The effect of visitors on stress-related behaviour
of zoo-housed red-faced spider monkeys (Ateles
Paniscus) compared in the inside and outside
compartment of the enclosure
Student
Student number Study
Major
University & Department Supervisor
Assessor and Examiner
Date of submission
Robyn Jansma 10699821
BSc Future Planet Studies Biology
University of Amsterdam - FNWI Peter Roessingh roessingh@uva.nl 020 5257732 Karline Janmaat karline_janmaat@eva.mpg.de +316 19 94 41 79 05 July 2018
Content
Abstract ... 1
1. Introduction ... 1
2. Materials and Methods ... 4
2.1 Study objects ... 4
2.2 Method of behavioral observation... 4
2.3 Visitor variables ... 5 2.4 Statistical analysis ... 5 3. Results ... 6 3.1 Stereotypic behaviour ... 6 3.2 Locomotive behaviour ... 6 3.3 Non-vigilant behaviour ... 7 3.4 Social behaviour ... 7
3.5 Possibly stereotypic behaviour ... 8
4. Discussion ... 9
5. Conclusion ... 11
6. Appendix ... 12
6.1 Study objects specifications ... 12
6.2 Exhibit specifications ... 12 6.3 Diet specifications ... 13 6.4 Ethogram ... 14 6.5 Model specifications ... 16 6.6 Model output... 16 References ... 17
1
Abstract
Animals held in captivity often endure more stress than wild congeners do, which might be harmful for their health and wellbeing. This stress is often expressed by showing stereotypic behaviour. Previous studies showed that visitors can have a negative impact on the stress levels of zoo-housed animals, referred to by the term ‘visitor effect’. The visitor effect is found to be strong in arboreal primate species, such as spider monkeys.
The current study examined the effect of visitors on the stress-related behaviour of red-faced spider monkeys (Ateles Paniscus) housed at ARTIS Amsterdam Royal Zoo. Two visitor variables were assessed; number of visitors and sound volume. Additionally, this study examined whether there was a difference in visitor effect in the inside and outside compartment of the enclosure since these exhibits differ strongly in design.
Behavioural observations were conducted for 3 weeks daily, during which all behaviour displayed by the spider monkeys was noted, along with the number of visitors present, the level of sound (in decibel) and the location of observation.
This study demonstrated that an increase in average level of decibel led to an increase in stereotypic behaviour. Additionally, spider monkeys were found to display more stereotypic behaviour in the outside compartment of their enclosure. However, this study did not find the number of visitors to have an influence on the stress-related behaviour. These results suggest that visitor noise can have a negative effect on the stress levels of red-faced spider monkeys and that this effect is dependent on the exhibit design.
This study hopefully provides more insight into the visitor effect by assessing the influence of two important visitor variables. These insights might improve future designs of zoo enclosures, thus enhancing animal welfare. However, future research is needed to further assess the cause of stress in zoo-housed animals.
Stress – visitor effect – exhibit design – stereotypic behaviour – spider monkeys
1.
Introduction
Zoo animals that live in captivity usually live in an environment that is very different from their natural habitat. Along with this artificial habitat come challenges that are potentially provocative and stressful for the animals. One study by Rangel-Negrin et al. (2009) showed that Yucatan spider-monkeys held in captivity endured more stress than congeners in the wild did.
According to Morgan and Tromborg (2006), stress is being expressed as the so called “fight or flight” response. During this response the animal’s body is preparing, by a cascade of physiological events, to maintain the status quo by either fighting or flighting the stressor. A stressor might be an actual challenge or the threat of a challenge.
Stress can be beneficial since it activates a mechanism that can help the organism survive. However, long-term exposure to a stressor might cause chronic stress. Besides lowering the animal’s welfare, chronic stress might result in health problems or eventually even death (Hoffman & Hercus, 2000).
Animals often express feeling stressed by showing stereotypical behaviour. This type of behaviour is seen as an indication of current discomfort or of a traumatic event in the past and abnormal for a healthy, wild individual. It specifically occurs when an animal is frustrated, stressed,
2 anxious or lacks stimulation (Mason, 1991). Evidence is accumulating that displaying stereotypic behaviour for an animal is a way of ‘coping’ with external stressors, by focussing on the stereotypic behaviour instead of the source of stress. By leading away the attention from the stressor, stress might be reduced (Mason, 1991).
