Content uploaded by Alvaro López Caicoya
Author content
All content in this area was uploaded by Alvaro López Caicoya on Nov 01, 2018
Content may be subject to copyright.
Giraffes’ physical cognition
1
© 2018, American Psychological Association. This paper is not the copy of record and may
not exactly replicate the final, authoritative version of the article. Please do not copy or cite
without authors' permission. The final article will be available, upon publication, via its DOI:
10.1037/com0000142
Accesing the original paper is strongly recommended, it includes videos of every test
Giraffes’ physical cognition
2
Object permanence in Giraffa camelopardalis: first steps in
giraffes’ physical cognition
Alvaro L. Caicoya a, Federica Amici b,c, Conrad Ensenyat d and Montserrat Colell a,e
a Department of Clinical Psychology and Psychobiology, Faculty of Psychology, University
of Barcelona, Barcelona, Spain.
b Institute of Biology, Faculty of Life Sciences, University of Leipzig, Leipzig, Germany
c Junior Research Group "Primate Kin Selection", Department of Primatology, Max Planck
Institute for Evolutionary Anthropology, Leipzig, Germany
d Barcelona Zoo, Spain.
e Institute of Neurosciences, University of Barcelona, Barcelona, Spain.
Corresponding author: Álvaro L. Caicoya. alocaico@gmail.com
Acknowledgments:
We thank the staff at the facilities of Barcelona and Leipzig, Oscar Quilez Peña,
Bibiana Martin Prat, Ruben Holland, Rene Forberg, Daniel Volkmann, Kathrin Dorn, Marco
Mehner, Stefan Lohmer, Michael Tempelhof and all the others for endless support and
cooperation. We also thank Fred Bercovitch for his clever commentaries on a previous
version of this manuscript. Last, we would like to thank Maria Luisa Caicoya for helping us
coding all the trials for inter-observer reliability.
This research was supported by a PRIC grant, convocatoria 2016/2017, Fundación
Zoo de Barcelona.
Giraffes’ physical cognition
3
Abstract
Although behaviour, biology and ecology of giraffes have been widely studied, little
is known about their cognition. Giraffes’ feeding ecology and their fission-fusion social
dynamics are comparable to those of chimpanzees (Pan troglodytes), suggesting that they
might have complex cognitive abilities. To assess this, we tested 6 captive giraffes on their
object permanence, short-term memory and ability to use acoustic cues to locate food. Firstly,
we tested whether giraffes understand that objects continue to exist even when they are out of
sight. Giraffes saw one of two opaque containers containing food, then containers were
closed, and 2 seconds later giraffes could choose one. Secondly, we measured giraffes’
memory repeating the procedure but with a delay of 30 seconds, 60 seconds or 2 minutes
between closing the containers and subjects’ choice. Finally we investigated whether giraffes
could locate food inside one of two identical opaque containers, when the only cue provided
was the sound made by food when shaking the baited container, or the lack of sound when
shaking the empty container. Our results show that giraffes form mental representations of
completely hidden objects, but may not store them for longer than 30 seconds. Moreover,
they rely on stimulus enhancement rather than acoustic cues to locate food, when no visual
cues are provided. Finally, we argue that giraffes and other ungulates might be a suitable
model to investigate the evolution of complex cognitive abilities from a comparative
perspective.
Key words: Object permanence, short-term memory, acoustic cues, giraffe, cognition,
bayesian statistics.
Giraffes’ physical cognition
4
Introduction
Giraffes are a neglected species in science, with only 400 scientific papers having
been written about them, as compared to 20,000 papers on white rhinos, for instance
(Dell'Amore, 2016). Besides this, all we know about giraffes is limited to their behaviour,
biology and ecology, while nothing is to our knowledge known about their cognition (Dagg,
2014; Seeber, Ciofolo, & Ganswindt, 2012; Shorrocks, 2016). During the past 15 years, the
number of wild giraffes has plummeted from an estimated 140,000 to around 9,000, and our
limited knowledge about them makes conservation efforts a very difficult task (Muller et al.,
2016).
In the wild, giraffes form fission-fusion groups, with individuals joining subgroups of
different size and composition, which vary from one individual (usually adult bulls) up to
100. Cows usually form small subgroups with calves and other cows, and the probability and
stability of a relationship between two giraffes is determined by kinship and shared space
(Bercovitch & Berry, 2013; Carter, Seddon, Frère, Carter, & Goldizen, 2013; VanderWaal,
Wang, McCowan, Fushing & Isbell, 2014). These fission-fusion social dynamics are thought
to be linked to the evolution of complex cognitive abilities, as for example in chimpanzees
(Pan troglodytes), although subtle differences in fission-fusion dynamics may posit different
cognitive challenges across animal taxa (Aureli et al., 2008).
