EFFECTS OF GUEST FEEDING PROGRAMS
ON CAPTIVE GIRAFFE BEHAVIOR
By
David A. Orban

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Zoology – Master of Science
2013

ABSTRACT
EFFECTS OF GUEST FEEDING PROGRAMS
ON CAPTIVE GIRAFFE BEHAVIOR
By
David A. Orban
Zoological institutions develop human-animal interaction opportunities for visitors to
advance missions of conservation, education, and recreation; however, the animal welfare
implications have yet to be evaluated. This behavioral study was the first to quantify impacts of
guest feeding programs on captive giraffe behavior and welfare, by documenting giraffe time
budgets that included both normal and stereotypic behaviors. Thirty giraffes from nine zoos (six
zoos with varying guest feeding programs and three without) were observed for three days each,
using both instantaneous scan sampling and continuous behavioral sampling techniques. All data
were collected during summer 2012 and analyzed using generalized linear mixed models. The
degree of individual giraffe participation in guest feeding programs was positively correlated
with increased time spent idle and marginally correlated to reduced time spent ruminating. When
time spent eating routine diets was combined with time spent participating in guest feeding
programs, individuals that spent more time engaged in total feeding behaviors performed less
oral stereotypic behavior such as object-licking and tongue-rolling. By extending foraging time
and complexity, guest feeding programs have the potential to act as environmental enrichment
and alleviate unfulfilled foraging motivations that may underlie oral stereotypic behaviors
observed in many captive giraffes. Additionally, management strategies can be adjusted to
mitigate idleness and other program consequences. Further studies, especially pre-and-post
comparisons, are needed to better understand the influence of human-animal interactions on zoo
animal behavior and welfare.

ACKNOWLEDGMENTS

Funding for this project was provided by the Michigan State University Department of
Zoology through its John R. Shaver Research Fellowships for Graduate Students in Zoology.
This research could not have been possible without the voluntary cooperation of each of
the nine participating zoos: Binder Park Zoo, Brookfield Zoo, Cincinnati Zoo & Botanical
Garden, Cleveland MetroParks Zoo, Detroit Zoo, Fort Wayne’s Children Zoo, Indianapolis Zoo,
Lincoln Park Zoo, and Pittsburgh Zoo & PPG Aquarium. The many zoological managers,
keepers, docents and other staff were great in familiarizing me to their animals and exhibits and
providing insight to their animals’ history and behavior. Not only were they knowledgeable and
helpful, they were incredibly welcoming during my visit to their institution. I am incredibly
grateful for their interest in me and this project.
I would like to acknowledge my graduate advisory committee: Dr. Richard Snider,
Dr. Janice Siegford, Greg Geise, and Scott Carter. Their guidance and support was instrumental
to this project’s design, implementation and ultimate success. “Doc” and Janice, in particular,
have mentored me through most of my undergraduate and graduate education, so I am especially
grateful for their constant support and motivation. With regard to the writing of my thesis, I need
to particularly thank Dr. Renate Snider; I know this thesis would not be what it turned out to be
without her helpful guidance and cheerful spirit. Special appreciation goes to my great friends,
Dr. Jeanette McGuire and Melissa Elischer (my thesis buddy), for their helpful reviews and
motivational “pushes” (and for Jeanette, her incredible assistance with statistical coding and data
analysis). Members of the MSU Animal Behavior and Welfare Group have also been a great
resource for guidance, inspiration, and enthusiasm throughout this whole process.

iii

I would like to finally thank my family and friends that kept me grounded during my
graduate school career. Special mentions go to my amazing sister, Amanda, my supportive
parents, Mark and Cheryl, my wonderful grandparents, Dan and Henrietta, my cousins, Kyle,
Carolyn, Kristin, Steven, Hanna, and Lance, and my best friends/roommates, Jen, Adam, and
Sarah. I also need to recognize the nation of Uganda and the friends I’ve made there during
multiple visits with MSU Study Abroad; Africa may not always give you what you want, but it
gives you what you need. To the countless other graduate students, undergraduate students,
faculty, staff, and zoo personnel that I’ve interacted with during my time at Michigan State
University, it has been an absolute privilege to work with, learn from, teach to, and befriend
you. Life as a Spartan has granted me countless memories, each of which I will treasure for years
to come.

iv

TABLE OF CONTENTS
LIST OF TABLES ....................................................................................................................... vi
LIST OF FIGURES .................................................................................................................... vii
KEY TO SYMBOLS OR ABBREVIATIONS ........................................................................ viii
INTRODUCTION..........................................................................................................................1
METHODS .....................................................................................................................................6
Behavioral Observations ......................................................................................................6
Management Survey ............................................................................................................9
Analysis................................................................................................................................9
Project Approval ................................................................................................................11
RESULTS .....................................................................................................................................12
DISCUSSION ...............................................................................................................................22
Effects of GFPs on Non-stereotypic Behaviors .................................................................22
Effects of GFPs on Stereotypic Behaviors.........................................................................24
Demographic Differences ..................................................................................................28
Study Limitations ...............................................................................................................30
Management Recommendations ........................................................................................31
Future Studies ....................................................................................................................33
CONCLUSION ............................................................................................................................35
APPENDICES ..............................................................................................................................36
Appendix A: Time Budget Data Sheet ..............................................................................37
Appendix B: Stereotypic Behaviors Data Sheet ................................................................38
Appendix C: Management Survey .....................................................................................39
LITERATURE CITED ...............................................................................................................45

v

LIST OF TABLES

Table 1.

Giraffe populations observed at nine zoos .....................................................................6

Table 2.

Ethogram of giraffe behaviors used to collect live observational data ..........................7

Table 3.

Sampling schedule for giraffe observations at all zoos (except Zoo B and Zoo
I where all times were shifted later by one hour ............................................................8

Table 4.

“Zoo” effect on giraffe behavior observed in nine zoos (n = 30 giraffes) ...................13

Table 5a. Average time budget (for total observation period) for giraffe sample
population (n = 9) at three zoos without GFPs. Total time exceeds 100%
because not all behaviors were mutually exclusive .....................................................14
Table 5b. Average time budget (for total observation period) for giraffe sample
population (n = 21) at six zoos with GFPs. Total time exceeds 100% because
not all behaviors were mutually exclusive ...................................................................14
Table 6.

Mixed model analysis results of feeding effects on giraffe behavior. Zoo was
the random effect and times spent feeding in various formats were fixed
effects in separately run models. Giraffe behaviors were the dependent
variables. Data from nine zoos combined (n = 30 giraffes) .........................................16

Table 7.

Frequency and duration of giraffe stereotypic behavior bouts observed during
the total observation period using continuous behavioral sampling. For
duration, only stereotypic behavior bouts that began and ended within a 30min sample session were analyzed. The number of giraffes from nine zoos
that performed the specific behavior is denoted by “n” ...............................................18

Table 8.

Mixed model analysis results of demographic effects on giraffe behavior. Zoo
was the random effect and each demographic was a fixed effect in separately
run models. Giraffe behaviors were the dependent variables. Data from nine
zoos combined (n = 30 giraffes) ..................................................................................20

vi

LIST OF FIGURES

Figure 1a. Average time budget for giraffe sample population (n = 9) at three zoos
without GFPs ...............................................................................................................14
Figure 1b. Average time budget for giraffe sample population (n = 21) at six zoos with
GFPs .............................................................................................................................14
Figure 2. Correlation (P = 0.007) between time spent idle and time spent participating
in guest feeding programs in giraffes (n = 30) from nine zoos. Horizontal line
indicates average for giraffes at zoos without guest feeding programs .......................17
Figure 3. Correlation (P = 0.09) between time spent ruminating and time spent
participating in guest feeding programs in giraffes (n = 30) from nine zoos.
Horizontal line indicates average for giraffes at zoos without guest feeding
programs ......................................................................................................................17
Figure 4. Correlation (P = 0.503) between time spent performing oral stereotypies and
time spent participating in guest feeding programs in giraffes (n = 30) from
nine zoos. Horizontal line indicates average performance of stereotypic
behavior by giraffes at zoos without guest feeding programs .....................................19
Figure 5. Correlation (P = 0.005) between time spent performing oral stereotypies and
combined time spent eating (routine diet) and participating in guest feeding
programs in giraffes (n = 30) from nine zoos. Horizontal line indicates
average performance of stereotypic behavior by giraffes at zoos without guest
feeding programs .........................................................................................................19
Figure 6. Correlation (P = 0.017) between time spent pacing and age in giraffes (n = 30)
from nine zoos..............................................................................................................21

vii

KEY TO SYMBOLS OR ABBREVIATIONS
AZA – Association of Zoos & Aquariums
ANOVA – Analysis of Variance
GFP  Guest Feeding Program
GLMM  Generalized Linear Mixed Model

viii

INTRODUCTION
The mission of many zoological institutions includes the advancement of conservation,
education, and recreation (Anderson et al., 2003; Fernandez et al., 2009; Kreger and Mench,
1995; Patrick et al., 2007; Reade and Waran, 1996). To accomplish these goals, many zoological
institutions (hereafter referred to as zoos) have developed human-animal interaction
opportunities for visitors to directly engage with wildlife. Although some educational benefits of
human-animal interactions have been documented, these interactions have yet to be evaluated
from the animal welfare perspective.
Animals that express natural behaviors or are in close proximity to visitors are more
interesting than inactive or hidden animals (Altman, 1998; Bitgood et al., 1988; Margulis et al.,
2003) and enhance the interpretive and entertainment experience (Anderson et al., 2003;
Swanagan, 2000; Woods, 2002). Therefore, an increasing number of zoos offer visitors an
opportunity to directly interact with selected species as a way of complementing educational
messages. Interaction with animals is the most direct form of engagement, because visitors
perceive animals through their own senses and can establish an emotional connection
(Beardsworth and Bryman, 2001). These interactions occur in a variety of formats such as
petting zoos, touch tanks, public feedings, animal rides, and live entertainment shows (Kreger
and Mench, 1995). Physical contact with live animals, combined with educational messaging,
has been found to improve cognitive learning and attitudinal empathy (Kidd et al., 1995; Morgan
and Gramann, 1989; Sherwood et al., 1989); therefore, zoos have the potential to significantly
advance their conservation education mission by providing human-animal interactions (Kreger
and Mench, 1995).
Now recognized by most zoos as a critical objective and responsibility, animal welfare
has been incorporated into the accreditation standards of the Association of Zoos & Aquariums
1