Research on this topic often focusses on primates since behavioural patterns are relatively easy observable in this class of animals. The effect is found to be strongest in arboreal primate species such as spider monkeys. Bernstein et al. (1976) observed wild white-bellied spider monkeys (Ateles
belzebuth) responding to the observers and recorded stereotypical behaviour such as forcefully
scratching their sides with alternately crossed over arm movements, showing aggression towards the observer and shaking -, braking off - and dropping branches while looking at the observer and being noisy. In addition, closely related species such as Howler monkeys and Wooly monkeys were seen to frequently yawn excessively when approached by humans (Bernstein et al., 1976).
A study by Carpenter (1935) similarly describes the vigorous scratching, dropping branches and growling or barking of wild red spider monkeys (Ateles geoffroyi) when being approached by a human. Another study on orangutans showed that when stressed or frustrated, orangutans would scratch and slap themselves, start biting the cage or an object and make sounds towards the observer (Elder & Menzel, 2001). It is furthermore shown in many studies that stress increases locomotor activity, such as pacing and sprinting, in many species of primates (Elder & Menzel, 2001).
In zoos, examples of potentially stressful challenges are abiotic factors such as exposure to loud sounds, provoking odours and undesirable temperatures. Furthermore, movement restriction due to enclosure size, community composition often being socially abnormal and the presence of and forced proximity to visitors are possible stressors (Morgan & Tromborg, 2006).
Several studies show indeed that visitors have a psychologically adverse effect on the animals that live in zoos, referred to by the term ‘visitor effect’. One study by Chamove et al. (1988) on the effect of zoo visitors on the social behaviour of different species of primates showed that their activity and aggression towards con-specifics was enhanced when visitors were present. Also, stereotyped behaviour increased proportional to the number of visitors (Chamove et al., 1988).
Another study on gorillas similarly showed that when visitor density was high, the apes spent less time resting and showed significantly more stereotyped behaviour and aggression towards one another, compared to when visitor density was low (Wells, 2005).
Moreover, a study by Lahm (1981) focussing on spider monkeys showed that when harassment by visitors was high, the monkeys showed less social behaviour. Furthermore, a study by Pérez-Galicia et
al. (2017) on red spider monkeys found that a high visitor density was linked to an increase in
self-directed behaviour such as self-scratching and a decrease in vocalization.
Lastly, Blaney & Wells (2004) found that when decreasing visitor visibility by placing a camouflage screen in front of a gorilla exhibit, the level of conspecific-directed aggression and stereotypic behaviour significantly lowered.
This study examines the effect of visitors and exhibit design on the stress-related behaviour of red-faced spider monkeys (Ateles Paniscus) housed at ARTIS Amsterdam Royal Zoo. The red-red-faced spider monkey is an arboreal species that mainly lives in the canopy of the tropical rainforest in northern South America. They are rarely found in the bottom layer of the forest (Roosmalen, 1980). The species is listed as Vulnerable on The IUCN Red List of Threatened Species (Mittermeier et al., 2008).
This research continues the studies of two other students of the University of Amsterdam, Edwin Sudibyo and Juul Verschuuren. Sudibyo (2017) examined the effect of partly lowering the visibility of visitors on the stress-related behaviour by placing a one-way vision film on the glass window of one of the compartments of the enclosure. Verschuuren (2018) examined the effect of visitor proximity by positioning a fence in front of the inside enclosure forcing visitors to keep a distance of 1 meter, whereas normally visitors are able to touch the glass window inside as well as the fence outside. In the current study, the effects of two visitor variables – visitor number and sound volume – are being examined. Because the enclosures inside and outside differ in design, a difference in the
3 visitor effect might be found. The inside enclosure contains a glass window, while the outside compartment separates the monkeys from the visitors by a fine black grid fence. Therefore, from the inside enclosure, the visibility of visitors is much higher than from the outside compartment. Along with the higher visibility, the sounds of the visitors might sound more hollow and unnatural to the monkeys in the inside compartment.
In the outside exhibit, voices are mixed with natural sounds such as chirping of birds. These differences probably create a difference in interaction with visitors, and therefore it was additionally assessed whether the effects of number and sound volume of visitors differ between the inside and outside enclosure.
Lastly, during this research the effect of visitors on some behaviour considered abnormal, yet not proven stereotypical was studied. These behaviour elements were appointed as possibly stereotypic and the possibility of these being also indicators of stress in spider monkeys was assessed.