From an ecological point of view, giraffes have a very diverse diet, eating different
specific leaves and flowers depending on the time of the year (Berry & Bercovitch, 2017).
Indeed, new long-term evidence has shown that giraffes’ diet is much more varied than
previously thought, and their dietary breadth is comparable to that of chimpanzees (Berry &
Bercovitch, 2017). Chimpanzees at Ngogo spend 77% of their feeding time eating 15 species
out of the 102 identified in their diet (Shannon–Weiner Diversity Index H´ = 3.282; Watts,
Giraffes’ physical cognition
5
Potts, Lwanda, & Mitani, 2012), while giraffes from the Luangwa Valley spend a comparable
66% of their feeding time eating 15 out of the 93 plant species in their diet (H´ = 3.699; Berry
& Bercovitch, 2017). Such dietary breadth may also be linked to enhanced cognitive skills,
with dietary complexity being one possible primary driver of cognitive evolution in primates
and other species (see MacLean et al., 2014).
Living in fission-fusion groups and having complex diets make giraffes excellent
candidates for having enhanced cognitive skills. Given our poor knowledge about giraffes’
cognition, a good starting point might be the study of object cognition. Indeed, the ability to
segment the world into discrete objects that exist independently of us through space and time
is one of the most fundamental conceptual structures, and therefore a widely studied area in
comparative psychology (Cacchione & Rakoczy, 2017). A very basic form of object
cognition is object permanence, i.e. the ability to understand that objects continue to exist
even when they are out of sight (Piaget, 1954). Piaget originally designed different object
permanence tasks to assess cognitive development in human infants, dividing them into 6
stages of increasing difficulty. Object permanence develops gradually through these 6 stages
also in a variety of species such as nonhuman great apes (Call, 2001), corvids (Bugnyar,
Stöwe, & Heinrich, 2007) and parrots (Pepperberg, Willner, & Gravitz, 1997). The stage that
an animal is capable of reaching is often used as a measure of the species’ cognitive abilities
as compared to others (e.g. Manger, 2013). Stage 4, for instance, is largely considered the
first true form of object permanence (Cacchione & Rakoczy, 2017). This stage is usually
tested by showing a reward to the subject and then hiding it under one of several identical
opaque containers: if subjects locate the hidden reward at first choice, they are assumed to
have acquired stage 4 of object permanence.
Previous studies have increased memory demands of object permanence tasks by
introducing a delay between the baiting procedure and the moment in which subjects can
Giraffes’ physical cognition
6
retrieve the food: after the reward has been hidden, the subject waits for a specific period of
time before selecting one of the containers (e.g. Barth & Call, 2006). In this way, memory
demands are increased, and individuals have to remember the food location in the face of
increasing delays. Unsurprisingly, performance declines with increasing delays across several
species (see Cacchione & Rakoczy, 2017). Another modification that has been done to this
paradigm is providing an acoustic cue (e.g. making the baited container produce a noise)
instead of a visual one (e.g. showing the reward being hidden: Call, 2004). In this task, if
subjects understand the causal connection between objects and the noise produced when they
move, they should infer that the noisy container is the one that contains the reward.
Moreover, when the empty container is shaken instead, subjects should use the absence of
noise to infer by exclusion that the unshaken container must be baited. This paradigm has
been already tested with children (e.g. Hill, Collier-Baker, & Suddendorf, 2012), all species
of great apes (Call, 2004), corvids (Shaw, Plotnik, & Clayton, 2013), pigs and boars
(Albiach-Serrano, Bräuer, Cacchione, Zickert, & Amici, 2012), among others. A similar
inference-by-exclusion paradigm (with visual instead of acoustic cues) has also been used
with two relatives of the giraffe, goats and sheep: subjects were shown either an empty or a
full container, and then had to search for food (Nawroth, von Borell, & Langbein, 2014).
Both species correctly solved the usual object permanence test (i.e. when the full container
was shown), but only goats solved it when being provided with mere indirect information and
inference by exclusion was needed (i.e. when the empty container was shown).
To our knowledge, none of these paradigms, nor any other paradigm assessing
cognition, has so far been used to test giraffes. In this study, we therefore aimed to start
exploring giraffes’ cognition with an object permanence task. This task may be an ideal
experimental tool, testing a basic cognitive skill (which is required for more complex skills to
emerge) and having been widely used in comparative psychology, therefore also allowing
Giraffes’ physical cognition
7
comparisons across species. In giraffes, object permanence may play an essential role when
dealing with social partners or predators, which may not always be visible but still exist.
However, object permanence may be ecologically less relevant when considering their diet,
as giraffes typically move from tree to tree to eat visible leaves (Leuthold & Leuthold, 1972).