(AZA). The AZA defines animal welfare as “an animal’s collective physical, mental, and
emotional states over a period of time, and is measured on a continuum from good to poor”
(AZA Animal Welfare Committee, 2013). The study of animal welfare is relatively new in the
zoo and aquarium industry, with comprehensive information still lacking for many exotic species
(Melfi, 2009; Wickens-Dražilová, 2006). Current welfare studies have primarily focused on
animal husbandry and environments (Melfi, 2009) with substantially less attention given to
welfare impacts of human-animal interactions within a zoo setting.
Zoo visitors can have three potential effects on animals: 1) stress, 2) enrichment, or 3)
none or neutral (reviewed by Hosey, 2000). Various species in zoos respond negatively to the
presence of guests (evidenced behaviorally and physiologically), with responses influenced by
visitor behavior, visitor group size, animal personality, and exhibit design (reviewed by: Davey,
2007; Fernandez et al., 2009; Hosey, 2000, 2008). For example, clouded leopards (Neofelis
nebulosa) placed on visitor display expressed higher levels of fecal corticoids than conspecifics
held off-exhibit (Wielebnowski et al., 2002). Other studies suggest that the experience is
enriching for other zoo animals, as evidenced by reciprocal behaviors or a desire for interaction
(reviewed by Claxton, 2011). Chimpanzees (Pan troglodytes) at Chester Zoo, for instance, were
found to elicit begging behavior and change locations within the exhibit to gain attention from
zoo visitors and solicit food (Cook and Hosey, 1995), while a Long-billed Corella (Cacatua
tenuirostris) spent more time at the front of its cage performing active behaviors when visitors
were nearby at Adelaide Zoo (Nimon and Dalziel, 1992). Additional research is needed to clarify
the true welfare implications of visitor effects (Davey, 2007), while also explicitly considering
species and individual variation.

2

According to the AZA, a program animal is described as “an animal whose role includes
handling and/or training by staff or volunteers for interaction with the public and in support of
institutional education and conservation goals” (AZA Program Animal Policy, 2011). Beyond
what is experienced by a normal exhibit animal, program animals are presented with novel
stimuli through changing environments, dynamic routines, and interactions with unfamiliar
people. The consequences of these specific interactive programs to animal welfare are relatively
unknown at this time, but are important to consider before continued emulation.
Charismatic megavertebrates such as the giraffe (Giraffa camelopardalis) are prime
candidates for human-animal interaction programs in zoos, largely due to their popularity and
prevalence. Additionally, their physical and behavioral attributes make them a relatively safe
species for guests to interact with. Over the last decade, many zoos have developed guest feeding
programs (GFPs) that offer visitors the opportunity to feed giraffes from platforms adjacent to
exhibits. An estimated 57% of AZA-accredited zoos exhibiting giraffes currently offer GFPs to
zoo guests (determined from zoos’ websites), but programs vary in schedule, platform structure,
food types, and regulations (Burgess, 2004). As more GFPs are developed, it becomes essential
to document their effects on giraffe behavior and welfare to not only justify their
implementation, but to also identify what modifications to programs may be necessary to
maximize benefits or minimize harm to giraffe welfare.
Wild giraffes spend about 70% of their day engaged in feeding and ruminating activity
(Ginnett and Demment, 1997; Leuthold and Leuthold, 1978a; Pellew, 1984), yet in captivity
these behaviors may comprise no more than half of the giraffe daily time budget (Koene, 1999,
Veasey et al., 1996). Captive giraffes also have less complex foraging patterns because, unlike in
the wild, diets do not contain ants or thorns that may require extensive oral manipulation (Koene

3

and Visser, 1996). Simple diets and a relatively large amount of unoccupied time may cause
distress for captive giraffes, leading them to develop replacement activities in the form of
stereotypies (Baxter and Plowman, 2001; Dittrich, 1976; Koene, 1999; Mason, 1991a; Mason
and Rushen, 2006).
Stereotypic behaviors are repetitive behavioral patterns that remain invariant and serve no
recognizable goal or function (Mason, 1991a). Their presence is often considered an indication
that the animal may not be appropriately adapting to its captive environment and is experiencing
suboptimal welfare (Broom, 1991a; Lawrence and Rushen, 1993; reviewed by Mason, 1991a;
reviewed in Mason and Latham, 2004). Captive giraffes perform multiple oral stereotypic
behaviors such as excessive object-licking and tongue-rolling, and locomotor stereotypic
behaviors such as pacing (Bashaw et al., 2001). A survey of AZA-accredited institutions found
that 79.7% of giraffes and okapis in the population performed stereotypies, 72.4% of which were
object-licking (Bashaw et al., 2001). In that study, giraffe subspecies, diet, type of feeder, and
feeding schedule were associated with stereotypic licking behavior, while giraffe subspecies,
diet, exhibit size, and environmental change were associated with the presence of stereotypic
pacing.
There is no fool-proof method for reducing prevalence of stereotypic behavior in giraffes,
and various attempts by zoos only hold a success rate of 51.9% (Bashaw et al., 2001). Claxton
(2011) argues that positive human-animal interactions could provide some necessary
environmental enrichment to decrease stereotypic behavior. Limited evidence shows that giraffes
fed by zoo visitors tend to express less stereotypic licking, however these conclusions were
drawn without quantification of behaviors (Bashaw et al. 2001). Other suggestions for decreasing
stereotypic behaviors in captive animals are to increase foraging time through environmental

4

enrichment programs (Newberry, 1995; Swaisgood and Shepherdson, 2006; Young, 1997) and
increase feeding duration by increasing food quantity, processing, or dispersal over larger areas
(Tarou et al., 2003). The implementation of GFPs may effectively encompass all of these
strategies and therefore contribute to creating a positive welfare state for giraffes.
Despite the critical gap in understanding effects of human-animal interactions on speciesspecific animal welfare, zoos have moved forward with implementation of guest feeding
programs. The primary goal of this behavioral observation study was to assess the impacts of
GFPs on captive giraffe behavior, and thus implications for welfare. The objective was to
quantify how daily time budgets of captive giraffes changed according to extent of individual
participation in a GFP, with emphasis on changes to stereotypic behaviors. We expected that an
increased time spent participating in GFPs by an individual giraffe would positively correlate to
increased active behaviors such as locomotion, rumination and social behavior, and negatively
correlate to prevalence of stereotypic behavior in that individual.

5

METHODS
Behavioral Observations
From June 22 to August 3, 2012, observations were conducted at nine Midwestern US,
AZA-accredited zoos (coded A through I in Table 1). Group size and animal demographics are
detailed in Table 1. At each zoo, behavioral data were collected for 3 days from 09:00 to 17:00,
with the exception of Zoo B and Zoo I where animals were observed from 10:00 to 18:00 to
comply with zoo operating hours. A single observer collected all data, with an intra-observer
reliability score of 87.24% as determined by comparison of 3 hours of live and video recordings.
The observer was positioned on a feeding platform or visitor pathway to blend in with visitors.
Giraffes were only observed while on public display.

Table 1. Giraffe populations observed at nine zoos.
Zoo

GFP
Number of
Sex
Presence* Individuals (male.female)
all-day
A
4
1.3
none
B
5
1.4
part-day
C
2
1.1
part-day
D
5
1.4
part-day
E
2
1.1
all-day
F
5
0.5
part-day
G
3
1.2
none
H
2
0.2
none
I
2
1.1
* Ordinal variables defined in Analysis section

Age Average
(years: mean ± SEM)
9.1 ± 3.4
11.9 ± 4.0
4.1 ± 0.0
8.8 ± 4.7
3.9 ± 0.8
11.5 ± 4.9
10.0 ± 4.4
18.6 ± 4.6
5.3 ± 0.6

Age Range
(years: min.  max.)
3.0  15.1
3.7  22.2
4.1  4.1
1.0  27.4
3.1  4.6
2.1  23.6
1.3  15.0
14.0  23.1
4.7  5.9

An ethogram (Table 2) was constructed by combining information obtained through
preliminary observations and published ethograms for giraffes (del Castillo et al., 2005; Ginnett
and Demment, 1997; Hosie and Turner, 2000; Kinahan and Marples, 2002; Seeber et al., 2012;

6

Veasey et al., 2006). Not all behaviors were mutually exclusive; for instance, rumination and
locomotion could occur simultaneously.

Table 2. Ethogram of giraffe behaviors used to collect live observational data.
Behavior
Idle:
Locomoting:
Eating:
Ruminating:

Drinking:
Socializing:
Grooming:
Objectmanipulating:
Object-licking:
Tongue-rolling:

Aberrant oral
behavior:
Pacing:
Time out:

Description
stationary position with no oral or locomotive activity, typically standing or
lying down (includes urination and defecation)
walking or running
tongue/mouth is in contact with non-cud food items for purpose of
consumption, includes mastication of non-cud
lower jaw moves horizontally to upper jaw and back to starting position in
succession of five or more occurrences resulting in the mastication of cud.
Also includes regurgitation and swallowing of cud
tongue/mouth is in contact with water
physical contact with conspecific, includes necking, allogrooming and
flehmen activities
rubbing of body against stationary object (e.g., wall, tree) or scraping of
teeth/tongue across own body
tongue/mouth/head is in physical contact with non-food item, typically
movable enrichment items
tongue is repetitively moved across non-food, stationary item, typically a
structural component of exhibit such as wall, fence, or tree
tongue is continuously moved (or “rolled”) in and out of mouth, most often
involving a food item that is not actively involved in mastication or
rumination
teeth or lips are repetitively moved across non-food, stationary item,
typically a structural component of exhibit such as wall, fence, or tree
locomotion occurring in a repeated pattern between two locations
animal or animal’s head is hidden from observer

One stereotypic behavior not identified previously, but recorded in this study, is “aberrant
oral behavior.” This was a catch-all type behavioral category that was neither tongue-playing nor
object-licking but still encompassed the repetitive, invariant nature of a stereotypic behavior and
seemed to have no apparent goal or function (Mason, 1991a). Typically, aberrant oral behavior
manifested as a grating of teeth or lips on some object such as rope fencing or a wooden wall.