This study was carried out in spring and with the temperatures rising, the monkeys were expected to spend more time outside. The one-way vision film was still present during this research, but the fence was removed.
It was hypothesised that in both exhibits an increase in visitor number will cause the spider monkeys to become more stressed. Similarly, when the sound volume rises, the stressful behaviour of the animals is expected to increase.
These effects are expected to be both strongest in the inside compartment since visitors are best visible from here. Additionally, the reverberating voices reaching the inside compartment are expected to be more stressful than the more natural sounding voices outside.
If these hypotheses are to be true, more activity (“locomotion”) and stereotypical behaviour and less non-vigilant and social behaviour are expected to be observed when visitor number and sound volume increase, intensified in the inside compartment.
This study will provide more insight into the visitor effect by assessing the influence of two important variables. These insights might improve future designs of zoo enclosures, thus enhancing animal welfare.
Many zoos claim to be important for the conservation of species, by conserving healthy and fertile individuals. Since health and welfare are inextricably linked, reducing the visitor effect in zoos might promote conservation efforts. Moreover, animal welfare is simply important for ethical reasons. Hopefully this research changes the way people think about zoos and enhances the pressure on zoos to make their animals’ welfare of paramount importance.
4
2. Materials and Methods
2.1 Study objects
For this study, the five red-faced spider monkeys (Ateles Paniscus) of ARTIS were being observed individually. The specifications of the individuals can be found in appendix 6.1.
On March 13, 2018 Shana gave birth to a baby that will cling to her belly for 6 months. The baby monkey is excluded from the study since it is still dependent on its mother.
The monkeys are housed in an enclosure inside the “Kleine Zoogdierenhuis” (Small Mammals house) containing inside and outside compartments of which the maximum height is approximately 3 meters (appendix 6.2). They are fed daily between 11:30-14:00 with a various diet (appendix 6.3). The food is offered in foraging devices, in stone baskets or freely in the enclosure, which varies each day in manner and where it is placed. The enclosure contains wood chips as ground substrate, ropes, static wooden poles, hay, a stuffed animal and a hammock.
2.2 Method of behavioral observation
All observation sessions lasted 10 minutes per individual, during which all behavioral elements carried out by the individual were noted one after another in Excel (Microsoft Corporation, 2010). Simultaneously, the starting time of each elements was noted automatically. Hence, the duration (in seconds) of every particular behaviour could be determined. All elements were noted in the true order of display and could therefore come up multiple times, resulting in a truthful representation of how an individual spider monkey spend its time.
Additionally, the location (inside or outside, appendix 6.2) and group size of the individual was mentioned simultaneously. Group size is defined as the number of con-specifics being within a radius of 1 meter of the observed individual.
Every individual was observed 3 times 10 minutes a day, 3-4 times a week. This was carried out for 3 consecutive weeks, which gives a total of 195 observations or 32,5 hours observational time. Being dependent on the position of the monkeys, 124 of all observations were carried out inside and 71 were carried out outside.
To avoid bias by non-randomly picking an individual that shows remarkable behaviour, the individuals were observed in a fixed order each time; Firstly, the male Mikey, subsequently Shana, Emma, Switi and lastly Priya.
Observations were carried out on the same time as was done by Verschuuren, between 10:00h-14:00h, to be able to compare results. The monkeys were being fed between 12:00h-14:00h, which varied daily. All observations were carried out by a single observer.
To assess the visitor effect on the stress-related behaviour of the spider monkeys, each individual was observed. During an observation all behaviour elements, i.e. a display of behaviour that is defined and reproducible, were noted and sorted in 5 functional categories; non-vigilant behaviour, social behaviour, locomotive behaviour, stereotypical behaviour and possibly stereotypical behaviour. The categories of stereotypic behaviour and locomotive behaviour are expected to be stress-related; the categories of non-vigilant and social behaviour are assumed to be related to tranquility or the absence of stress; the category of possibly stereotypic behaviour will be tested during this research on being stress-related. Appendix 6.4 shows the complete list of behaviour elements in each category, including a description.
5
2.3 Visitor variables
To assess the factors that cause stress to spider monkeys, two visitor variables were measured. Firstly, during an observation, the number of visitors was counted manually. Visitors passing by were not included, only the visitors that stopped and viewed the spider monkeys were considered as a possible influence. At the end of an observation, visitor counts were summed, and used as an indication of crowdedness.