In this respect, testing object permanence in giraffes can be important to understand how
cognitive skills really map the socio- and/or ecological challenges faced by giraffes in their
every-day life and how modular brains are (Amici, Barney, Johnson, Call, & Aureli, 2012).
We conducted 3 different experiments on giraffes. Firstly, we tested stage 4 of object
permanence by visibly hiding food in one of two opaque containers. Secondly, we tested their
short-term memory skills, by increasing the delay between baiting and retrieval up to 2
minutes. Finally, we tested their inferential skills by testing their ability to use the presence or
absence of acoustic cues to locate food. If giraffes’ ecological (i.e. dietary breadth) or social
characteristics (i.e. fission-fusion social dynamics) are linked to their cognitive skills, they
should perform above chance across all experiments, similarly to other species with similar
socio-ecological characteristics that have been shown to master these tasks (e.g. Barth & Call,
2006; Call, 2004).
Method
Ethics statement
The Barcelona Zoo and the Leipzig Zoo controlled and approved all the procedures.
Given that giraffes participated on a completely voluntary basis, and no invasive procedures
were used, no formal approval was required. During the task, moreover, individuals were
never food deprived, and motivation to participate was ensured exclusively by the use of
highly-preferred food (i.e. carrots, carob pellets and apples). Before testing started, we
assessed participant’s food preferences by presenting them with two food types and making
Giraffes’ physical cognition
8
them choose one, with each possible comparison being repeated 12 times per individual. The
experiments thus provided a form of enrichment for the subjects and did not present any risks
or adverse effects.
Subjects and materials
We tested 6 giraffes (Giraffa camelopardalis) ranging from 1 to 21 years of age and
housed at the zoos of Barcelona and Leipzig (see Table 1). All study subjects were
consistently fed a diet of fruit and vegetables. None of the subjects had previous experience
with the materials used in our experiments, and none had ever been tested in any cognitive
task. The tests were carried out in the interior facilities, after isolating the participant giraffe
from the group. Subjects kept visual, auditory and potentially tactile contact with the rest of
the group in all cases. Only the experimenter remained in the enclosure during the tests.
For all the experimental conditions we used two identical opaque containers of
approximately 15x15x3cm. Using only two containers allowed us to test naïve subjects with
an easier set-up, as often is done in literature (e.g. Albiach-Serrano et al., 2012; Chiandetti &
Vallortigara, 2011; Nawroth, von Borell, & Langbein, 2015). Depending on the result of
previous individual preference tests, a piece of carrot or apple was used as a reward in case of
a correct choice; in order to fill the container and facilitate visual discrimination, the food
reward was laid on a bed of carobs or pellets, depending on the diet restrictions at the two
facilities. Every trial was recorded from a video-camera fixed one meter behind the
experimenter’s back. All experimental conditions were administered in a pseudo-randomized
order, except from the habituation phase, which was administered at the beginning to all
subjects. Trials always started when the subject’s head was in front of the experimenter, with
its head approximately between the two containers.
Giraffes’ physical cognition
9
All giraffes belonged to the subspecies Giraffa camelopardalis rothschildi. All of them were
born in captivity.
Habituation phase
Only one container was used. The experimenter baited the container out of the
subject’s view, and then showed the container to the subject. After 5 seconds the
experimenter closed the container and pushed it toward the subject. If the subject touched the
container, the experimenter opened the lid and let the subject eat the food. After 4 successful
retrievals out of 5 consecutive trials, the giraffe started the experimental phase.
Olfactory control condition
We used the same procedure as in the Habituation phase, but two containers were
used. Out of the subject’s view, the experimenter baited one of the two containers and closed
both of them. Then, the experimenter showed both containers to the subject, holding each one
in one hand, approximately at 80 cm from each other and around 50 cm from the subject.
After 5 seconds, the experimenter simultaneously moved both containers toward the subject,
and let the subject choose. The experimenter made this movement with his eyes closed to
avoid providing inadvertent cues to the subject: he could notice the choice of the giraffe
Table 1
Subjects participating in the study.
Name
Sex
Age (years)
Zoo
Rearing history
Nuru
F
8
Barcelona
Mother
Yalinga
F
13
Barcelona
Mother
Nakuru
M
1
Barcelona
Mother
Max
M
21
Leipzig
Nursery
Andrea
F
9
Leipzig
Mother
Ashanti
F
16
Leipzig
Mother
Giraffes’ physical cognition
10
because both containers were light enough to be clearly moved by the giraffe during the
election. If the subject touched the correct container, the experimenter opened the lid and let
the subject eat the food, while removing the unchosen container. If the subject touched the
wrong container, the experimenter opened its lid and showed its content to the subject, then
showed the content of the correct container and removed both. In this condition, the subject
had neither visual nor acoustic cues to locate the food, and could only rely on possible
olfactory cues coming from the container containing the reward. Chance performance (50%
of the trials) in this condition therefore indicated that subjects could not rely on olfactory cues
to locate food.