7

Time budgets were calculated from data collected during 30-minute sessions of
instantaneous scan sampling (one scan per minute; Altmann, 1974; Martin and Bateson, 2007)
that took place over 3 days at each institution. Based on preliminary observations, 1 minute was
sufficient for scanning up to six giraffes in the presence of maximum guest crowds. Five to six
sampling sessions took place per day, staggered throughout the 3 recording days (Table 3). The
time budget data sheet allowed recording of information from multiple giraffes simultaneously
(Appendix A).
In addition to instantaneous scans, the frequency and duration of stereotypies were
recorded by continuous behavioral sampling (Altmann, 1974; Martin and Bateson, 2007). Each
giraffe group was continuously observed for 30 minutes, and the start and end times of
stereotypic behaviors of individual giraffes were logged. A stereotypy was only recorded if its
duration exceeded 5 seconds. Five to six sessions of continuous behavioral sampling took place
per day, staggered throughout the 3 recording days (Table 3). Sessions were staggered to provide
ample time for data collection by both sampling methods, to account for circadian rhythms of
giraffes, and to record all management and GFP events. The stereotypic behavior data sheet
allowed recording of multiple giraffes simultaneously (Appendix B).

Table 3. Sampling schedule for giraffe observations at all zoos (except Zoo B and Zoo I
where all times were shifted later by one hour).
Day Type of Sampling
Scan (Time Budget)
1
Behavioral (Stereotypies)
Scan (Time Budget)
2
Behavioral (Stereotypies)
Scan (Time Budget)
3
Behavioral (Stereotypies)

Start Time
09:00
10:30
09:30
11:00
09:30
11:00
10:00
11:30
10:00
11:30
09:00
10:30

8

12:00
12:30
12:30
13:00
13:00
12:00

13:30
14:00
14:00
14:30
14:30
13:30

15:00
15:30
15:30
16:00
16:00
15:00

16:30
16:30

Management Survey
To document differences among zoos, qualitative data were collected from giraffe care
staff (keeper or curator) via a structured interview (Appendix C). Survey variables included
environment (size, features, proximity to humans and other species), animal care (diet, feeding
schedule, training, and enrichment), and animal demographics (sex, age, rearing, and health
status). Details were also collected regarding various aspects of the GFPs including platform
size, food type, and food price.

Analysis
Data collected from 30 giraffes were used for analysis. A total of 32 giraffes were
observed throughout the study period, but two giraffes were excluded from analysis because of
their age and physical limitations. Two behaviors, drinking and object-manipulating, were also
excluded due to exhibit characteristics that prevented consistent, accurate sampling. One
individual in the dataset was an outlier in its performance of aberrant oral behavior, as
determined through a Grubbs test. The individual was a Reticulated/Masai/Rothschild’s hybrid,
separated early from its mother, aggressive to both keepers and conspecifics, and performed a
head-rolling movement while at the feeding platform. Since we do not fully understand how
these variables influence giraffe behavior, values with and without this outlying individual are
presented where appropriate.
Generalized linear mixed models (GLMM) were used for analysis because data were
non-independent (clustered) and unbalanced. Preliminary analysis tested influence of
demographic and environmental variables on behavior, and the results were used to select
random and fixed effects for the model. Specifically, analysis of variance (ANOVA) tests were

9

conducted to determine whether giraffe behaviors differed significantly between zoos. Each zoo
housed a separate cluster of giraffes and was therefore used as a random effect in the model. No
other random effects were necessary because “zoo” encompassed all other significant
environmental and animal care factors (e.g., exhibit size, indoor housing size, and browse
availability).
Behavioral data collected on each giraffe from the 3 observation days were compiled.
Time spent participating in a GFP was converted to a percentage of total observation time to
facilitate comparisons among giraffes with different total sampling times. Because not all zoos
had GFPs, and because some giraffes did not participate when GFPs were available, time budget
behavioral data were standardized across all zoos (those with GFPs and those without GFPs) by
converting time spent engaged in behaviors to percentages of total observation time with the time
spent participating in GFPs removed. Percentage of time spent participating in a GFP was used
as the fixed effect, and behaviors from the time budget, including stereotypic behaviors, were
analyzed separately as dependent variables. All percentages for all models were arcsine
transformed prior to data analysis.
Sex (ordinal variable) and age (continuous variable) were separately entered into
generalized linear mixed models as fixed effects because preliminary analysis showed that they
independently influenced behavior. For these two models, “zoo” was the only random effect and
behaviors were analyzed separately as the dependent variables.
One additional set of GLMM analyses was conducted to assess how duration of
availability of GFPs, regardless of individual participation, affected giraffe behavior. Zoos were
grouped by an ordinal variable based on GFP duration: “no GFP”, “part-day GFP”, and “all-day
GFP” (Table 1). The designation of “all-day GFP” was indicative of a GFP being open during all

10

zoo operating hours, whereas a “part-day GFP” was a more limited program that only remained
open for certain segments of the day, which in these cases extended no more than 2 hrs total. In
this model, “zoo” was the only random effect, duration of GFPs was the fixed effect, and
behaviors were analyzed separately as the dependent variables.
All statistical analyses were performed using the statistical software R v. 2.13.2011 (R
Core Development Team, 2011), chiefly the “lme4” package for generalized linear mixed
models (Bates et al., 2011). The outlier test was performed using the “outliers” package (Komsta,
2011).

Project Approval
Observational data were collected without manipulation of animals, humans, or
environments occurring. Therefore, this study received an exemption from filing an Animal Use
Form with the Michigan State University Institutional Animal Care and Use Committee or
requiring human subject research approval from the Michigan State University Institutional
Review Board. All zoos voluntarily participated in this study.

11

RESULTS
Results of the two sampling methods used to measure stereotypic behavior were highly
correlated, indicating that despite the presumed “event” status (Altmann, 1974; Martin and
Bateson, 2007) of stereotypic behavior, scan sampling was adequate for capturing the amount of
time spent performing these behaviors. Observations between instantaneous scan sampling and
continuous behavioral sampling were highly correlated for object-licking (Pearson’s r = 0.957, P
< 0.001), tongue-rolling (Pearson’s r = 0.928, P < 0.001), and pacing (Pearson’s r = 0.799, P <
0.001). A single outlier was apparent in the aberrant oral behavior category (Grubbs Test G =
5.166, P < 0.001); after removing this outlier, observations of this behavior were also highly
correlated (Pearson’s r = 0.927, P < 0.001). Because results derived from the two sampling
techniques were correlated, data from only one method were analyzed further. The instantaneous
scan sampling dataset was chosen because other behaviors used to calculate time budgets, as
well as stereotypic behaviors, were included.
Several giraffe behaviors significantly differed between zoos including idleness,
rumination, social, total oral stereotypies, and pacing (ANOVA, Table 4). A trend was observed
for time spent eating, and no effect was found on locomotion and grooming behaviors. Because
multiple behaviors were influenced by the zoo the giraffes resided in, zoo was included as a
random effect in subsequent generalized linear mixed model analyses.

12

Table 4. “Zoo” effect on giraffe behavior observed in nine zoos (n = 30 giraffes).
F8,20
Behavior
P-value
Idleness
4.441
0.003***
Locomotion
1.822
0.129
Eating
2.118
0.081*
Rumination
5.316
0.001***
Social
2.523
0.043**
Grooming
0.967
0.487
Total Oral Stereotypies
3.053
0.019**
Pacing
4.945
0.002***
* Indicates trend (0.1 > P ≥ 0.05), ** Indicates significance (0.05 > P ≥ 0.01),
*** Indicates high significance (P < 0.01)
The average time budget for giraffes at zoos without GFPs was comprised mostly of
idleness and eating, but rumination, locomotion, and oral stereotypies also occupied a large
portion of the observation period (Table 5a, Fig. 1a). At zoos with GFPs, the average time budget
for giraffes was similarly distributed (Table 5b, Fig. 1b).

13

Table 5a. Average time budget (for total observation
period) for giraffe sample population (n = 9) at three zoos
without GFPs. Total time exceeds 100% because not all
behaviors were mutually exclusive.
Behaviors
Idleness
Locomotion
Eating
Rumination
Social
Grooming
Oral stereo.
Pace stereo.

% of Observation Period ± SEM
Total (n=9) Male (n=2) Female (n=7)
29.6 ± 3.6
30.1 ± 1.7
29.4 ± 4.7
11.5 ± 1.3
16.7 ± 1.1
10.0 ± 1.1
21.3 ± 3.4
33.1 ± 1.0
17.9 ± 3.4
18.4 ± 3.0
12.0 ± 5.8
20.2 ± 3.4
1.4 ± 0.6
3.0 ± 2.3
1.0 ± 0.4
0.8 ± 0.2
0.8 ± 0.1
0.8 ± 0.3
17.0 ± 4.4
5.5 ± 5.0
20.2 ± 4.9
1.3 ± 0.7
0.0 ± 0
1.6 ± 0.9

Oral
stereotypy
17.0%
Grooming
0.8%

Behaviors
Idleness
Locomotion
Eating
Rumination
Social
Grooming
Oral stereo.
Pace stereo.
Part. in GFP

% of Observation Period ± SEM
Total (n=21) Male (n=5) Female (n=16)
35.4 ± 3.8
28.9 ± 1.6
37.4 ± 4.9
11.6 ± 1.2
13.6 ± 2.7
10.9 ± 1.3
19.6 ± 1.8
27.6 ± 4.0
17.2 ± 1.6
11.9 ± 1.7
15.9 ± 3.2
10.6 ± 2.0
0.5 ± 0.2
0.5 ± 0.3
0.5 ± 0.2
1.3 ± 0.4
2.6 ± 1.2
0.9 ± 0.4
12.5 ± 0.4
3.7 ± 2.2
15.2 ± 4.8
1.4 ± 0.5
1.8 ± 1.4
1.2 ± 0.6
7.8 ± 1.9
6.6 ± 1.8
8.2 ± 2.4

Pacing
stereotypy
Oral
1.4%
stereotypy
12.5%
Grooming
1.3%

Pacing
stereotypy
1.3%
Idleness
29.6%

Social
1.4%

Rumination
18.4%

Table 5b. Average time budget (for total observation
period) for giraffe sample population (n = 21) at six zoos
with GFPs. Total time exceeds 100% because not all
behaviors were mutually exclusive.

Social
0.5%

Eating
21.3%

Locomotion
11.5%

Rumination
11.9%

Part. in GFP
7.8%

Idleness
35.4%

Eating
19.6%

Locomotion
11.6%

Figure 1b. Average time budget for giraffe sample
population (n = 21) at six zoos with GFPs.