Secondly, the average and maximum sound volume around the enclosure was measured during the observation by a decibel meter (Decibel X: dB, dBA Noise Meter (version 6.2.3), SkyPaw Co. Ltd, iOS 11.4). The average and maximum decibel level during an observation was determined, giving an indication for audible volume.
2.4 Statistical analysis
The data derived from the observations was analysed in RStudio (RStudio Team, 2016) using R version 3.1.2 (R Core Team, 2014). From each observation, the duration of all behavioural elements pertaining to one of the behavioural categories was summed, hence determining the total duration of each category. To analyse these values, a Generalized linear mixed-effects model was created for each category (Bates et al., 2015). The data was assigned to the model as Poisson-distributed since this was visually determined and common for counted data.
For each model, the residuals were visually checked for being normally distributed and independent of fitted values. All models appeared good to use without violating any assumptions.
For specifications of the models used, see appendix 6.5.
As fixed factors, the factors expected being of influence on the stress-related behaviour were chosen (visitors, average level of sound, maximum level of sound and location). All fixed factors except for location were z-transformed for a better fit to the model. An offset term was added to the model, to correct the sums for the time the individual was not visible during the observation.
Additionally, random slopes were added for all fixed factors, which corrects for the different behavioural response of each individual to each of the factors.
The explanatory power of a model was tested using a chi square test for independence, comparing each full model to a null model which lacked all fixed factors. A model would appear as ‘of influence’ when the outcome of the full model differed significantly from the null model.
Subsequently, a chi square test was used to drop each fixed factor alternately and see whether it contributed significantly to the influence of the full model.
For interpretation of the results, a significance level of α = 0.05 was chosen. For specific outcomes of the Generalized linear mixed-effects model, see appendix 6.6.
6
3. Results
All models appeared to differ significantly from the null models, unless stated otherwise.
The total number of visitors counted during one observation ranged from 1 to 58 visitors. The average level of sound ranged from 57 to 85 dB. The maximum level of sound measured was 107 dB.
3.1 Stereotypic behaviour
No significant association was found between number of visitors and the amount of stereotypic behaviour (P = 0.699) and between the maximum level of sound and the amount of stereotypic behaviour (P = 0.341)(appendix 6.6).
There was however a significant positive relationship between the average level of sound and the amount of stereotypic behaviour displayed by the spider monkeys (P = 0.006, coefficient = 0.339)(Figure 1). Therefore, when the average level of decibel increases, the display of stereotypic behaviour increases in the inside compartment as well as in the outside compartment.
Furthermore, a significant influence of location on the amount of stereotypic behaviour was found (P = 0.019). The duration of stereotypic behaviour in each of the two compartments (mean inside = 33.55s, mean outside = 53.80s) clearly shows that in the outside compartment, the monkeys show significantly more stereotypic behaviour than in they do in the inside compartment.
Figure 1. Positive relationship between average sound level and the amount of stereotypic behaviour displayed.
3.2 Locomotive behaviour
Surprisingly, all fixed factors collectively had no significant influence on the amount of locomotive behaviour displayed. The P-value of the full model compared to the null model was extremely high (P = 1), showing that the amount of locomotive behaviour in spider monkeys is influenced by something entirely different than any of these visitor variables.
7
3.3 Non-vigilant behaviour
No significant influence of any of the researched factors on the amount of non-vigilant behaviour was found. However, the association between the average level of sound and the amount of non-vigilant behaviour was almost significant (P = 0.091, coefficient = - 0.142)(Figure 2), so a negative trend can be detected. Therefore, when the average level of decibel increases, the non-vigilant behaviour tends to slightly decrease.
Figure 2. Negative trend of the average level of sound on the amount of non-vigilant behaviour displayed.
3.4 Social behaviour
Similarly, no significant influence of any of the researched factors on the amount of social and possibly stereotypic behaviour was found. However, the positive relationship between the number of visitors and the amount of social behaviour appeared to be almost significant (P = 0.082, coefficient = 0.498). Therefore, there seems to be a trend for which the amount of social behaviour increases when the number of visitors increases (Figure 3).
8
3.5 Possibly stereotypic behaviour
Additionally, the location of observation tends to have an almost significant influence on the possibly stereotypic behaviour (P = 0.075). When looking at the mean duration of possibly stereotypic behaviour in each compartment (inside = 34.80s, mean outside = 47.66s), it is clearly shown that in the outside compartment, the monkeys tend to show more possibly stereotypic behaviour than in they do in the inside compartment.
