. Written notes were taken from a chair
placed approximately 1 to 3 feet away from the recorder (Figure 1).
Figure 1 Diagram of Wolf Haven study site
23
ANALYSIS
Recordings were imported into Cornell Lab’s “Raven” bioacoustic software for
spectrogram analysis. Captured sound files were displayed as spectrograms in 30 second
intervals at a time and manually searched for the signature of the haa call (Figure 2). The
haa call was selected as the focus of the data collection for this study because it has
reliably been shown to be associated with food (Szipl et al., 2015; Bugnyar et al., 2001;
Conner, 1985).
Figure 2. Haa call spectrogram (recorded October 27, 2018 at Wolf Haven International)
Figure 3. Two examples of non-haa raven calls (recorded January 12, 2019 at Wolf Haven International)
24
Calls that were ambiguous or overlapping with other calls were not included in
the final call count. Additionally, days during which an unusual event causing more raven
vocalizations than normal (e.g. the feeding of an entire deer carcass to one wolf
enclosure) were noted as outliers and not included in the final data tables. The accepted
haa calls were then sorted by start time. The accepted calls were then reviewed a second
time for accuracy. On days when wolf feeding was scheduled, the feeding time was
noted. It was also noted when a feeding was scheduled but not observed, and when
scheduled feeding days and fasting days were swapped by Wolf Haven animal care staff
for the benefit of this study. Other disturbances, such as train whistles or noise from
airplanes, were also recorded, as well as any wolf behaviors suggesting that the wolves
were anticipating being fed soon.
A paired t-test for two sample means was used to determine whether the
difference in the average number of haa calls per hour on feeding days and on fasting
days was statistically significant. A paired two sample t-test for means on observed
feeding days comparing the number of haa calls before and after feeding was also
conducted. Additionally, a two-sample t-test assuming unequal variances was conducted
to compare the average number of haa calls per hour on feeding days when feeding went
as planned to scheduled feeding days where feeding was not observed. This analysis did
not include weekends where feeding and fasting days were switched by Wolf Haven
staff, nor one Sunday on a weekend during which the wolves were fed on both Saturday
and Sunday but the feeding was unobserved. A more fine-grained analysis was also
employed via a one-way ANOVA to examine the number of calls in the 30 minutes
25
before and after feedings, and before and after the average feeding time of noon on nonfeeding days and on days during which feeding was not observed. Finally, for the two
weekends during which Wolf Haven staff switched the feeding and fasting days, the
average number of calls per hour on the “switched feeding day,” the Sunday in both
cases, was compared to a typical feeding day and the average number of calls per hour on
the “switched fasting day” (the Saturday) was compared to a typical fasting day.
The haa call frequency from each recording day was also plotted as a scatterplot
and a generalized additive model (“GAM”) was employed to visually detect patterns in
the recorded vocalizations and the time of observed feeding, if applicable, was marked
(Appendix B).
26
CHAPTER 4: RESULTS
The paired two sample t-test for means comparing the average number of haa
calls per hour on feeding days (including days upon which feeding was scheduled to
occur but was not observed) to the average number of haa calls per hour on fasting days
yielded the following results: the mean number of haa calls on feeding days (15 ± 2 SE)
was similar to the mean number of calls on fasting days (13 ± 3 SE; paired t16 = 0.60, p =
0.56).
Table 1. Average number of haa calls on feeding and fasting days
WEEKEND
FEEDING DAY
AVERAGE
9/29 & 9/30
10/6 & 10/7
10/13 & 10/14
10/20 & 10/21
10/27 & 10/28
11/3 & 11/4
11/10 & 11/11
11/17 & 11/18
12/1 & 12/2
12/8 & 12/9
12/15 & 12/16
12/22 & 12/23
1/5 & 1/6
1/12 & 1/13
1/19 & 1/20
1/26 & 1/27
2/2 & 2/3
7.98
22.51
8.03
22.24
4.02
12.59
10.78
25.15
9.39
14.10
11.60
12.89
26.08
15.47
2.37
21.54
28.43
FASTING DAY
AVERAGE
1.04
15.02
35.37
11.54
11.18
3.04
2.15
41.71
7.12
31.02
15.38
6.28
13.42
6.15
9.32
10.89
1.03
For the paired two sample t-test for means on observed feeding days comparing
the number of haa calls before and after feeding, the mean number of haa calls before
feeding (13.9 ± 4.4 SE) was similar to the mean number of haa calls after feeding (11.5 ±