Figure 1a. Average time budget for giraffe sample
population (n = 9) at three zoos without GFPs.
14

The amount of time spent participating in GFPs influenced rates of idleness and
rumination (GLMM, Table 6). Idleness significantly increased and rumination tended to decrease
as a function of participation in GFPs (Fig. 2, Fig. 3). Additional analysis of these two behaviors
showed that both behaviors differed based on type of GFP. Idleness significantly increased for
all-day feeding programs (GLMM, t = 3.303, P = 0.003) but was not influenced by part-day
feeding programs (GLMM, t = 0.239, P = 0.813). Similarly, rumination significantly decreased
for giraffes in all-day feeding programs (GLMM, t = 2.206, P = 0.036) but was not influenced by
part-day feeding programs (GLMM, t = 0.092, P = 0.927). Locomotion, eating (routine diet),
social and grooming behaviors were not affected by time spent participating in GFPs.
In contrast to the positive correlation between time spent participating in GFPs and
idleness, there was a negative correlation between time spent eating (routine diet) and idleness
(GLMM, Table 6). The combination of time spent eating (routine diet) and participating in GFPs
tended to be positively correlated with increased time spent locomoting, and had no effect on
idleness or rumination (GLMM, Table 6). Social and grooming behaviors were not influenced by
total time spent feeding.

15

Table 6. Mixed model analysis results of feeding effects on giraffe behavior. Zoo was the
random effect and times spent feeding in various formats were fixed effects in separately
run models. Giraffe behaviors were the dependent variables. Data from nine zoos
combined (n = 30 giraffes).
Fixed Effect

Dependent Variable (Behavior)
t-value
Idleness
2.902
Locomotion
1.067
Eating (routine diet)
-1.272
Rumination
-1.756
Social
-0.845
Grooming
-1.678
Participation in GFP
Tongue-rolling
-1.561
Object-licking
-0.735
Aberrant oral behavior (outlier removed) -1.387
Aberrant oral behavior (with outlier)
2.755
Pacing
-0.764
Total oral stereotypy (outlier removed)
-0.981
Total oral stereotypy (with outlier)
-0.678
Idleness
-2.427
Eating (routine diet)
Total oral stereotypy
-2.816
Idleness
-0.203
Locomotion
1.817
Rumination
0.137
Social
1.207
Grooming
-0.698
Total feeding (routine
Tongue-rolling
-2.667
diet + participation in
Object-licking
-1.958
GFP)
Aberrant oral behavior (outlier removed) -0.804
Aberrant oral behavior (with outlier)
1.735
Pacing
-0.113
Total oral stereotypy (outlier removed)
-3.228
Total oral stereotypy (with outlier)
-3.043
* Indicates trend (0.1 > P ≥ 0.05), ** Indicates significance (0.05 > P ≥ 0.01),
*** Indicates high significance (P < 0.01)

16

P-value
0.007***
0.295
0.214
0.090*
0.405
0.104
0.130
0.468
0.177
0.010**
0.451
0.335
0.503
0.022**
0.009***
0.841
0.080*
0.892
0.237
0.491
0.013**
0.060*
0.429
0.094*
0.911
0.003***
0.005***

Idleness
(% of Observation Period)

100
90
80
70
60
50
40
30
20
10
0
0

5

10

15

20

25

30

35

Participation in Guest Feeding Programs
(% of Observation Period)
Figure 2. Correlation (P = 0.007) between time spent idle and time spent participating in
guest feeding programs in giraffes (n = 30) from nine zoos. Horizontal line indicates
average for giraffes at zoos without guest feeding programs.
35

Rumination
(% of Observation Period)

30
25
20
15
10
5
0
0

5

10

15

20

25

30

35

Participation in Guest Feeding Programs
(% of Observation Period)
Figure 3. Correlation (P = 0.09) between time spent ruminating and time spent
participating in guest feeding programs in giraffes (n = 30) from nine zoos. Horizontal line
indicates average for giraffes at zoos without guest feeding programs.

17

Twenty-eight of the 30 giraffes expressed one or more stereotypic behavior (19
performed tongue-rolling, 24 performed object-licking, 8 performed aberrant oral behavior, and
10 performed pacing). Performance of stereotypies varied in frequency and duration between
giraffes (Table 7).
Table 7. Frequency and duration of giraffe stereotypic behavior bouts observed during the
total observation period using continuous behavioral sampling. For duration, only
stereotypic behavior bouts that began and ended within a 30-min sample session were
analyzed. The number of giraffes from nine zoos that performed the specific behavior is
denoted by “n”.

Stereotypic Behavior
Tongue-rolling (n = 16)
Object-licking (n = 20)
Aberrant oral behavior (n = 6)
Pacing (n = 10)

Frequency
Average ±
Min.
SEM
17.5 ± 4.9
1
15.5 ± 3.1
1
8.7 ± 3.3
1
6.3 ± 1.0
1

Max.
59
45
23
11

Duration (mm:ss)
Average ±
Min. Max.
SEM
03:43 ± 02:06 00:07 16:46
04:35 ± 01:31 00:06 21:30
00:49 ± 00:47 00:05 05:38
02:07 ± 00:22 00:13 16:29

Time spent participating in GFPs had no effect on time spent engaged in stereotypies:
tongue-rolling, object-licking, aberrant oral behavior, or pacing (GLMM, Table 6). When all oral
stereotypies were combined, analysis continued to show no significant correlation with
participation in GFPs (Fig. 4, Table 6). Time spent eating (routine diet) was negatively correlated
to time spent engaged in total oral stereotypies (GLMM, Table 6). When time spent eating
(routine diets) and participating in GFPs were combined, total oral stereotypies significantly
decreased further (Fig. 5), and particularly significant negative correlations between total time
spent feeding and tongue-rolling and object-licking were observed (GLMM, Table 6).

18

Oral Stereotypies
(% of Observation Period)

60
50
40
30
20
10
0
0

5

10

15

20

25

30

35

Participation in Guest Feeding Programs
(% of Observation Period)
Figure 4. Correlation (P = 0.503) between time spent performing oral stereotypies and time
spent participating in guest feeding programs in giraffes (n = 30) from nine zoos.
Horizontal line indicates average performance of stereotypic behavior by giraffes at zoos
without guest feeding programs.

Oral Stereotypies
(% of Observation Period)

60
50
40
30
20
10
0
0

10

20

30

40

50

60

Total Feeding
(% of Observation Period)
Figure 5. Correlation (P = 0.005) between time spent performing oral stereotypies and
combined time spent eating (routine diet) and participating in guest feeding programs in
giraffes (n = 30) from nine zoos. Horizontal line indicates average performance of
stereotypic behavior by giraffes at zoos without guest feeding programs.
19

Sex and age were significantly correlated with several behaviors (GLMM, Table 8). Male
giraffes spent more time eating (routine diet) throughout the observation period than females.
There were also trends for increased locomotion and grooming among male giraffes. With regard
to stereotypic behaviors, female giraffes performed oral stereotypies more frequently than males.
Among older giraffes, idleness increased, while time spent engaged in locomotion, eating
(routine diet), and grooming decreased. Performance of oral stereotypies was not affected by age,
but pacing, the only locomotor stereotypy measured, was observed significantly more often in
younger giraffes (Fig. 6). Rumination and social behavior were affected by neither sex nor age.
Table 8. Mixed model analysis results of demographic effects on giraffe behavior. Zoo was
the random effect and each demographic was a fixed effect in separately run models.
Giraffe behaviors were the dependent variables. Data from nine zoos combined (n = 30
giraffes).
Fixed Effect

Dependent Variable (Behavior) t-value
P-value
Idleness
-0.347
0.731
Locomotion
1.766
0.088*
Eating (routine diet)
3.72
0.001***
Rumination
-0.163
0.872
Sex
Social
1.392
0.175
Grooming
1.987
0.057*
Total oral stereotypy
-2.868
0.008***
Pacing
0.233
0.818
Idleness
2.118
0.043**
Locomotion
-2.404
0.023**
Eating (routine diet)
-3.282
0.003***
Rumination
-0.247
0.807
Age
Social
-0.603
0.551
Grooming
-2.826
0.009***
Total oral stereotypy
1.533
0.137
Pacing
-2.549
0.017**
* Indicates trend (0.1 > P ≥ 0.05), ** Indicates significance (0.05 > P ≥ 0.01),
*** Indicates high significance (P < 0.01)

20

Pacing
(% of Observation Period)

8
7

6
5
4
3
2
1
0
0

5

10

15

20

25

30

Age (years)
Figure 6. Correlation (P = 0.017) between time spent pacing and age in giraffes (n = 30)
from nine zoos.

21

DISCUSSION
Human-animal interactions can influence the behavior and welfare of program animals in
zoological institutions and must be closely evaluated to ensure a positive experience for both
program animals and human participants. Welfare can be assessed through behavioral,
physiological, and health measures, and is unique to species, individuals, and contexts. Poor
welfare is associated with reduced fitness and is usually brought about by stress or an inability to
cope with environmental conditions (Broom, 1991b). In this research, stereotypies were used as
indicators of sub-optimal welfare, assuming that they arose from frustration of not being able to
appropriately perform a desired, natural behavior  foraging. A guest feeding program (GFP)
may help to restore some aspects of natural giraffe foraging by extending time spent feeding
throughout the day, increasing food acquisition complexity, expanding distributional range of
food, and providing additional food quantities. This was the first study to document
quantification of behavioral changes (stereotypic and non-stereotypic) as a function of individual
giraffe participation in GFPs, and results may indicate whether GFPs can be considered a form of
environmental enrichment for captive giraffes.