We followed the same procedure as in the Olfactory control condition, but in this case
the subject could see the baiting procedure. In full view of the subject, the experimenter
showed the content of both containers, holding each one in one of his hands as above, and
after 5 seconds he simultaneously closed the lids of both containers, making their content
invisible. After 2 seconds, the experimenter closed his eyes and approached both containers
to the subject, letting her choose (see an example in Video 1).
Giraffes’ physical cognition
11
Figure 1. (A) The experimenter shows the content of both recipients to one of the subjects.
(B) The subject makes a choice.Object permanence condition
A
B
Giraffes’ physical cognition
12
Memory condition
We used the same procedure as in the object permanence condition. The only
difference was the time elapsed between closing the lid and letting the subject choose.
Depending on the condition, the time elapse was 30 sec, 60 sec or 120 sec, instead of 2 sec.
The experimenter stared at the ground during this time, observing his watch (see an example
of the 30sec condition in Video 2).
Acoustic cues condition
We followed the same procedure as in the Object permanence condition, but this time
the opaque lids were closed before being presented to the subject, so that no visual cue was
given with regards to the food location. In the Shake full condition, the experimenter held
both containers slightly beyond the subject’s reach, and then shook 3 times the container
containing the reward. In this way, the carob/pellets inside the container made a loud noise,
similar to the sound of a rattle. After 2 seconds, the experimenter simultaneously pushed both
containers toward the subject and let her choose (see an example in Video 3). In the Shake
empty condition the procedure was identical, but this time the experimenter shook the empty
container, which thus made no sound. In this last condition, a correct choice was coded when
the giraffe selected the unshaken container, as this one contained the food reward.
Design
Each subject went first through a habituation phase. After this, each subject was
administered 12 trials for each of the 7 conditions, in a pseudo-randomized order. Within
each condition, the position of the food reward was also pseudo-randomized, being on the
right side in half of the trials. There was always a 2 minute break between trials. Sessions
continued until the subject stopped approaching the experimenter, usually around 30 minutes
after the first trial.
Giraffes’ physical cognition
13
Data analyses
An external observer coded 15% of all the trials from the video-recordings. This
observer was naïve to the hypotheses: as soon as the subject made a choice, she stopped the
tape and coded before seeing the experimenter’s reaction to the subject’s choice. Inter-
observer reliability was excellent (κ = 1, n = 75).
We used Bayesian statistics to analyse the results, instead of the traditional null-
hypothesis significance testing (NHST) (see NHST analyses in supplementary material 1).
This was due to various reasons (see Kruschke, 2014). Firstly, Bayesian statistics provide
more information about the analysed parameters, as compared to traditional hypothesis tests
(Kruschke, 2013). Secondly, Bayesian statistics require no corrections for multiple
comparisons and thus provide more statistical power, which is an advantage in case of
multiple testing. Moreover, Bayesian analyses use the highest density interval (HDI) instead
of the confidence interval employed for frequentist analysis. In particular, the HDI informs
about the probability that a certain hypothesis is true, given the data, and does not simply
accept or reject the null hypothesis, as NHST instead do. Therefore, the HDI reduces
uncertainty, indicating the most credible values and covering 95% of data distribution.
To assess performance, we used the Bayesian alternative to a one-sample t-test and
assessed whether each condition had its 95% HDI above chance (i.e. 6 out of 12 correct
choices, 50%). Our dependant variable was the percentage of correct responses. A correct
choice was scored if the subject chose the baited container by touching it with her lips or
tongue. In those conditions resulting in a preference for the baited container, we repeated the
analysis including only the first three trials, to assess performance before extensive learning
could take place. If subjects in these conditions had simply learned to associate cups with
food (e.g. the noisy cup and the food in the Acoustic cues condition), their performance
Giraffes’ physical cognition
14
should drop to chance levels in the first three trials. We applied a t distribution, as this has
fatter tails than the normal distribution and can better accommodate outliers in the model. We
used minimally informative priors as described by Kruschke (2013), i.e. normal priors with
large standard deviation for (µ), broad uniform priors for (σ), and a shifted-exponential prior
for (ν).
Since giraffes moved freely during the tests, we coded their body orientation in order
to know if they used it as an aid to make their choices (e.g. by maintaining their head oriented
toward the baited container throughout the trial). We thus counted the number of trials in
which giraffes maintained their heads oriented towards the same container from the end of
stimuli presentation (i.e. closing the lids in the visual conditions, or stopping containers’
movements in acoustic conditions) until the choice was made.
Results
In the Olfactory control condition (95% HDI: .32 - .57) giraffes performed at chance
level (50% correct choices), showing that they could not rely on olfactory cues to locate
hidden food.