3.4 SE; paired t11 = 0.36, p = 0.73).
27
Table 2. Haa calls on observed feeding days
DATE
9/29/2018
10/6/2018
10/13/2018
10/27/2018
11/3/2018
11/10/2018
11/17/2018
12/1/2018
12/22/2018
1/6/2019
1/27/2019
2/2/2019
HAA CALLS
BEFORE
FEEDING
10
46
1
7
7
13
42
11
10
0
20
0
HAA CALLS
AFTER
FEEDING
5
1
15
1
19
8
8
7
0
33
7
34
The results of the two-sample t-test assuming unequal variances comparing the
average number of haa calls per hour on feeding days when feeding went as planned
(14.81 ± 2.61 SE) to scheduled feeding days where feeding was not observed (13.16 ±
3.22 SE) resulted in t = 0.25 and p = 0.81.
Table 3 Haa calls per hour on scheduled feeding days
DATE
9/29
10/6
10/13
10/27
11/3
11/10
11/17
12/1
12/22
2/2
FEEDING AS
SCHEDULED
7.98
22.51
8.03
4.02
12.59
10.78
25.15
9.39
12.89
28.43
DATE
10/20
12/8
12/15
1/13
1/19
FEEDING
UNOBSERVED
22.24
14.10
11.60
15.47
2.37
The mean number of calls divided into three groups: A) 60 minutes surrounding
observed feedings (30 minutes before and 30 minutes after), B) 60 minutes surrounding
28
the average feeding time of noon on non-feeding days (from 11:30 a.m. to 12:30 p.m.),
and C) 60 minutes surrounding noon on days during which feeding was scheduled but not
observed, were statistically similar (F2, 29 = 0.29, p = 0.75). On observed feeding days, the
average number of haa calls was 12.33 ± 2.43 SE. On non-feeding days the average was
12.29 ± 4.17 SE, and on days when feeding was scheduled but not observed, the average
was 8.16 ± 3.10 SE.
Table 4 Summary of data used for ANOVA
Groups
Feeding days
Scheduled but unobserved
feeding days
Non feeding days
Count
12
6
14
Sum
148
49
Average
12.33
8.17
172
12.29
Variance
70.97
57.77
243.30
Regardless of the specific comparison, the differences in haa call frequencies in
all described scenarios were not statistically significant. Additionally, review of the
scatterplots with added GAM analysis did not yield any detectable pattern in frequency of
haa calls.
29
CHAPTER 5: CONCLUSION AND DISCUSSION
As shown in the previous section, both the difference in number of haa calls
between feeding days and fasting days and the difference between the number of haa
calls when feeding went on as scheduled versus when feeding was schedule but not
observed were not statistically significant. Additionally, no obvious pattern was apparent
when examining plots of the data. This result could be for a number of reasons.
As has been established, ravens are extremely intelligent birds that exhibit insight
and complex thought processes, as well as the ability to remember other birds, people,
and events. As described by White, ravens have been observed as being capable of
learning to associate certain auditory cues with food; in the case of White’s study, ravens
associated the sound of a hunting rifle with the opportunity to feed on fresh carcasses
(2005). As White points out, due to the longevity of ravens and their relative slowness to
reproduce when compared to some other animals, it is not possible for ravens to have
evolved to be attracted to the sounds of gunshots, so some other mechanism must be at
play. He hypothesizes that due to the fact that ravens already associate certain sounds,
such as the calls of other ravens and possibly that of wolf howls, with an accessible food
source, ravens have also simply learned that the sound of a hunting rifle firing also is
associated with food. According to studies of raven foraging styles, once ravens are
alerted to the presence of food – such as they might be by a gunshot – they often times
return to their roosting location to alert other ravens of the presence of a food source
(Dall & Wright, 2009). Due to Wolf Haven’s relatively small area and the ease of
accessing food for resident ravens treating the grounds as a roost, it is possible that due to
30
the plentiful food and lack of motivation to travel the behavior of resident ravens has
changed due to learning and becoming adapted to their living situation, just like ravens
living near hunters become accustomed to their gunshots signaling food.