Effects of GFPs on Non-stereotypic Behaviors
Idleness comprised the largest portion of the average giraffe time budget in this study,
regardless of whether giraffes were housed at a GFP or non-GFP institution, and observed levels
of idleness were higher than those reported for wild giraffes (Pellew, 1984; Veasey et al., 1996).
Data indicated a strong positive correlation between idleness and degree of participation in
GFPs; however when analyzed categorically, time spent idle was only significantly higher in

22

cases of all-day feeding programs, not in cases of part-day feeding programs. Locomotion levels
were substantially lower than idleness and lower than those reported for wild giraffes (Pellew,
1984; Veasey et al., 1996). The elevated levels of idleness seem to be a consequence of open but
not active feeding programs (often the case in all-day feeding programs) in which giraffes waited
near platforms in anticipation of being fed. However, time spent engaged in GFPs may not
impact overall activity levels, if time spent eating routine diets is accounted for. When time spent
feeding from routine diets and GFPs were combined, total time spent feeding had no influence
on idleness and tended to correlate to increased locomotion. Therefore idleness levels may be
irrespective of total feeding activities, and may only potentially be problematic under all-day
GFP situations.
Giraffes that spent more time engaged in GFPs had lower levels of rumination, which
might be indicative of elevated stress because rumination is thought to transpire only when in a
relaxed state (Trunkfield and Broom, 1990). Rumination is a very important part of the digestive
process (Baxter and Plowman, 2001) and is the second most time-consuming activity in the daily
time budget of wild giraffes (Pellew, 1984). Captive giraffes have been shown to increase
rumination when fibrous food items (Baxter and Plowman, 2001) and an ad-libitum browse wildlike feeding regime (Veasey et al., 1996) were offered. Yet, giraffes exposed to a GFP did not
exhibit the anticipated higher levels of rumination. Because giraffes still need to ruminate, it is
plausible that rumination bouts may have shifted outside the observation period, specifically to
night-time, indoor holding. Rumination has previously been found to occur more often at night in
both captive and wild giraffes (Baxter and Plowman, 2001; Pellew, 1984). Furthermore, the time
spent ruminating was only significantly reduced in giraffes in all-day GFPs, signifying that
continuous feedings may have prevented initiation of rumination bouts (which were often
23

comprised of more than 20 continuous instantaneous scans in individuals with no or part-day
GFPs). Without assessing the 24-hr time budget (see Study Limitations), it is difficult to ascertain
whether time spent engaged in a GFP altered the total amount of rumination or simply the
timing.

Effects of GFPs on Stereotypic Behaviors
Stereotypies can be defined as repetitive behavioral patterns that remain invariant and
serve no recognizable goal or function (Mason, 1991a). Although the absence or reduction of
stereotypic behaviors does not necessarily guarantee good welfare (Mason and Latham, 2004;
Melfi, 2009), these behaviors are often used as indicators of suboptimal welfare (Broom, 1991a;
Lawrence and Rushen, 1993; reviewed by Mason, 1991a, Mason and Latham, 2004). Animal
welfare can only be improved by mitigating the motivation that initiates stereotypic behavior 
not merely reducing performance of the behavior (Mason and Latham, 2004).
Believed to stem from abnormal animal–environment interactions (Carlstead, 1998),
stereotypic behavior does not appear in the same context or frequency as it would in nature
(Fraser and Broom, 1990; Veasey et al., 1996). The behavior is typically elicited in captive
situations where an animal experiences stress, conflict, and/or frustration from a lack of
stimulation in or control over its environment (Mason, 1991a; Morgan and Tromberg, 2007). In
response to suboptimal conditions, stereotypies may act as a coping mechanism (Broom,
1991a&b; Fraser and Broom, 1990), allowing the animal to express a behavioral need, receive
some benefit (e.g. salivation), lower responsiveness (calming feeling), or establish predictability

24

in an unpredictable environment (Mason, 1991a). Thus through coping, stereotypic behavior may
work to alleviate stressful conditions and rectify individual welfare (Mason and Latham, 2004).
Many have hypothesized that oral stereotypies form in captivity as a result of frustrated
feeding (and ruminating) motivation that is unable to be fulfilled as it would in nature (reviewed
in Bergeron et al., 2006; Baxter and Plowman, 2001; Dittrich, 1976). Oral stereotypic behavior is
especially prevalent in ungulate species due to adaptations necessary for herbivory: extensive
amount of time needed for foraging (to attain bulk and nutritional gain), need to search for and
select specific nutrients, and ability to overcome physical and chemical plant defenses (Bergeron
et al., 2006). In captivity, ungulate diets are often concentrated, relatively homogenous, and
directly offered in a fixed delivery (time and location), and consequently reduce time spent
searching for food, consuming food, and (for ruminants) ruminating food (Baxter and Plowman,
2001; Bergeron et al., 2006; Dittrich, 1976; Kinahan and Marples, 2000; Koene, 1999; Young,
1997). As the largest ruminant in the world, feeding activity is a major component of a giraffe’s
biology, comprising a majority of its time budget (Ginnett and Demment, 1997; Leuthold and
Leuthold, 1978; Pellew, 1984); the amount of time captive giraffes dedicate to feeding and
rumination may only comprise roughly half of what is performed in the wild (Koene, 1999;
Veasey et al., 1996). Both quantity and quality of food in captivity may have significant impacts
on giraffe behavior, with time not dedicated to feeding leading to inactivity or a replacement
stereotypic behavior (Kinahan and Marples, 2000; Koene, 1999).
Captive ungulates exhibit oral stereotypies that often resemble normal mouth movements
made when feeding (Bergeron et al., 2006; Dittrich, 1976) and for giraffes, these include objectlicking, tongue-rolling/tongue-playing, and vacuum chewing (Baxter and Plowman, 2001;
Seeber et al., 2012). In a survey of the AZA giraffe and okapi population, 79.7% of giraffids
25

surveyed performed stereotypies with 72.4% of these being some sort of oral stereotypy (Bashaw
et al., 2001). In this study, 93.3% of giraffes surveyed performed oral stereotypies at some point
during the observation period; 33.3% performed pacing stereotypies. This higher occurrence
could indicate that managers were unaware and thus underreported the prevalence of stereotypic
behavior among their giraffes.
We expected GFPs to decrease prevalence of stereotypic behavior because program
qualities (i.e., extend time spent feeding, increase feeding complexity) would mimic natural
foraging characteristics and reduce frustration. Tarou et al. (2003) suggested that extending
foraging duration would ultimately fulfill the foraging motivation causing oral stereotypies, and
limited reports have shown that captive giraffes that spent more time feeding were observed to
perform less oral stereotypic behavior (Koene, 1999; Koene and Visser, 1996). Increasing
complexity of attaining food (through feeder design) can reduce oral stereotypy (Fernandez et
al., 2008; Bashaw et al., 2001), and adding extra portions of browse and fibrous food items to the
diet decreases object-licking and tongue-playing (Baxter and Plowman, 2001; Bashaw et al.,
2001; Kinahan and Marples, 2000; Koene, 1999; Koene and Visser, 1996). A survey study by
Bashaw et al. (2001) indicated a tendency for giraffes fed by zoo visitors to display less
stereotypic licking (although the study did not measure feeding duration or quantity of feed).
Individual time spent participating in GFPs had no significant effect on performance of
stereotypic behaviors. Although each of the four stereotypic behaviors decreased among giraffes
that increasingly participated in GFPs, these relationships were non-significant. However, the
difficulty in abolishing stereotypic behaviors once established must be taken into consideration
(Mason and Latham, 2004; Mason, 1991a). Furthermore, because stereotypies can become habits
over time regardless of stimuli, or be a result of central nervous system dysfunction, the
26

performance of stereotypic behavior may not be directly linked to current environmental
conditions and welfare (Mason and Latham, 2004; Mason 1991b). In zoo animals particularly,
stereotypic behaviors may be a result of past exhibits or experiences. For these reasons, guest
feeding programs may not be able to fully eliminate an engrained stereotypic behavior even if
enriching. Yet, since GFPs were not found to correlate to increased stereotypic behavior, we can
infer that this human-animal interaction is not an apparent cause of stress or frustration and does
not negatively contribute to giraffe welfare.
Time spent eating routine diets significantly correlated to reduced performance of oral
stereotypies, and this correlation was strengthened when combined with time spent participating
in GFPs. This relationship supports the hypothesis that increased time spent feeding reduces time
spent performing oral stereotypies (Fernandez et al., 2008; Koene, 1999; Tarou et al., 2003).
Guest feeding programs, in combination with increased feeding duration of routine diets, can
significantly reduce performance of oral stereotypic behavior, reinforcing the proposition that
these programs can serve as environmental enrichment for captive giraffes.
Time spent engaged in GFPs had no effect on performance of stereotypic pacing,
supporting the idea that locomotor stereotypies are more influenced by environmental variables
than by feeding variables (Bashaw et al., 2001). However, neither exhibit size nor indoor holding
size showed significant influence on time spent stereotypic pacing in this study. Pacing was most
often observed at the end of the observation period, directly prior to giraffes being allowed entry
into indoor holding for the night. This observation was similar to that made by Koene and Visser
(1996). Food was frequently available to giraffes upon return indoors, indicating that in this case,
food or routine moving may have contributed to the cause of anticipatory pacing. Another
explanation for the pacing behavior could be that, at two zoos, pacing individuals were separated
27

from a conspecific(s) kept indoors. It has been previously reported that females separated from a
male conspecific at Zoo Atlanta showed increases of object-licking and pacing stereotypies
(Tarou et al., 2000), and so observations from this study may further indicate that social
separation is a stressful experience that can influence prevalence of stereotypic behavior.

Demographic Differences
Giraffe age positively correlated to levels of idleness, likely because older individuals
were larger and could more readily participate in GFPs (previously shown to correlate to
idleness). Presumably, locomotion decreased in older giraffes for the same reason. Younger,
subordinate giraffes moved around the platform to find available, non-competitive space, or were
too small to participate and spent time moving around the rest of the exhibit. Locomotion
differences were also found based on sex, with males tending to locomote more than females,
which corroborates data from wild giraffes (Pellew, 1984).
In the wild, females typically spend more time feeding than males, about 50-70%
compared to about 25-50% of 24-hr activity, respectively (Ginnett and Demment, 1997;
Leuthold and Leuthold, 1978; Pellew, 1984), with differences attributed to reproductive costs
(Leuthold and Leuthold, 1978). In this study, females spent significantly less time eating than
males but this difference might be largely influenced by the significantly higher stereotypic
behavior levels exhibited by females which took up a larger portion of the time budget. Baxter
and Plowman (2001) also reported that two females spent less time eating and more time oral
stereotyping than their male companion. Older giraffes spent significantly less time eating

28

routine diets than younger giraffes, likely because older giraffes spent a larger portion of their
day idle and participating in GFPs.
Female giraffes in the study performed stereotypic behaviors significantly more often
than male giraffes. This observation was similar to those of Baxter and Plowman (2001), Hosie
and Turner (2000), and Fernandez et al. (2008), although these authors observed few individuals
and presented no reason behind the difference. Because female giraffes spend more time engaged
in feeding behaviors in the wild (Ginnett and Demment, 1997; Leuthold and Leuthold, 1978a
Pellew, 1984), and a strong unfulfilled motivation to feed is believed to induce oral stereotypic
behavior, we would expect higher levels of oral stereotypies among captive female giraffes.
In contrast to observations of Veasey et al. (1996), younger giraffes in the present study
paced more often than older giraffes. It has been suggested that stereotypies develop and become
more frequent as animals mature (reviewed in Mason, 1991a), so differences in pacing behavior
between young and older giraffes is difficult to explain. Pacing often preluded reuniting with
separated conspecifics in indoor housing, suggesting that younger individuals may experience
more stress from social separation than older individuals. There is recent evidence to suggest
giraffes do not randomly associate as previously believed (Estes, 1999; Leuthold, 1979) but form
social relationships, with social preference for individuals differing most in age (Bashaw et al.,
2007). Younger giraffes in this study may have preferred to associate with separated older
individuals, and prevention from doing so elicited frustration which caused pacing behavior.
Alternatively, feeding motivations may become more pronounced at different ages, and thus the
corresponding types of stereotypic behavior expressed in the time budget will shift accordingly
as giraffes age.