Subjects performed above chance level in the Object permanence condition (95%
HDI: .67 - .75) and in the 30 sec Memory condition (95% HDI: .55 - .72). Equivalent results
were obtained if only including the three first trials of these two conditions (95% HDI > .5).
In the 60 sec Memory condition (95% HDI: .43 - .71) and in the 120 sec Memory condition
(95% HDI: .4 - .82) giraffes performed at chance levels (Figure 2).
In the Shake full condition (95% HDI: .68 - .92) subjects performed above chance
level but in the Shake empty condition giraffes performed below chance (95% HDI: .34 -
.37). See supplementary material for the complete data (See Figure 2).
Giraffes’ physical cognition
15
Figure 2. Mean ± SEM of correct choices in the forced-choice task. Values above 0.5
indicate a preference for the baited container. A 95% HDI above chance level (50%) is
indicated by an asterisk.
Figure 3. In gray, the percentage of trials in which giraffes maintained their heads oriented
toward the correct (Correct) or the incorrect (Errors) container, from the end of the stimuli
presentation, until their choice. In black, the percentage of trials in which giraffes did not
maintain head orientation toward the container throughout the trial.
Giraffes’ physical cognition
16
The percentage of trials in which giraffes maintained the head orientation toward the same
container throughout the trial was relatively high in the Object permanence (43%), Shake full
(58%) and Shake empty (76%) conditions, but much lower in the 30 sec (7%), 60 sec (1%)nd
120 sec (0%) Memory conditions (Figure 3).
Discussion
Chance performance in the Olfactory condition confirmed that giraffes in the other
conditions were not simply relying on smell to locate hidden food. Giraffes were able to
locate food both in the Object permanence condition (2 sec delay) and in the 30 sec Memory
condition, but not when delays were increased in the 60 sec and 120 sec Memory conditions,
suggesting a memory or an attention limit. Giraffes could further use the presence of acoustic
cues to locate food in the Shake full condition, but not the absence of acoustic cues to infer
food location in the Shake empty condition.
Chance performance in the olfactory condition demonstrates that subjects failed to
select the container containing food in the absence of visual and acoustic information.
Therefore, our results in the other conditions cannot be attributed to giraffes’ relying on
inadvertent cues provided by the experimenter or by food smell to solve the task. Moreover,
they are also not the result of learning, as giraffes either correctly solved the task from the
very first trials (Object permanence, 30 sec Memory and Shake full conditions), and correctly
performed also in the following trials, or failed to master a condition, even after being
administered multiple trials (60 sec and 120 sec Memory, and Shake empty conditions).
Giraffes reached stage 4 of object permanence, as they consistently performed above
chance level in our first condition. Understanding that objects exist even if being completely
hidden may be the result of different socio-ecological pressures. In giraffes, for example,
object permanence may be useful to track conspecifics – something important when living in
Giraffes’ physical cognition
17
groups (Zucca, Milos, & Vallortigara, 2007). However, object permanence is also a
widespread ability in vertebrates (Cacchione & Rakoczy, 2017). Therefore, it is likely that
this ability is shared with other species as a result of homology, and does not reflect specific
selective pressures experienced since giraffes diverged from other taxa. Future studies should
address whether giraffes also show further stages of object permanence. Goats, for instance,
have been tested with a similar experiment, showing stage 6 of object permanence (i.e. being
able to successfully track invisible displacements; Nawroth et al., 2015). Given the socio-
ecological challenges faced by giraffes, it would come as no surprise if they could also reach
stage 6.
The Memory conditions tested whether giraffes can use their short-term memory
(Carruthers, 2013) to remember the location of hidden food. Their success in the 30 sec
condition indicates that giraffes can store information for up to 30 seconds later. Their failure
in the 60 sec and 120 sec conditions, however, indicates that their short-term memory might
not last long when compared with other mammals (Lind, Enquist, & Ghirlanda, 2015). Dogs,
for example, can store information for at least 240 seconds (Fiset, Beaulieu, & Landry, 2003),
and cats for at least 60 seconds (Fiset & Doré, 2006). Even in the 30 sec condition, their
accuracy (i.e. 72%) is far from the one shown by chimpanzees (i.e. 86%) with the same delay
and three containers (Barth & Call, 2006). From an evolutionary perspective, however, this
limit in short-term memory is surprising, as giraffes’ fission-fusion social system would
especially benefit from the ability to remember the position of others through time (see Aureli
et al., 2008). Moreover, it is possible that giraffes may show a much better performance when
tested in a long-term memory task. For instance, fights between bulls at first encounters are
common, but these fights create dominance hierarchies and the same bulls rarely fight again
(Shorrocks, 2016; but see Bercovitch & Berry, 2015). Given that adult bulls seldom
Giraffes’ physical cognition
18
encounter each other, this ability to recognize other individuals and remember the outcome of
previous fights may indicate a good long-term memory.