Just like the multifaced role of Raven in myth – a trickster, a knowledge bringer,
and even a god – ravens in the natural world are also multifaceted and can be difficult for
humans to understand and are easy for us to misread (Munday, 2013). There is a
widespread belief in many cultures that, due to their vocal nature and association with
humans, ravens bring news to or alert people of future events (Drive, 1999). Ravens in all
of their ranges have been noted to be extremely vocal and endowed with a large
vocabulary, as well as exhibiting regional “dialects;” that is, certain calls that are only
present in discreet populations (Conner, 1985). It is possible that the population of ravens
at Wolf Haven has been stationary for long enough to form their own dialect, leading to
the potential for misunderstandings of their vocalizations. However, more studies of this
specific group and their vocalizations would need to occur before any claims to the
validity of this theory could be made.
The choice of Wolf Haven as the site for the study also brought about challenges.
Although the feeding of the wolves took place on a regular schedule on weekends,
sometimes the feeding days during the week were more variable and did not occur in
regular intervals. This could impact the ability of the ravens to internalize a feeding
schedule and their ability to “know” when the weekend was coming or tell Saturday
feeding days apart from Sunday fasting days. Additionally, as touched upon earlier, even
31
on the weekend some unexpected, unscheduled events occurred that likely interfered with
data collection. These included two occasions upon which an entire deer carcass was fed
to a group of wolves, increasing the number of haa calls, and days during which feeding
was scheduled but did not occur during the observation period, which had an unknown or
undetectable effect on the number of haa calls.
In addition to unplanned scheduling abnormalities, there were environmental
factors that may have impacted the calling habits and call frequency of the ravens. For
example, the food for the wolves was delivered via an older pickup truck that had a
distinctive noise both when it was first started and upon driving. This sound was audible
from the observation point in the sanctuary. The wolves, with their superior sense of
hearing, visibly reacted to the sound of the food truck by assuming a physically alert
posture and in some occasions whimpering. The ravens could easily observe the wolves
and it is possible that their behavior was influenced by that of the wolves. Enrichment for
the wolves also posed an unexpected challenge, as when enrichment was delivered the
food truck was used to transport the enrichment items from the main office to the wolf
enclosures. This triggered the same reaction in the wolves as food delivery, which in turn
could have altered the behavior of the ravens.
Finally, the behavior of the ravens themselves could have posed a problem with
the study design. Ravens, like all corvids, are known to cache food, observe others
caching food, and raid the caches of others (Bugnyar & Kotrschal, 2002). Due to the
advanced ages of many of the wolves at Wolf Haven, some are slow eaters, leaving
32
ample time for ravens to steal away food for eating or for caching. Additionally, some of
the oldest wolves have smaller appetites and thus leave leftovers, allowing the ravens
even more chances to eat or cache what the wolves leave behind. These circumstances
allow for food to be available to ravens even on fasting days, which could cause the
ravens to make calls signaling food or hunger less than they typically would in a situation
where food, both to eat and to cache, was less available.
The study design outlined here may yield more conclusive results when
conducted under laboratory settings. The controlled conditions of a laboratory would
eliminate any external environmental factors that could have had an effect on the
behavior of the ravens such as the aforementioned scheduling issues, unusual
occurrences, environmental “clues,” and copious opportunities for storing food. Under a
laboratory setting, a strictly adhered to feeding schedule could be implemented, and any
haa calls made by the ravens in the study would be uninfluenced by the myriad of
external factors encountered at Wolf Haven.
However, a laboratory setting cannot mimic natural behavior in the wild.