29

Study Limitations
The present study only analyzed the period of the day from 9:00 to 17:00, which leaves a
significant portion of the 24-hr time budget undocumented. Especially because stereotypic
behaviors have been recorded more at night and while in indoor housing (Bashaw et al., 2001;
Baxter and Plowman, 2001; Hosie and Turner, 2000), additional observations need to be made
during the remainder of the 24-hr day. Fernandez et al. (2008) found that changing the feeding
habits indoors affected stereotypic behavior expressed outdoors; the opposite is also possible.
Extending the observation period to 24 hrs would help to clarify effects on overall stereotypic
behavior, as well as rumination rate and timing.
Behavioral studies on animals within zoological settings present unique challenges. For
example, it is often difficult to attain sample sizes sufficient for meaningful statistical inferences
and also to attain balance with regard to sex and age in a sample population. The sample
population in this study consisted of 7 males and 23 females and had a median age of 5.7 years.
Although the composition of a giraffe herd within zoo exhibits is typically female-biased, the
sample population from this study may not adequately reflect the age structure of the whole
giraffe population. Giraffes in captivity can live into their mid- to late-20s, therefore caution
should used when drawing inferences from this study about age- or sex-based effects on giraffe
behavior.
Cross-institutional analysis can also present problems when assessing behavior, as
individual animals and their ability to express certain behaviors will likely differ depending on
exhibit and management characteristics (e.g., Carlstead et al., 1999; Koene and Visser, 1996;
Montaudouin and Pape, 2004; Perkins, 1992). In contrast to Veasey et al. (1996) and Bashaw

30

(2011), significant differences in multiple behaviors were found between zoos in this study. With
high variation in exhibit size (ranging from 0.3 acres to 16 acres) and complexity (i.e., provision
of foliage, browse, and enrichment), social composition (intra- and inter-specific), GFP
characteristics, and other factors, this was expected. Because animal welfare is influenced by
many factors, more research emphasis needs to be placed on understanding exhibit and
management characteristics’ effect on giraffe behavior.
The six GFPs observed in this study were also physically and temporally quite different.
Platforms (railings included) reached as high as 5.3 m and had a functional length ranging from
1.7 to 25.9 m. Food items for sale included leaves of romaine lettuce, sweet potato pieces, ryecrisp crackers, carrots, and browse leaves. Some program sessions were as short as 20 min (but
offered three times per day), while other programs operated throughout the entire business day.
Diversity within GFPs likely offered different experiences for both guests and giraffes, so future
studies should distinguish how specific aspects of GFPs influence the guest experience and
giraffe behavior and welfare.

Management Recommendations
Guest feeding programs at giraffe exhibits are becoming increasingly common as
reflected by their presence at approximately 57% of AZA-accredited zoos that exhibit giraffes
(determined from zoos’ websites and not including VIP tours). Aside from the presumed
educational and recreational benefits of up-close human-animal interactions to the visitors
(Beardsworth and Bryman, 2001; Kidd et al., 1995; Kreger and Mench, 1995; Woods, 2002),
these programs also provide a financial incentive to the institution. Within this study alone, a
31

single food portion was sold for as much as $5.00 (although most sold for $1.00-$2.00), and one
zoo reported a $1,600 income from their GFP during a single observation day. This easy-toimplement, lucrative program may provide extra funds to general zoo operations, animal care, or
specific conservation programs.
One of the most significant effects of GFPs was increased idleness observed in
participating giraffes. Inactivity, while not necessary harmful, may lead to health problems and
development of replacement stereotypic behavior (Veasey et al., 1996); furthermore, idleness
also does not instill interest among zoo guests (Altman, 1998; Bitgood et al., 1988; Margulis et
al., 2003). Since results showed that idleness was only significantly higher at zoos that offered
all-day GFPs, this problem can be avoided by structuring GFPs into multiple short sessions.
Shorter sessions may also help to assure a steady crowd of visitors during the GFP, therefore
limiting the opportunity for giraffes to remain idle.
Anecdotal problems cited by multiple animal managers during the survey portion of the
study included feeding of undesirable items by visitors (non-approved foods such as poisonous
plants and human snackfood), teasing by visitors, giraffe reluctance to participate in programs,
and giraffe dominance over conspecifics at feeding platforms (also described by Burgess, 2004).
While proximate zoo staff surveillance may help reduce guest-driven incidents, special stationor target-training may be necessary to encourage timid giraffes and control those more dominant.
Two zoos were working to station-train their giraffes to stand at specific parts of the platform
and were successful at keeping giraffes at their stations as long as there was a fairly continuous
flow of food being offered. A quiet, calm atmosphere at the feeding platform was also observed
to help maintain giraffe presence. Loudspeaker systems and whistles were not observed to be

32

effective at attracting and maintaining giraffe presence at feeding platforms; these only added to
the disturbance brought about by visitor crowds.
Managers should recognize that animals (at the species and individual level) will react
differently to contact and interaction contexts. Although this study suggests that no reduction of
giraffe welfare resulted from GFPs, managers should not assume that all individuals will be
affected in the same way or to the same degree, nor will other species in other human-animal
interactive programs. Other interactive experiences, such as stingray touch pools,
parakeet/lorikeet aviaries, and petting zoos, need to be considered as unique opportunities with
each meriting scientific research to determine effects to animal welfare.

Future Studies
The next step in understanding effects of GFPs on giraffe welfare would be to perform
pre-and-post comparisons of physiological and behavioral indicators of welfare. By evaluating
the cause-and-effect relationship GFPs have on these parameters, further evidence will be gained
for comparison to the correlations presented in this study. Specifically, one can determine
whether GFPs cause a direct reduction in stereotypic behavior both short- and long-term after a
new GFP is established. Physiological indicators of stress, such as fecal cortisol levels, should
also be measured in connection with a newly established GFP.
More research also needs to be conducted on understanding the unfulfilled motivations
that cause oral stereotypic behavior. Excess salivation was observed during many bouts of oral
stereotypic behavior in this study; this has been one suggested purpose of stereotypic licking as
the production of excess saliva may help to buffer gastrointestinal acidity (Baxter and Plowman,
33

2001; Bergeron et al., 2006; Nicol et al., 2002; Wiepkema et al., 1987). Some giraffes were also
observed to involve a small volume of food, such as a single leaf or small portion of hay, in their
oral stereotypy. The food items were not actively ingested or ruminated, only manipulated. This
observation was also made by Koene and Visser (1996), and yet, the purpose of the food items’
involvement in oral stereotypic behavior, specifically how it may relate to salivation, is not
understood. These observations should be studied further with specific investigation to the
relationship of captive diets and gastrointestinal health with performance of oral stereotypies in
ungulates.
One interesting finding of this study was that grooming behavior significantly decreased
with age, and had a tendency to be more frequent in male giraffes. Grooming has not been
identified as a key behavior in a giraffe’s behavioral repertoire, but it may be interesting to know
how grooming affects fitness levels, especially for younger, male giraffes (e.g. mate attractivity).

34

CONCLUSION
Results from the present study suggest that guest feeding programs at zoos have no
apparent detrimental impact on giraffe welfare, as indicated by behavioral time budgets
encompassing observations of both normal and stereotypic behaviors. Giraffe participation in
guest feeding programs does not significantly influence their performance of stereotypic
behavior. However, increased time spent engaged in feeding did significantly correlate with
decreased levels of oral stereotypic behavior, and these levels further decreased with increased
individual participation in guest feeding programs. This indicates that managers can potentially
reduce oral stereotypies, an indicator of sub-optimal welfare, by increasing opportunities for
giraffes to fulfill feeding motivation. Giraffe idleness can be problematic at guest feeding
platforms, but resolution may come from adjusted management and program scheduling. As zoos
strive to enhance the educational and entertainment value of a guest’s experience through
human-animal interactions, they must also ensure optimal welfare of individual animals
involved. If properly implemented, guest feeding programs can provide financial benefits to the
institution and potential enrichment to giraffe welfare.

35

APPENDICES

36

Time
0:00
0:01
0:02
0:03
0:04
0:05
0:06
0:07
0:08
0:09
…
0:25
0:26
0:27
0:28
0:29

Comments:
37

Behavior
Active Feeds at
Platform (y/n)

> 3m of Platform

< 3m of Platform

Time Out

Other

Pacing

Aberrant
oral behavior

Tonguerolling

Humidity: ______

Objectlicking

-

Objectmanipulating

Time Budget Data Sheet

Grooming

Socializing

Drinking

Ruminating

-

Eating

Temp: ______

Locomoting

Idle

APPENDIX A

Date:
Site:
Start Time:
Platform

APPENDIX B
Stereotypic Behaviors
Data Sheet
Temp: ______
Humidity: ______

Giraffe ID

Start
Time

End Time

Legend:
T - Tongue-rolling
O - Object-licking
A - Aberrant oral behavior
P - Pacing

Date:
Site:
Start Time:

Type of
Stereotypic
Behavior

Additional
Comments:

38

Comments

APPENDIX C
Management Survey

Institution: _______________________________

1) Animals
Individual Sex

Age
(years)

Years
at Zoo

Reproductive/Health
Status

Hand-Reared vs.
Mother-Reared

□M □F
□M □F
□M □F
□M □F
□M □F
□M □F
□M □F
□M □F
Individual Personality Type (check all that apply):
□ Docile
□ Docile
□ Docile
□ Docile
□ Docile
□ Docile
□ Docile
□ Docile

□ Aggressive
□ Aggressive
□ Aggressive
□ Aggressive
□ Aggressive
□ Aggressive
□ Aggressive
□ Aggressive

□ Playful
□ Playful
□ Playful
□ Playful
□ Playful
□ Playful
□ Playful
□ Playful

□ Timid
□ Timid
□ Timid
□ Timid
□ Timid
□ Timid
□ Timid
□ Timid

Other
Comments
□ Investigative
□ Investigative
□ Investigative
□ Investigative
□ Investigative
□ Investigative
□ Investigative
□ Investigative

2) Giraffe Keeper Staff
a. How many different keepers / managers directly care for giraffes?

b. Does keeper staff have tactile interaction with giraffes? If so, estimate the amount of
time this occurs on a weekly basis.