On each of the behaviour types appointed as possibly stereotypic separate, being ‘glancing’, ‘stretching’, ‘tongue sticking’ and ‘interacting’, none of the visitor variables did have any effect (P > 0.05).
9
4. Discussion
This research attempted to examine the effect of visitor number and sound level on spider monkeys and tested whether these effects were different in the two compartments of their exhibit.
This study found that the average level of sound increases the amount of stereotypic behaviour significantly, consistent with the hypothesis. This result reinforces the previous findings of Larsen et al. (2014), who found that an increase in noise levels induced an increase in koala vigilance, which is an indication of stress.
Similarly, Birke (2002) describes how noisy visitors caused zoo-housed orang-utans to display more abnormal behaviour. Furthermore, a study of Cooke and Schillaci (2007) demonstrated that an increase in noise level significantly increased stereotypic behaviour in gibbons. Therefore, this study clearly suggests that spider monkeys in ARTIS Amsterdam Royal Zoo endure more stress when the average level of sound surrounding their enclosure is high, confirming the general idea in literature that a high level of visitor noise can cause prolonged stress to captive spider monkeys, which might result in health problems (Hoffman & Hercus, 2000) and hence a lower animal welfare.
The findings of this research might give zoos an approach to enhance their animals’ wellbeing by changing enclosure designs or educating its visitors.
Furthermore, the effect of visitors appears to be significantly stronger in the outside compartment than in the inside compartment. This confirms the idea that design of the exhibit might influence the effect of visitors, as stated by Choo (2011).
During this research, observations were conducted either inside or outside, dependent on the location of the spider monkeys. The fact that 124 observations were carried out inside and only 71 outside while the monkeys are free to go where they want, might confirm the finding that the monkeys endure more stress in the outside compartment.
However, it was expected that the unnatural, echoing sound of visitor voices, that characterizes the inside compartment, would increase the adverse effect of visitors on spider monkeys.
A possible explanation for the contrasting findings might be that the glass window of the inside enclosure actually obstructs most of the visitor noise and therefore attenuates the voices, making the effect of visitor noise less adverse here than in the outside compartment.
Moreover, there might be other factors that influence the higher stress levels in the outside compartment, such as a difference in temperature, exhibit composition, the ability to smell odours or the different nature of sounds outside. The latter might be partially caused by a new jaguar exhibit that was constructed only recently near the spider monkeys’ enclosure. These jaguar growls are possibly an additional reason for the monkeys to feel threatened and therefore endure more stress in the outside compartment of their enclosure.
Hence, the higher stress-levels in the outside compartment might be explained by the higher input of stressors, however further research is also needed to identify other stressors to outside-housed spider monkeys.
Surprisingly, the number of visitors appeared to have little effect on the stress levels in spider monkeys, only slightly, but not significantly, increasing the amount of social behaviour displayed.
This confirms the findings of Choo (2011), who also found the visitor number to have little effect on orangutans. However, it is in contrast with other studies on the effect of the number of visitors on primates (Chamove et al. (1988); Wells (2005); Pérez-Galicia (2017); Larsen et al. (2014)).
This finding might be explained by the fact that in this study, visitor numbers were recorded only during the observation, not reflecting the true number of visitors that had visited the exhibit earlier.
Hence, this research was carried out under the assumption that visitor numbers led to immediate behavioural responses of the monkeys.
10 However, a more reliable relationship could be determined when the total number of visitors prior to the observation are known. Future research could utilize an automatic people counting device to test whether a significant relationship is present when exact visitor numbers prior to the observation are incorporated.
Furthermore, this research did not show any influence of all visitor variables on the amount of locomotive behaviour displayed. These findings do not support those of Chamove et al. (1988) who found an increase in locomotion in different species of primates when visitors were present. Additionally, many other studies show that in many species of primates, stress increases locomotor activity (Elder & Menzel, 2001).
On the other hand, Pérez-Galicia et al. (2017) found in their study on spider monkeys (Ateles Geoffroyi) that the change in locomotor activity as a result of the presence of visitors, was strongly dependent on the social position of the individual. However, this finding suggests that the visitor variables do have an effect on some individuals, which should be present in the outcome of the full model since each individual holds a different social position.
Nevertheless, the findings of the current study are questioning the former research suggesting that visitors would have an influence on the amount of locomotor activity displayed, however it cannot provide an alternative suggestion. Future research is required to further examine the effect of the visitor variables and the difference in exhibit design on the locomotor activity in red-faced spider monkeys and to test whether other factors might have an influence on it.