Although a wolf sanctuary also cannot perfectly mimic conditions unimpacted by human
schedules and feedings, is does allow for wild ravens to react to novel situations and can
give insights into their natural problem solving and social behaviors. Additional studies,
such as those examining the relationship between wolves and ravens more closely, would
benefit from a setting such as Wolf Haven. Although this study did not find a significant
difference between the number of haa calls on feeding days versus those on fasting days,
33
perhaps with an altered study design or a longer length of time for data collection there
would be a different outcome. Another possible avenue of investigation is to apply this
study design to other sanctuary or zoo settings that have a local wild raven population,
ideally a raven population that interacts with wolves residing at said sanctuary or zoo.
Additional research into the scavenging behavior of wild ravens interacting with
animals in captivity could be leveraged into applications such as preventing food loss at
zoological parks and lessening or eliminating undesirable behaviors of wild ravens that
are in close contact with humans. As human populations continue to grow, contact with
wild animals that are usually reclusive will continue to increase. Further studies of such
animals, like ravens, may contribute to a more cooperative human-animal relationship
with increased understanding of the animal cousins that we share the planet with.
34
Bibliography
Adamson, P., & Edwards, G. (Eds.). (2018). Animals: A history. Oxford: Oxford
University Press.
Allen, J. (2009). The lives of the brain : Human evolution and the organ of mind.
Cambridge, Mass.: Belknap Press of Harvard University Press.
Anderson, J. R. (2002). Gone fishing: tool use in animals. Biologist, 49(1), 15–18.
Balda, R. P., & Kamil, A. C. (1992). Long-term spatial memory in clark's nutcracker
nucifraga columbiana. Animal Behaviour, 44(4), 761-769.
Boarman, W. I. and B. Heinrich (1999). Common Raven (Corvus corax), version 2.0. In
The Birds of North America (A. F. Poole and F. B. Gill, Editors). Cornell Lab of
Ornithology, Ithaca, NY, USA.
Boddice, R. (2011). Anthropocentrism : Humans, Animals, Environments. Leiden:
BRILL.
Boeckle, M., & Bugnyar, T. (2012). Long-term memory for affiliates in ravens. Current
Biology: CB, 22(9), 801–806.
Boeckle, M., & Clayton, N. (2017). A raven's memories are for the future. Science (New
York, N.Y.), 357(6347), 126-127.
Boeckle, M., Szipl, G., & Bugnyar, T. (2012). Who wants food? Individual
characteristics in raven yells. Animal Behaviour, 84, 1123–1130.
Boeckle, M., Szipl, G., & Bugnyar, T. (2018). Raven food calls indicate sender’s age and
sex. Frontiers in Zoology, 15(1), 1–10.
Braun, A., & Bugnyar, T. (2012). Social bonds and rank acquisition in raven nonbreeder
aggregations. Animal Behaviour, 84(6), 1507–1515.
35
Bugnyar, T., & Heinrich, B. (2005). Ravens, Corvus corax, differentiate between
knowledgeable and ignorant competitors. Proceedings: Biological Sciences,
272(1573), 1641–1646.
Bugnyar, T., & Kotrschal, K. (2001). Movement coordination and signalling in ravens.
Acta Ethologica, 3(2), 101–110.
Bugnyar, T., & Kotrschal, K. (2002). Observational learning and the raiding of food
caches in ravens, Corvus corax: Is it “tactical” deception? Animal Behaviour,
64(2), 185–195.
Bugnyar, Kijne, & Kotrschal. (2001). Food calling in ravens: Are yells referential
signals? Animal Behaviour, 61(5), 949-958.
Bugnyar, T., Stöwe, M., & Heinrich, B. (2004). Ravens, Corvus corax, follow gaze
direction of humans around obstacles. Proceedings: Biological Sciences,
271(1546), 1331–1336.
Carroll, M. (1984). The Trickster as Selfish-Buffoon and Culture Hero. Ethos, 12(2),
105-131.