39

3) Exhibit Characteristics
Outdoor
Area size:
Substrate (check all that apply):
Features (check all that apply):

□ Dirt □ Sand □ Concrete □ Grass □ Other:
□ Pool
□
□ Running water
□
□ Elevation changes
□
□ Trees
□
□ Fields
□ Shade structure
Is the overall surface plane slanted?
□ Yes □ No
Are the animals able to hide from public view? □ Yes □ No
Indoor/Holding
Area size:
Substrate (check all that apply): □ Dirt □ Sand □ Concrete □ Wood □ Tile □
Hay/Straw □Other: __________________
Features (check all that apply): □ Hay/straw
□
□ Separate stalls
□
How many:
____________
Is the floor slanted?
□ Yes □ No
Are the animals able to hide
□ Yes □ No
from public view?
What form of lighting is used?
□ Sunlight □ Fluorescent □ Incandescent
□ Wide Spectrum □ Other: _______________
How many hours of artificial
lighting per day?
Containment
Features (check all that apply): □ Mesh or chain link fence □ Metal bars □ Moat
□ Glass □ Hot wire □ Gunite □ Cable □ Other:
Surroundings
Proximity to predator (y/n):
Visual:
Proximity to visitor conveniences
Visual:
(e.g., food, retail, rides, shows) (y/n):
Public-Animal Interface
Minimum distance between public and
animals at normal viewing area:
% of exhibit under public view:
Number of sides (max 4) that public
surrounds exhibit:
Animal Composition:
Is this a mixed-species exhibit?
□ Yes □ No If so, what species:
How many hours per summer day
are the animals put on exhibits:
40

4) Diet & Feeding Schedule
Please list all components of the animals’ normal diet:
Food (Type and/or Brand)

Quantity per Day

What % of Food is Consumed

Browse

Quantity per Day

# Days given per week

Vitamin Supplements

# Days given per Week

Outside of any feeding program:
a. When and where do feedings take place?

b. In what sort of container is food presented?

c. In what sort of container is water presented?
d. Please list all “treats” given (and how often) to the animals:

5) Veterinary Care
Please list animals and the medical care they receive or have received that may alter
behavior:
Individual

List any medical care given to each giraffe:
(e.g., medications for disease, injury, parasite)

41

6) Training
Please list animals and the behaviors trained:
Ind.

Frequency
of training
sessions

Amount of
time per day
being trained

Level of
Desired
Training
(1-5)

Animal Response to Training
(retention of trained behavior)
□ Poor
□ Poor
□ Poor
□ Poor
□ Poor
□ Poor
□ Poor
□ Poor

□ Fair
□ Fair
□ Fair
□ Fair
□ Fair
□ Fair
□ Fair
□ Fair

□ Good
□ Good
□ Good
□ Good
□ Good
□ Good
□ Good
□ Good

a. What key behaviors have been trained:

b. Number of personnel training animals:
c. What training techniques are used (Check all that apply)?
Target training
Clicker training
Positive reinforcement
Negative reinforcement

d. What type of reward is offered during training (Check all that apply)?
Vocal praise
Tactile praise (e.g., pat or rub)
Food treat

e. Does any of the training have direct public education purposes?

42

□ Excellent
□ Excellent
□ Excellent
□ Excellent
□ Excellent
□ Excellent
□ Excellent
□ Excellent

7) Enrichment
Please list all enrichment items used (if applicable):
Frequency Used Duration Used

Animal Response (Interest)
□ Poor □ Fair □ Good □ Excellent
□ Poor □ Fair □ Good □ Excellent
□ Poor □ Fair □ Good □ Excellent
□ Poor □ Fair □ Good □ Excellent
□ Poor □ Fair □ Good □ Excellent
□ Poor □ Fair □ Good □ Excellent
□ Poor □ Fair □ Good □ Excellent
□ Poor □ Fair □ Good □ Excellent

8) Stereotypic Behaviors
Please list any stereotypies associated with your animals:
Behavior

Estimated
% Time of
Occurrence

Animal/Location/
Time of Day

History (Were stereotypies
performed at previous
institution?)

Object licking
Tongue playing
Mouthing/cribbing
Conspecific licking
Pacing

9) Giraffe Public Feeding Programs
Please identify the behavioral experience each giraffe has with feeding program:
Individual
Docile
Aggressive
Timid
Approach spontaneously
Approach when called
Other (please specify)
Years at present program
Years at previous program
a. Is touching of giraffes permitted? Do giraffes allow it?

43

Please describe the conditions of the giraffe feeding platform and program:
Feeding Platform
Length (adjacent to exhibit):
Height (from giraffe’s perspective, top of rail):
Composition of platform (e.g., wood, metal)
Substrate at base of platform (where giraffes stand):
Max number of people it can hold:
Feeding Program
Type of feed offered
Daily frequency of feeding events:
Duration of each feeding event:
Daily average amount of people participating
in giraffe feeding:
Is there any record of visitor aggravation
(e.g., illegal feedings, illegal entry to
exhibit)?
a. Do any other demonstrations or keeper chats occur in or adjacent to giraffe exhibits?

b. Have there been any issues or problems in adapting giraffes to your feeding program?

c. Does your feeding program and a giraffe’s participation in it affect how you would
manage your giraffe herd? This includes breeding and animal transfers.

d. How does the zoo benefit financially from the guest feeding program?

10) Zoo Attendance
Or Contact Information:

Total for ___ / ___ / 2012:
Total for ___ / ___ / 2012:
Total for ___ / ___ / 2012:

11) Additional Questions or Comments

44

LITERATURE CITED

45

LITERATURE CITED
Altmann, J. 1974. Observational Study of Behavior: Sampling Methods. Behaviour 49(3-4):
227-267.
Altman, J.D. 1998. Animal activity and visitor learning at the zoo. Anthrozoös 11(1): 12-21.
Anderson, U.S., A.S. Kelling, R. Pressley-Keough, M.A. Bloomsmith, and T.L. Maple. 2003.
Enhancing the Zoo Visitors’ Experience by Public Animal Training and Oral
Interpretation at an Otter Exhibit. Environment and Behavior 35(6): 826-841.
AZA Animal Welfare Committee. 2013. Association of Zoos & Aquariums.
URL http://www.aza.org/Membership/detail.aspx?id=378
AZA Program Animal Policy. 2011. Accreditation Standards and Related Policies. Association
of Zoos & Aquariums. p. 59-60.
Bashaw, M.J., L.R. Tarou, T.S. Maki, and T.L. Maple. 2001. A survey assessment of variable
related to stereotypy in captive giraffe and okapi. Applied Animal Behaviour Science
73(3): 235-247.
Bashaw, M.J., M.A. Bloomsmith, T.L. Maple, and F.B. Bercovitch. 2007. The Structure of
Social Relationships Among Captive Female Giraffe (Giraffa camelopardalis). Journal of
Comparative Psychology 121(1): 46-53.
Bashaw, M.J. 2011. Consistency of captive giraffe behavior under two different management
regimes. Zoo Biology 30(4): 371-378.
Bates, D., M. Maechler, and B. Bolker. 2011. lme4: Linear mixed-effects models using S4
classes. R package version 0.999375-42. http://CRAN.R-project.org/package=lme4.
Baxter, E. and A.B. Plowman. 2001. The effect of increasing dietary fibre on feeding,
rumination and oral stereotypies in captive giraffes (Giraffa camelopardalis). Animal
Welfare 10(3): 281-290.
Beardsworth, A. and A. Bryman. 2001. The wild animal in late modernity: The case of the
Disneyization of zoos. Tourist Studies 1(1): 83-104.
Bergeron, R., A.J. Badnell-Waters, S. Lambton, and G. Mason. 2006. Stereotypic Oral
Behaviour in Captive Ungulates: Foraging, Diet and Gastrointestinal Function. In: G.
Mason and J. Rushen, editors. Stereotypic Animal Behaviour: Fundamentals and
Applications to Welfare, Second Edition. Wallingford, UK: CABI. p. 19-57.
Bitgood, S. 1988. Exhibit Design and Visitor Behavior: “Empirical Relationships”. Environment
and Behavior 20(4): 474-491.

46

Broom, D.M. 1991a. Animal welfare: concepts and measurement. Journal of Animal Science
69(10): 4167-4175.
Broom, D.M. 1991b. Assessing welfare and suffering. Behavioural Processes 25(2-3): 117-123.
Burgess, A. 2004. The Giraffe Resource Husbandry Manual. Antelope/Giraffe Taxon Advisory
Group, The Association of Zoos and Aquariums.
Carlstead K. 1998. Determining the Causes of Stereotypic Behavior in Zoo Carnivores. In: D.J.
Shepherdson, J.D. Mellen, and M. Hutchins, editors. Second Nature: Environmental
Enrichment for Captive Animals. Washington, DC: Smithsonian Institution Press. p.
172-83.
Carlstead, K., J. Fraser, C. Bennett, and D.G. Kleiman. 1999. Black Rhinoceros (Diceros
bicornis) in U.S. Zoos: II. Behavior, Breeding Success, and Mortality in Relation to
Housing Facilities. Zoo Biology 18(1): 35-52.
Cook, S. and G.R. Hosey. 1995. Interaction Sequences Between Chimpanzees and Human
Visitors at the Zoo. Zoo Biology 14(5): 431-440.
Claxton, A.M. 2011. The potential of the human–animal relationship as an environmental
enrichment for the welfare of zoo-housed animals. Applied Animal Behaviour Science
133(1): 1-10.
Davey, G. 2007. Visitors’ Effects on the Welfare of Animals in the Zoo: A Review. Journal of
Applied Animal Welfare Science 10(2): 169-183.
del Castillo, S.M., M.G. Bashaw, M.L. Patton, R.R. Rieches, and F.B. Bercovitch. 2005. Fecal
steroid analysis of female giraffe (Giraffa camelopardalis) reproductive condition and the
impact of endocrine status on daily time budgets. General and Comparative
Endocrinology 141(3): 271-281.
Dittrich, L. 1976. Food presentation in relation to behaviour in ungulates. International Zoo
Yearbook 16(1): 48-54.
Estes, R.D. 1999. The Safari Companion: A Guide to Watching African Mammals. White River
Junction, VT: Chelsea Green Publishing Company. Print.
Fernandez, L.T., M.J. Bashaw, R.L. Sartor, N.R. Bouwens, and T.S. Maki. 2008. Tongue
Twisters: Feeding Enrichment to Reduce Oral Stereotypy in Giraffe. Zoo Biology 27(3):
200-212.
Fernandez, E.J., M.A. Tamborski, S.R. Pickens, and W. Timberlake. 2009. Animal-visitor
interactions in the modern zoo: Conflicts and interventions. Applied Animal Behaviour
Science 120 (1-2): 1-8.