No variable was found to influence the amount of non-vigilant and social behaviour significantly in this study. There were however some trends visible.
Firstly, an increase in the level of sound from visitors tends to decrease the amount of non-vigilant behaviour displayed. These findings support the hypothesis, as well as the findings of Wells (2005), who found in a study on gorillas that a high visitor density caused the apes to spend less time resting. Since this research was based on a data set of n = 195, a larger data set would possibly result in a statistically significant relationship. A negative relationship should be of concern for zoos, since a lack of rest can cause serious health problems (Rangel-Negrin et al., 2009). Therefore, further research on this topic is required.
Secondly, an increase in visitor numbers was found to slightly increase the amount of social behaviour displayed. Although this study hypothesized the opposite and research conducted by Lahm (1981) on spider monkeys also found that when visitor density was high, the monkeys showed less social behaviour, an explanation for this positive trend could be that social interactions between individuals can reduce stress. Therefore, when an individual is stressed, it might start displaying more social behaviour towards conspecifics in order to lower its stress level (Kanopy, 2015). Further research is required to assess the role of social behaviour in captive spider monkeys.
Subsequently, this study suggests a non-significant trend in which the monkeys appear to display more often behaviour elements being possibly stereotypic – glancing at visitors, stretching limbs, sticking tongue or interacting with visitors – when in the outside compartment.
This finding agrees with the finding that the display stereotypic behaviour is most abundant in the outside compartment. However, none of the visitor variables seems to have an influence on the amount of possibly stereotypic behaviour and therefore, there is likely to be another factor not assessed in this study to be of influence on the stress level of spider monkeys.
This finding also suggests therefore that some of these possibly stereotypic behaviour elements are linked to stress. However, when testing each of the four behaviour elements separately, the location of observation appeared not to be of influence on any of them. This might be explained by the fact that each behaviour element apart occurred very rarely and therefore the models did not have enough power to show the relationship. Further research on these specific behaviour elements with larger data sets is therefore needed.
11 This research was carried out with the greatest care; however, future studies should take some of the following limitations into account to be able to generalize findings even better.
For instance, this study focussed solely on the number and sound level of visitors, without considering the level of harassment by visitors.
According to the findings of Lahm (1981), high levels of harassment by visitors cause a decrease in social behaviour, therefore excluding this variable from the current study might have led to neglecting any existing relationships.
Additionally, the findings of this study do not provide information about the individual monkeys’ responses to the variables, although this is accounted for in the model.
However, evaluating these individual responses is required for a better understanding of the visitor effect and influence of enclosure design. Lastly, despite of significant results, the regression coefficients were relatively low, indicating not the strongest relationships. Increasing the number of observations in further research might strengthen these relationships.
5. Conclusion
This study found that visitor noise can have a negative effect on the stress levels of red-faced spider monkeys and that this effect is dependent on the exhibit design. This study also suggests that the number of visitors is likely to have no effect on the stress levels of red-faced spider monkeys but slightly increases the amount of social behaviour displayed. No reason was found in this study to allocate the possibly stereotypic behaviour elements to being stress-related.
This study hopefully provides more insight into the visitor effect by assessing the influence of two important visitor variables. These insights might improve future designs of zoo enclosures, thus enhancing animal welfare. However, future research is needed to further assess the causes of stress in zoo-housed animals.
12
6. Appendix
6.1 Study objects specifications
Table 1. The five study objects with their name, sex, date of birth, zoo code and chip code
Figure 4. Family tree of the five study objects
6.2 Exhibit specifications
name sex date of birth Zoo code Chip code Mikey M 21-1-1999 M07071 95000000261601 Shana F 30-3-2000 M00048 00-0611-BC15 Emma F 25-4-2001 M01060 00-0631-BE5F Switi F 14-2-2010 M10028 00-06DF-3DEC Priya F 1-11-2016 M16102 900-110-000-251-079
Figure. 5 Map of the red-faced spider monkeys enclosure in ARTIS. A, B, C, D, E, F and G are the locations of the enclosure. A, B and C are the inside compartments and E and F are the outside compartments. The door between B and C is always open and is a see-through grid fence. G is an inside compartment for occasional feeding and care by caretakers. Location D is a grid covered lamp above the glass which is not visible to the observer and visitors. Visitors can stand on location a and b. The location of the visitors indoors is a, and outdoors is b. The observations will also take place from location a and b. Location c is a hallway for the caretakers. During the observations a one-way vision film is placed on the glass window of compartment C.