Church, R. M. (2001). Animal cognition: 1900–2000. Behavioural Processes, 54(1-3),
53-63
Cohen, H., & Stemmer, Brigitte. (2007). Consciousness and cognition fragments of mind
and brain. London: Academic Press.
Conner, R. N. (1985). Vocalizations of Common Ravens in. The Condor, 87(3), 379–388.
Dall, S. R. X., & Wright, J. (2009). Rich pickings near large communal roosts favor
“gang” foraging by juvenile common ravens, corvus corax. PLoS ONE, 4(2).
36
De Waal, F. (2016). Are we smart enough to know how smart animals are? New York:
W.W. Norton & Company.
Driver, J. (1999). Raven Skeletons from Paleoindian Contexts, Charlie Lake Cave,
British Columbia. American Antiquity, 64(2), 289-298.
Emery, N. J. (2006). Cognitive ornithology: The evolution of avian intelligence.
Philosophical Transactions Biological Sciences, 361(1465), 23–43.
Emery, N. J., & Clayton, N. S. (2004). The Mentality of Crows: Convergent Evolution of
Intelligence in Corvids and Apes. Science, 306(5703), 1903-1907.
Fitch, W. T., Porteš, M., Němec, P., Herculano-Houzel, S., Lučan, R. K., Kocourek, M.,
& Olkowicz, S. (2016). Birds have primate-like numbers of neurons in the
forebrain. Proceedings of the National Academy of Sciences, 113(26), 7255–7260.
Frame, P. (2010). Observations of a Possible Foraging Tool used by Common
Ravens. The Wilson Journal of Ornithology, 122(1), 181-182.
Fraser, O. N., & Bugnyar, T. (2010). The quality of social relationships in ravens. Animal
Behaviour, 79(4), 927–933.
Gould, J. L. (2004). Animal cognition. Current Biology, 14(10), R372-R375
Griffin, D. R. (1977). The question of animal awareness: Evolutionary continuity of
mental experience. New York, NY: Rockefeller University Press.
Grodzinski, U., & Clayton, N. S. (2010). Problems faced by food-caching corvids and the
evolution of cognitive solutions. Philosophical Transactions of the Royal Society
B: Biological Sciences.
Heinrich, B. (1989). Ravens in winter (2014 print ed.). New York, NY: Simon &
Schuster Paperbacks.
37
Heinrich, B. (1994). Does the Early Common Raven Get (and Show) the Meat? The Auk,
111(3), 764-769.
Heinrich, B. (1999). Mind of the raven: An investigation into the mind of the raven. New
York: Cliff Street Books.
Heinrich, B., & Marzluff, J. M. (1991). Do Common Ravens Yell because They Want to
Attract Others ? Behavioral Ecology and Sociobiology, 28(1), 13–21.
Hunt, G. R., & Gray, R. D. (2004). Direct observations of pandanus-tool manufacture and
use by a new caledonian crow ( corvus moneduloides).Animal Cognition, 7(2),
114-120.
Iwaniuk, A. N., & Nelson, J. E. (2002). Can endocranial volume be used as an estimate of
brain size in birds? Canadian Journal of Zoology, 80(1), 16–23.
Kabadayi, C., & Osvath, M. (2017). Ravens parallel great apes in flexible planning for
tool-use and bartering. Science.
Kelley, T. D., Thomson, R., & Milton, J. R. (2018). Standard model of mind: Episodic
Memory. Procedia Computer Science, 145, 717–723.
Lee, W. Y., Lee, S., Choe, J. C., & Jablonski, P. G. (2011). Wild birds recognize
individual humans: Experiments on magpies, pica pica. Animal Cognition, 14(6),
817-25.
Loretto, M.-C., Reimann, S., Schuster, R., Graulich, D. M., & Bugnyar, T. (2016). Shared
space, individually used: spatial behaviour of non-breeding ravens (Corvus corax)
close to a permanent anthropogenic food source. Journal of Ornithology, 157,
439–450.
38
Mackintosh, N. (1994). Animal learning and cognition (Handbook of perception and
cognition (2nd ed.)). San Diego: Academic Press.