47

Fraser, A.F. and D.M. Broom. 1990. Farm animal behaviour and welfare. London, UK: Ballière
Tindall. Print.
Ginnett, T.F. and M.W. Demment. 1997. Sex differences in giraffe foraging behavior at two
spatial scales. Oecologia 110(2): 291-300.
Horwich R.H., C. Kitchen, M. Wangel, and R. Ruthe. 1983. Behavioral Development in Okapis
and Giraffes. Zoo Biology 2(2): 105-125.
Hosie, L. and H. Turner. 2000. Oral behaviours performed by giraffes at Chester Zoo.
Proceedings of the 2nd Annual Symposium on Zoo Research. Federation of Zoological
Gardens of Great Britain and Ireland. Paignton, Devon: 137-142.
Hosey, G.R. 2000. Zoo animals and their audiences: What is the visitor effect? Animal Welfare
9(4): 343-357.
Hosey, G. 2008. A preliminary model of human–animal relationships in the zoo. Applied Animal
Behavior Science 109(2): 105-127.
Hutchins, M., B. Smith, and R. Allard. 2003. In defense of zoos and aquariums: the ethical basis
for keeping wild animals in captivity. Journal of the American Veterinary Medical
Association 223(7): 958-966.
Kidd, A.H., R.M. Kidd, and R.L. Zasloff. 1995. Developmental factors in positive attitudes
toward zoo animals. Psychological Reports. 76(1): 71-81.
Kinahan, A. and N. Marples 2000. Behavioural studies on captive tapirs Tapirus terrestris and
giraffes Giraffa camelopardalis, with particular attention to feeding. Proceedings of the
2nd Annual Symposium on Zoo Research. Federation of Zoological Gardens of Great
Britain and Ireland. Paignton, Devon. p. 127-136.
Koene, P. and E.K. Visser. 1996. Tongue playing behavior in captive giraffes. Proceedings of the
1st International Symposium on Physiology and Ethology of Wild and Zoo Animals.
Berlin, Germany. 2: 106-111.
Koene, P. 1999. When feeding is just eating: how do farm and zoo animals use their spare time?
Regulation of Feed Intake. Wallingford, UK: CABI. Print.
Komsta, L. 2011. outliers: Tests for outliers. R package version 0.14.
http://CRAN.R-project.org/package=outliers.
Kreger, M.D. and J.A. Mench. 1995. Visitor–animal interactions at the zoo. Anthrozoös 8(3):
143-158.
Lawrence, A.B. and J. Rushen. 1993. Stereotypic Behaviour, Fundamentals and Applications to
Welfare. Wallingford, UK: CABI. Print.
48

Leuthold, B. M. and W. Leuthold. 1978. Daytime activity patterns of gerenuk and giraffe in
Tsavo National Park, Kenya. East African Wildlife Journal 16(4): 231-243.
Leuthold, B.M. 1979. Social organization and behaviour of giraffe in Tsavo East National Park.
East African Wildlife Journal 17: 19-34.
Margulis, S.W., C. Hoyos, and M. Anderson. 2003. Effect of Felid Activity on Zoo Visitor
Interest. Zoo Biology 22(6): 587-599.
Martin, P., and P. Bateson. 2007. Measuring Behavior: An Introductory Guide. New York:
Cambridge University Press. Print.
Mason, G. 1991a. Stereotypies: a critical review. Animal Behaviour 41(6): 1015-1037.
Mason, G. 1991b. Stereotypies and Suffering. Behavioural Processes.25: 103-115.
Mason, G.J., and N.R. Latham, 2004. Can’t stop, won’t stop: is stereotypy a reliable animal
welfare indicator? Animal Welfare 13(1): 57-69.
Mason, G. and J. Rushen 2006. Stereotypic Animal Behaviour: Fundamentals and Applications
to Welfare, Second Edition. Wallingford, UK: CABI. Print.
Melfi V.A. 2009. There Are Big Gaps in Our Knowledge, and Thus Approach, to Zoo Animal
Welfare: A Case for Evidence-Based Zoo Animal Management. Zoo Biology 28(6):
574-588.
Mitlöhner, F.M., J.L. Marrow-Tesch, S.C. Wislon, J.W. Dailey, and J.J. McGlone. 2001.
Behavioral sampling techniques for feedlot cattle. Journal of Animal Science 79(5):
1189-1193.
Montaudouin S., and G. Le Pape, 2004. Comparison between 28 zoological parks: stereotypic
and social behaviours of captive brown bears (Ursus arctos). Applied Animal Behaviour
Science 92(1-2): 129-141.
Morgan, J.M., and J.H. Gramann 1989. Predicting Effectiveness of Wildlife Education
Programs: A Study of Students' Attitudes and Knowledge toward Snakes. Wildlife
Society Bulletin 17(4): 501-509.
Morgan, K.N., and C.T. Tromberg. 2007. Sources of stress in captivity. Applied Animal
Behaviour Science 102(3-4): 262-302.
Newberry, R.C. 1995. Environmental enrichment: Increasing the biological relevance of captive
environments. Applied Animal Behavior Science 44(2): 229-243.

49

Nicol. C.J., H.P.D. Davidson, P.A. Harris, A.J. Waters, and A.D. Wilson. 2002. Study of cribbiting and gastric inflammation and ulceration in young horses. Veterinary Record
151(22): 658-662.
Nimon, A.J., and F.R. Dalziel, 1992. Cross-species interaction and communication: a study
method applied to captive siamang (Hylobates syndactylus) and long-billed corella
(Cacatua tenuirostris) contacts with humans. Applied Animal Behaviour Science 33(2):
261-272.
Patrick, P., C. Matthews, D. Ayers, and S. Tunnicliffe. 2007. Conservation and education:
prominent themes in zoo mission statements. The Journal of Environmental Education
38(3): 53-60.s
Pellew, R.A. 1984. The feeding ecology of a selective browser, the giraffe (Giraffa
camelopardalis tippelskirchi). Journal of Zoology 202(1): 57–81.
Perkins, L.A. 1992. Variables That Influence the Activity of Captive Orangutans. Zoo Biology
11(3): 177-186.
R Development Core Team. 2011. R: A language and environment for statistical computing.
R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0,
URL http://www.R-project.org/.
Read, L.S., and N.K. Warren. 1996. The modem zoo: How do people perceive zoo animals?
Applied Animal Behaviour Science 47(1): 109-118.
Seeber, P.A., I Ciofolo, and A. Ganswindt. 2012. Behavioural inventory of the giraffe (Giraffa
camelopardalis) BioMed Central Research Notes 5: 650.
Sherwood, K.P., S.F. Rallis, and J. Stone. 1989. Effects of Live Animals vs. Preserved
Specimens on Student Learning. Zoo Biology 8(1): 99-104.
Swaisgood, R. and D. Shepherdson. 2006. Environmental Enrichment as a Strategy for
Mitigating Stereotypy in Zoo Animals: a Literature Review and Meta-analysis. In: G.
Mason and J. Rushen, editors. Stereotypic Animal Behaviour: Fundamentals and
Applications to Welfare, Second Edition. Wallingford, UK: CABI. p. 256-285.
Swanagan, J.S. 2000. Factors Influencing Zoo Visitor’s Conservation Attitudes and Behavior.
The Journal of Environmental Education 31(4): 26-31.
Tarou, L.R., M.J. Bashaw, and T.L. Maple. 2000. Social Attachment in Giraffe: Response to
Social Separation. Zoo Biology 19(1): 41-51.
Tarou, L.R., M.J. Bashaw, and T.L. Maple. 2003. Failure of a Chemical Spray to Significantly
Reduce Stereotypic Licking in a Captive Giraffe. Zoo Biology 22(6): 601-607.

50

Tarou, L.R and M.J. Bashaw. 2007. Maximizing the effectiveness of environmental enrichment:
Suggestions from the experimental analysis of behavior. Applied Animal Behaviour
Science 102 (3): 189-204.
Trunkfield, H.R. and D.M. Broom. 1990. The welfare of calves during handling and transport.
Applied Animal Behaviour Science 28 (1): 135-152.
Veasey, J.S., N.K. Waran, and R.J. Young. 1996. On comparing the behavior of zoo housed
animals with wild conspecifics as a welfare indicator, using the giraffe (Giraffa
camelopardalis) as a model. Animal Welfare 5(2): 139-153.
Wickens-Dražilová, D. 2006. Zoo Animal Welfare. Journal of Agricultural and Environmental
Ethics 19(1): 27–36.
Wielebnowski, N.C., N. Fletchall, K. Carlstead, J.M. Busso, and J.L. Brown. 2002.
Noninvasive Assessment of Adrenal Activity Associated with Husbandry and Behavioral
Factors in the North American Clouded Leopard Population. Zoo Biology 21(1): 77-98.
Wiepkema, P.R., K.K. Van Hellemond, P. Roessingh, and H. Romberg. 1987. Behaviour and
Abomasal Damage in Individual Veal Calves. Applied Animal Behaviour Science 18(3):
257-268.
Woods, B. 2002. Good zoo/bad zoo: Visitor experiences in captive settings. Anthrozoös 15(4):
343-360.
Young, R.J. 1997. The importance of food presentation for animal welfare and conservation.
Proceedings of the Nutrition Society 56(3): 1095-1104.

51