Giraffes used their body orientation as a cue to choose in 52% of their correct trials in
the Object permanence condition, but only in 9% of their correct trials in the 30 sec condition
(see Figure 3). This indicates that, although body orientation may be used as a cue in these
tasks, it is not essential for giraffes to succeed. Also note that the most successful subject,
Ashanti, barely used body orientation as a cue to solve Object permanence and Memory trials
(only in 1 out of 48 trials), confirming that relying on body orientation is no key to success.
In the Acoustic cues condition we tested whether giraffes were able to make
inferences about food position by relying on acoustic cues. Although they performed very
well in the Shake full condition, performance in the Shake empty condition suggests that
giraffes relied on stimulus enhancement (i.e. go toward the container being shaken) rather
than acoustic cues to make their choices, and were not able to infer by exclusion the location
of food. In these conditions, giraffes more heavily relied on body orientation to make their
choice, keeping their head oriented toward the shaken container in 90% of the wrong Shake
empty trials (Figure 3). This may indicate that their attention got trapped by the movement of
the container. However, it is worth noting that this is an especially challenging test. For
instance, not a single chimpanzee out of a sample of 12 performed significantly above chance
levels when an empty container was shaken in an analogous experiment (Call, 2004). Only
with larger sample sizes (e.g. 30 individuals) do primates start to achieve significant results in
these tasks (Call, 2004). Similar patterns have been found in other primate species, being
unusual to find individual performances above chance levels (Maille & Roeder, 2012; Marsh,
Vining, Levendoski, & Judge, 2015; Sabbatini & Visalberghi, 2008). One of the possible
explanations for our outcome is that giraffes have cognitive restraints: although they could
use acoustic cues to locate food when the full container was shaken, inference by exclusion
Giraffes’ physical cognition
19
was cognitively too demanding to allow them success in the Shake empty condition. In
contrast to apes, however, giraffes did not simply select the baited container at chance levels,
in the Shake empty condition, but went for the empty shaken container significantly more
than chance, showing evidence of stimulus enhancement (rather than simply lack of inference
by exclusion). It is therefore likely that, even in the Shake full condition, subjects simply
went for the baited container because it was shaken, and not because it made noise. This
would also be in line with the leaf-based diet of wild giraffes: as leaves produce no sound
when shaken, the ability to associate sound with food may have not been especially selected
for.
Using stimulus enhancement to locate food, of course, need not be the result of
evolutionary pressures, but may depend on our giraffes’ captive condition. Being captive
animals, they could have learned throughout their lives to associate food to human cues (e.g.
shaking), rather/more than to physical cues (e.g. acoustic cues). Regarding this last
explanation, it is usually thought that domesticated species are better able to understand
human communicative cues (e.g. Kaminski, Riedel, Call, & Tomasello, 2005). Our results,
however, would rather suggest that developmental experience may easily allow also wild
animals to rely on human cues to locate food. Indeed, stimulus enhancement is traditionally
considered an early indicator of social learning (Spence, 1937; Thorpe, 1956). Further
experiments are needed to test these different possibilities, by for example testing giraffes’
ability to use human cues, testing their abilities to make inferences by exclusion in the visual
rather than acoustic modality, or providing them with acoustic cues while separately
controlling for stimulus enhancement.
Overall, our results indicate that giraffes can successfully form mental representations
of objects and store them in memory for short periods of time, and likely rely on stimulus
enhancement to locate food, when only acoustic cues are provided. Our results are in line
Giraffes’ physical cognition
20
with recent, perhaps unexpected findings of complex cognitive capacities in other ungulates
(e.g. Briefer, Haque, Baciadonna, & McElligott, 2014; Nawroth et al., 2014; Nawroth et al.,
2015; Marino & Allen, 2017), and converge in suggesting that ungulates might have better
cognitive skills than previously thought. Importantly, our results also show that giraffes are
attentive and motivated enough to become valuable subjects in future cognitive tests. Given
the variety of socio-ecological characteristics across ungulate taxa (see Shultz & Dunbar,
2006), this opens up to the possibility of comparing cognitive performance across ungulate
species, to test evolutionary hypotheses about the emergence of complex cognitive abilities.
Recently, evolutionary hypotheses have been mainly contrasted by comparing performance
across primates and corvids: extending this approach to other taxa may not only provide an
easier alternative, but also allow us to better understand the limits of evolutionary theories
that have so far only been tested across few species. Such investigations may help to
reconstruct the evolution of cognitive skills and to gain a real comprehensive view of their
socio-ecological foundations.