Marzluff, J. M., Walls, J., Cornell, H. N., Withey, J. C., & Craig, D. P. (2010). Lasting
recognition of threatening people by wild American crows. Animal Behaviour,
79(3),
Müller, J. J. A., Massen, J. J. M., Bugnyar, T., & Osvath, M. (2017). Ravens remember
the nature of a single reciprocal interaction sequence over 2 days and even after a
month. Animal Behaviour, 128, 69–78.
Munday, P. (2013). Thinking through ravens: Human hunters, wolf-birds and embodied
communication. Perspectives on Human-Animal Communication: Internatural
Communication, 207–225.
Newton, I., Davis, P. E., & Davis, J. E. (1982). Ravens and Buzzards in Relation to
Sheep Farming and Forestry in Wales. Journal of Applied Ecology, 19(3), 681–
706.
Nomura, T., & Izawa, E. I. (2017). Avian brains: Insights from development, behaviors
and evolution. Development Growth and Differentiation, 59(4), 244–257.
Nottebohm, F., & Liu, W. C. (2010). The origins of vocal learning: New sounds, new
circuits, new cells. Brain and Language, 115(1), 3–17.
Parssmen, T. M. (1974). Popular Science and Society: The Phrenology Movement in
Early Victorian Britain. Journal of Social History, 8(1).
Pepperberg, I. M. (2010). Alex: A Study in Avian Cognition. Encyclopedia of Animal
Behavior, 44–49.
39
Plec, E. (2013). Perspectives on human-animal communication : Internatural
communication (Routledge studies in rhetoric and communication ; 12). New
York: Routledge.
Reber, S. A., Boeckle, M., Szipl, G., Janisch, J., Bugnyar, T., & Fitch, W. T. (2016).
Territorial raven pairs are sensitive to structural changes in simulated acoustic
displays of conspecifics. Animal Behaviour, 116, 153–162.
Salwiczek, L. H., Watanabe, A., & Clayton, N. S. (2010). Ten years of research into
avian models of episodic-like memory and its implications for developmental and
comparative cognition. Behavioural Brain Research, 215(2), 221-234.
Scheid, C., Range, F., & Bugnyar, T. (2007). When, What, and Whom to Watch?
Quantifying Attention in Ravens (Corvus corax) and Jackdaws (Corvus
monedula). Journal of Comparative Psychology, 121(4), 380–386.
Schwab, C., Bugnyar, T., Schloegl, C., & Kotrschal, K. (2008). Enhanced social learning
between siblings in common ravens, Corvus corax. Animal Behaviour, 75(2),
501–508.
Serjeantson, D., & Morris, J. (2011). Ravens and Crows in Iron Age and Roman
Britain. Oxford Journal of Archaeology, 30(1), 85–107.
Seyfarth, R. M., & Cheney, D. L. (2010). Production, usage, and comprehension in
animal vocalizations. Brain and Language, 115(1), 92–100.
Stahler, D., Heinrich, B., & Smith, D. (2002). Common ravens, Corvus corax,
preferentially associate with grey wolves, Canis lupus, as a foraging strategy in
winter. Animal Behaviour, 64(2), 283-290.
40
Stegmann, U. (2013). Animal Communication Theory : Information and Influence. New
York: Cambridge University Press.
Suzuki, T. N. (2016). Semantic communication in birds: evidence from field research
over the past two decades. Ecological Research, 31(3), 307–319.
Szipl, G., & Bugnyar, T. (2014). Craving Ravens: Individual ‘haa’ Call Rates at Feeding
Sites as Cues to Personality and Levels of Fission-Fusion Dynamics? Animal
Behavior and Cognition, 1(3), 265.
Szipl, G., Boeckle, M., Wascher, C. A. F., Spreafico, M., & Bugnyar, T. (2015). With
whom to dine? Ravens’ responses to food-associated calls depend on individual
characteristics of the caller. Animal Behaviour, 99, 33–42.
Taylor, A. H. (2014). Corvid cognition. Wiley Interdisciplinary Reviews: Cognitive
Science, 5(3), 361–372.