13
6.3 Diet specifications
Table 2. Diet of the study objects with their daily food and additional food which varies per day.
Monday Tuesday Wednesday Thursday Friday Saturday Sunday
Dry food 350 g Primacon + 200 g leafeater
Vegetables + fruit 1/2 zucchini
1/2 paksoi 1/2 crop iceberg lettuce
Rosettes of 1 broccoli 1 and 1/2 tomato 1 capsicum 1/2 of a papaya a couple of grapes 2 or 3 bananas 1 carrot 1 crop chicory leaves 1 crop endive lettuce Additional food gum walnuts
hazelnuts boiled meat
Cat food Nutrilon (covering the fruit)
Cat food Insects
14
6.4 Ethogram
Category Abbreviation Behaviour Description
So
cial
Bg Com Being groomed CommunicatingIndividual is being groomed, and does not groom other individual; it is not reciprocal
Individual communicates with another individual by making soft sounds
Cu Cuddling Individual holds or is being held with the arms wrapped around the other
Gr Grooming Individual picks fur off other individual for cleaning Ma Mating Individual mates with other individual (Mikey and Shana)
Pl Playing Individual exhibits playful behaviour with or without another individual; grabbing tail, biting, pushing, wrestling
Non
-
vi
gilant
E L N R Eating Lying Neutral Resting Individual is eatingIndividual is lying down on the ground or on a branch Individual does not show any typical behaviour
Individual is either sleeping, relaxing or resting with its eyes closed
Lo
comoti
ve
M Sp Moving SprintingIndividual is moving (walking, climbing, swinging) through the enclosure, i.e. showing activity
Individual suddenly runs very fast without showing an obvious goal or purpose
Stereotyp
ic
Ag Aggressive Individual shows aggressive behaviour towards visitors or con-specifics
Bo Biting object Individual sets teeth in object, e.g. branch or rope Bs Gt Nt Biting self Gnashing teeth Neck twisting
Individual puts mouth around, and sets teeth in own arm or leg or other part of body
Individual opens mouth, showing bottom and top teeth touching each other, sometimes grinding
Individual lets head hang from neck, usually tilted backwards, and twists neck in circular motion
15 Pa Sc Sh So Y Pacing Scratching Scratching head Sounds Yawning
Individual moves around in fixed pattern, little peripheral awareness, locomotion and route show no obvious goal or purpose
Individual scratches excessively own body using fingers Individual scratches his head using an object, not its fingers Individual vocalizes
Individual opens mouth wide, followed by (what appears as) inhalation and exhalation of breath
Po
ss
ibly
Ster
eotyp
ic
Gl I St Ts Glancing Interacting Stretching Tongue stickingIndividual glances towards something outside of the enclosure without special interest
Individual interacts with visitor(s), e.g. standing in front of the glass and looking towards visitors, or pursing lips into a round circle opening, sometimes making noises
Individual stretches one of its limbs Individual sticks out tongue
Table 3. Ethogram of the behavior observations containing the social, non-vigilant, locomotive, stereotypic and possibly stereotypic behaviour.
16
6.5 Model specifications
Figure 6. Generalized linear mixed-effects model.
The Category_sum represent the duration of all behavioural elements pertaining to a particular behavioural category summed for each observation in seconds. The 5 categories are social, non-vigilant, locomotive, stereotypic and possibly stereotypic behaviour.
The fixed factors are the factors expected being of influence on each behavioural category (visitors, average level of sound, maximum level of sound and location).
The offset term corrects the Category_sum for the time the individual was not visible during the observation.
A random slope was added for all fixed factors, to correct for the different behavioural response of each individual to each of the factors.
The dataset was called spidermonkeys and each Category_sum was visually determined as being Poisson-distributed.
6.6 Model output Stereotypic behaviour Locomotive behaviour Non-vigilant behaviour
Social behaviour Possibly stereotypic behaviour Visitors 0.698964029 1 0.2200908 0.08240941 0.3426603 Location 0.019449733 * 1 0.1136284 0.61185514 0.0751997 Average decibel 0.006484092 ** 1 0.0908387 0.75071914 0.5875534 Maximum decibel 0.340600011 1 0.7618972 0.23518987 0.1742127 Figure 7. Results from the Generalized Mixed-Effect models. Significant values are in bold, almost significant values are in
italic.