Giraffes’ physical cognition
21
References
Albiach-Serrano, A., Bräuer, J., Cacchione, T., Zickert, N., & Amici, F. (2012). The effect of
domestication and ontogeny in swine cognition (Sus scrofa scrofa and S. s.
domestica). Applied Animal Behaviour Science, 141, 25-35.
Amici, F., Barney, B., Johnson, V. E., Call, J., & Aureli, F. (2012). A modular mind? A test
using individual data from seven primate species. PloS one, 7, e51918.
Aureli F, Schaffner, C. M., Boesch, C., Bearder, S. K., Call, J., Chapman, C. A., Connor R.
C., Di Fiore, A., Dunbar, R. I. M., Henzi, S. P., Holekamp, K. E., Korstjens, A. H.,
Layton, R. H., Lee, P.C., Lehman, J., Manson, J. H., Ramos-Fernandez, G., Strier, K.
B., van Schaik, C. P. (2008). Fission-fusion dynamics: new research frameworks.
Current Anthropology, 49, 627-654.
Barth, J., & Call, J. (2006). Tracking the displacement of objects: a series of tasks with great
apes (Pan troglodytes, Pan paniscus, Gorilla gorilla, and Pongo pygmaeus) and
young children (Homo sapiens). Journal of Experimental Psychology: Animal
Behavior Processes, 32, 239-252.
Bercovitch, F. B., & Berry, P. S. (2013). Herd composition, kinship and fission–fusion social
dynamics among wild giraffe. African Journal of Ecology, 51, 206-216.
Bercovitch, F.B., & Berry, P. S. M. (2015) The composition and function of all-male herds of
Thornicroft’s giraffe, Giraffa camelopardalis thornicrofti, in Zambia. African Journal
of Ecology, 53, 167–174.
Giraffes’ physical cognition
22
Berry, P. S., & Bercovitch, F. B. (2017). Seasonal and geographical influences on the feeding
ecology of giraffes in the Luangwa Valley, Zambia: 1973–2014. African Journal of
Ecology, 55, 80-90.
Briefer, E. F., Haque, S., Baciadonna, L., & McElligott, A. G. (2014). Goats excel at learning
and remembering a highly novel cognitive task. Frontiers in Zoology, 11, 20.
Bugnyar, T., Stöwe, M., & Heinrich, B. (2007). The ontogeny of caching in ravens, Corvus
corax. Animal Behaviour, 74, 757–767.
Cacchione, T., & Rakoczy, H. (2017). Comparative metaphysics: Thinking about objects in
space and time. In: Handbook of Comparative Psychology (J. Call, Ed.), pp. 579-599,
American Psychological Association.
Call, J. (2001). Object permanence in orangutans (Pongo pygmaeus), chimpanzees (Pan
troglodytes), and children (Homo sapiens). Journal of Comparative Psychology, 115,
159 –171.
Call, J. (2004). Inferences about the location of food in the great apes (Pan paniscus, Pan
troglodytes, Gorilla gorilla, and Pongo pygmaeus). Journal of Comparative
Psychology, 118, 232-241.
Carruthers, P. (2013). Evolution of working memory. Proceedings of the National Academy
of Sciences, 110, 10371-10378.
Carter, K. D., Seddon, J. M., Frère, C. H., Carter, J. K., & Goldizen, A. W. (2013). Fission–
fusion dynamics in wild giraffes may be driven by kinship, spatial overlap and
individual social preferences. Animal Behaviour, 85, 385-394.
Giraffes’ physical cognition
23
Chiandetti, C., & Vallortigara, G. (2011). Intuitive physical reasoning about occluded objects
by inexperienced chicks. Proceedings of the Royal Society of London B: Biological
Sciences, 278, 2621-2627.
Dagg, A. I. (2014). Giraffe: biology, behaviour and conservation. Cambridge University
Press, Cambridge.
Dell'Amore, C. (2016, September 8). Africa may have new giraffe species and this could help
protect them. National Geographic.
Fiset, S., Beaulieu, C., & Landry, F. (2003). Duration of dogs' (Canis familiaris) working
memory in search for disappearing objects. Animal Cognition, 6, 1-10.
Fiset, S., & Doré, F. Y. (2006). Duration of cats’ (Felis catus) working memory for
disappearing objects. Animal Cognition, 9, 62-70.
Hill, A., Collier-Baker, E., & Suddendorf, T. (2012). Inferential reasoning by exclusion in
children (Homo sapiens). Journal of Comparative Psychology, 126, 243-254.
Kaminski, J., Riedel, J., Call, J., & Tomasello, M. (2005). Domestic goats, Capra hircus,
follow gaze direction and use social cues in an object choice task. Animal Behaviour,
69, 11-18.
Kruschke, J. K. (2013) Bayesian estimation supersedes the t test. Journal of Experimental
Psychology: General, 142, 573–603.
Kruschke, J. (2014) Doing Bayesian data analysis: a tutorial with R, JAGS, and Stan.