Unwin, M. (2011). The Atlas of Birds : Diversity, Behavior, and Conservation. Princeton:
Princeton University Press.
Voros, S., & Markic, O. (2014). Photograph of the Soul: Towards a Critical
Neuroscience. Teorija in Praksa, 51(5), 811–831.
Vucetich, J. A., Peterson, R. O., & Waite, T. A. (2004). Raven scavenging favours group
foraging in wolves. Animal Behaviour, 67(6), 1117-1126.
Walker, L. E., Marzluff, J. M., Metz, M. C., Wirsing, A. J., Moskal, L. M., Stahler, D. R.,
& Smith, D. W. (2018). Population responses of common ravens to reintroduced
gray wolves. Ecology and Evolution, 8(22), 11158–11168.
White, C. (2005). Hunters Ring Dinner Bell For Ravens: Experimental Evidence Of A
Unique Foraging Strategy. Ecology, 86(4), 1057-1060.
41
Zentall, T. R. (2005). Animals may not be stuck in time. Learning and Motivation, 36,
208–225.
42
Appendices
Appendix A
9/15 Saturday1
9/16 Sunday2
9/29 Saturday
RECORDING
START TIME
11:01
11:10
11:11:45
Not observed
Not observed
12:31:45
9/30 Sunday
10/06 Saturday3
11:01:32
10:54:54
n/a
12:32:15
10/07 Sunday
10/13 Saturday
10:57:32
10:56:56
n/a
12:16
10/14 Sunday
10/20 Saturday4
10/21 Sunday
10/27 Saturday
10:00:00
10:57:03
11:00:26
10:55
n/a
Not observed
n/a
12:38
10/28 Sunday
11/03 Saturday
11/04 Sunday
11/10 Saturday
11:01:51
11:00
11:04:37
11:06
n/a
11:32
n/a
12:35
11/11 Sunday5
11/17 Saturday
11:09
10:58
n/a*
12:39
11/18 Sunday
12/01 Saturday6
12/02 Sunday
12/08 Saturday
12/09 Sunday
12/15 Saturday7
12/16 Sunday8
12/22 Saturday
12/23 Sunday
01/05 Saturday
01/06 Sunday
01/12 Saturday
01/13 Sunday
01/19 Saturday
01/20 Sunday
01/26 Saturday
01/27 Sunday
11:06
11:03
11:10
11:07
11:07
11:04
11:05
11:08
11:04
11:00
11:11
11:04
11:05
11:04
11:04
11:03
11:07
n/a
12:31
n/a
Not observed
n/a
Not observed
Not observed
12:29
n/a
Food switch week
11:53
Not observed
n/a
Not observed
n/a
Food switch week
11:51
DAY
43
FEEDING TIME
FEEDING IN MINTUES
AND SECONDS
n/a
n/a
1 hr 20 min / 80 min / 4800
sec
n/a
1 hr 37 min 21 sec / 97 min
/ 5841 sec
n/a
1 hr 19 min 4 sec / 79 min /
4744 sec
n/a
n/a
n/a
1 h4 43 min / 103 min /
6180 sec
n/a
32 min / 1920 sec
n/a
1 hr 29 min / 89 min / 5340
sec
n/a
1 hr 41 min / 101 min /
6060 sec
n/a
1 hr 28 min / 88 min / 5280
n/a
n/a
n/a
n/a
n/a
1 hr 21 min / 81 min / 4860
n/a
n/a
42 min / 2520 sec
n/a
n/a
n/a
n/a
n/a
44 min / 2640 sec
02/02 Saturday
02/03 Sunday
11:08
11:10
11:08
n/a
0
n/a
1, 2: No haa calls were observed on 9/15 or 9/16
3: Full deer carcass given to wolf group outside of observation area
4: Animal care emergency delayed feeding of wolves
5: Wolves given food-based enrichment
6: Full deer carcass given to wolf group outside of observation area
7, 8: Wolves fed on both Saturday and Sunday
* 12/29 and 12/30 recording skipped due to high wind warning
44
Appendix B
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