Academic Press, Cambridge.
Giraffes’ physical cognition
24
Leuthold, B. M., & Leuthold, W. (1972). Food habits of giraffe in Tsavo National Park,
Kenya. East African Wildlife Journal, 10, 129-142.
Lind, J., Enquist, M., & Ghirlanda, S. (2015). Animal memory: A review of delayed
matching-to-sample data. Behavioural Processes, 117, 52-58.
MacLean, E. L., Hare, B., Nunn, C. L., Addessi, E., Amici, F., Anderson, R. C., ... &
Boogert, N. J. (2014). The evolution of self-control. Proceedings of the National
Academy of Sciences, 111, E2140-E2148.
Maille, A., & Roeder, J. J. (2012). Inferences about the location of food in lemurs (Eulemur
macaco and Eulemur fulvus): a comparison with apes and monkeys. Animal
Cognition, 15, 1075-1083.
Manger, P. R. (2013). Questioning the interpretations of behavioral observations of
cetaceans: is there really support for a special intellectual status for this Mammalian
order?. Neuroscience, 250, 664-696.
Marino, L., & Allen, K. (2017). The psychology of cows. Animal Behavior and Cognition, 4,
474-498.
Marsh, H. L., Vining, A. Q., Levendoski, E. K., & Judge, P. G. (2015). Inference by
exclusion in lion-tailed macaques (Macaca silenus), a hamadryas baboon (Papio
hamadryas), capuchins (Sapajus apella), and squirrel monkeys (Saimiri sciureus).
Journal of Comparative Psychology, 129, 256-267.
Muller, Z., Bercovitch, F., Brand, R., Brown, D., Brown, M., Bolger, D., Carter, K., Deacon,
F., Doherty, J.B., Fennessy, J., Fennessy, S., Hussein, A.A., Lee, D., Marais, A.,
Giraffes’ physical cognition
25
Strauss, M., Tutchings, A., & Wube, T. (2016). Giraffa camelopardalis. The IUCN
Red List of Threatened Species 2016: e.T9194A109326950.
Nawroth, C., von Borell, E., & Langbein, J. (2014). Exclusion performance in dwarf goats
(Capra aegagrus hircus) and sheep (Ovis orientalis aries). PLoS One, 9, e93534.
Nawroth, C., von Borell, E., & Langbein, J. (2015). Object permanence in the dwarf goat
(Capra aegagrus hircus): perseveration errors and the tracking of complex
movements of hidden objects. Applied Animal Behaviour Science, 167, 20-26.
Pepperberg, I. M., Willner, M. R., & Gravitz, L. B. (1997). Development of Piagetian object
permanence in a grey parrot (Psittacus erithacus). Journal of Comparative
Psychology, 111, 63–75.
Piaget, J. (1954). The construction of reality in the child. Basic Books, New York.
Sabbatini, G., & Visalberghi, E. (2008). Inferences about the location of food in capuchin
monkeys (Cebus apella) in two sensory modalities. Journal of Comparative
Psychology, 122, 156-166.
Seeber, P. A., Ciofolo, I., & Ganswindt, A. (2012). Behavioural inventory of the giraffe
(Giraffa camelopardalis). BMC research notes, 5, 650-657.
Shaw, R. C., Plotnik, J. M., & Clayton, N. S. (2013). Exclusion in corvids: The performance
of food-caching Eurasian jays (Garrulus glandarius). Journal of Comparative
Psychology, 127, 428-435.
Shorrocks, B. (2016). The giraffe: Biology, ecology, evolution and behaviour. John Wiley &
Sons, New York.
Giraffes’ physical cognition
26
Shultz, S., & Dunbar, R. I. M. (2006). Both social and ecological factors predict ungulate
brain size. Proceedings of the Royal Society of London B, 273, 207-215.
Spence, K. W. (1937). Experimental studies of learning and higher mental processes in infra-
human primates. Psychological Bulletin, 34, 806–850.
Thorpe, W. H. (1956). Learning and instinct in animals. Methuen, London.
VanderWaal, K. L., Wang, H., McCowan, B., Fushing, H., & Isbell, L. A. (2014). Multilevel
social organization and space use in reticulated giraffe (Giraffa camelopardalis).
Behavioral Ecology, 25, 17-26.
Watts, D. P., Potts, K. B., Lwanga, J. S., & Mitani, J. C. (2012). Diet of chimpanzees (Pan
troglodytes schweinfurthii) at Ngogo, Kibale National Park, Uganda, 1. Diet
composition and diversity. American Journal of Primatology, 74, 114-129.
Zucca, P., Milos, N., & Vallortigara, G. (2007). Piagetian object permanence and its
development in Eurasian jays (Garrulus glandarius). Animal Cognition, 10, 243-258.