Visitors, Location, Average decibel and Maximum decibel depict the fixed factors that were expected being of influence on each of the stress-related behaviour categories of spider monkeys.
A chi square test for independence was used to drop each fixed factor from a full model alternately and see whether it contributed significantly to the influence of the full model. A significance level of α = 0.05 was chosen.
Fixed factors Poissonmodel = glmer (Category_sum ~ z.Visitors + Location + z.Av_Decibel + z.Max_Decibel +
Offset term offset(log(Visible/Total_Time) +
Random slope (1+ z.Visitors + Location + z.Av_Decibel + z.Max_Decibel ||Individual),
17
References
Altmann, J. (1974). Observational Study of Behavior: Sampling Methods. Behaviour, 49(3/4), 227-267. Bates, B., Maechler, M., Bolker, B. & Walker S (2015). Fitting Linear Mixed-Effects Models Using lme4.
Journal of Statistical Software, 67, 1-48.
Bernstein, I. S., Balcaen, P., Dresdale, L. Gouzoules, H., Kavanagh, M., Patterson, T. (1976). Differential Effects of Forest Degradation on Primate Populations. Primates, 17(3), 401-411.
Birke, L. (2002). Effects of browse, human visitors and noise on the behaviour of captive orang utans,
Animal Welfare, 11(2), 189-202.
Blaney, E. C., Wells, D. L. (2004). The influence of a camouflage net barrier on the behaviour, welfare and public perceptions of zoo-housed gorillas. Animal Welfare, 13(2), 111-118.
Carpenter, C. R. (1935). Behavior of Red Spider Monkeys in Panama. Journal of Mammalogy, 16(3), 171-180.
Chamove, A.S., Hosey, G.R. & Schaetzel, P. (1988). Visitors Excite Primates in Zoos. Zoo Biology, 7, 359-369
Choo, Y., Todd, P. A., Li, D. (2011). Visitor effects on zoo orangutans in two novel, naturalistic enclosures, Applied Animal Behaviour Science, 133, 78-86.
Cooke, C. M., Schillaci, M. A. (2007). Behavioral responses to the zoo environment by white handed gibbons, Applied Animal Behaviour Science, 106(1-3), 125-133.
Elder, C. M., Menzel, C. R. (2001). Dissociation of Cortisol and Behavioral Indicators of Stress in an Orangutan (Pongo pygmaeus) During a Computerized Task. Primates, 42(4), 345-357.
Hoffman, A.A. & Hercus, M.J. (2000). Environmental stress as an evolutionary force. BioScience, 50(3), 217-226.
Kanopy (Firm). (2015). Understanding Psychological Stress.
Lahm, S.A. (1981). The effects of public density and harassment on primate behavior at the San Diego Zoo. (unpublished manuscript).
Larsen, M. J., Sherwen, S. L., Rault, J. (2014). Number of nearby visitors and noise level affect vigilance in captive koalas. Applied Animal Behaviour Science, 154, 76-82.
Mason, G. J. (1991). Stereotypies: a critical review. Animal Behaviour, 41, 1015-1037.
Mittermeier, R.A., Rylands, A.B. & Boubli, J. P. (2008). Ateles paniscus. The IUCN Red List of Threatened Species 2008: e. T2283A9392691.
18 Morgan, K. N., Tromborg, C. T. (2007). Sources of stress in captivity. Applied Animal Behaviour Science, 102, 262-302.
Pérez-Galicia S, Miranda-Anaya M, Canales-Espinosa D, Muñoz-Delgado J. Visitor effect on the behavior of a group of spider monkeys (Ateles geoffroyi) maintained at an island in Lake Catemaco, Veracruz/Mexico. Zoo Biology. 2017; 36: 360–366.
Rangel-Negrin, A., Alfaro, J.L., Valdez, R.A., Romano, M.C. & Serio-Silva, J.C. (2009). Stress in Yucatan spider monkeys: effects of environmental conditions on fecal cortisol levels in wild and captive populations. Animal Conservation, 12, 496-502.
R Core Team (2014). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/.
RStudio Team (2016). RStudio: Integrated Development for R. RStudio, Inc., Boston, MA URL http://w ww.rstudio.com/.
Van Roosmalen, M.G.M. (1980). Habitat preferences, diet, feeding strategy and social organization of the black spider monkey (Ateles paniscus paniscus Linnaeus 1758) in Surinam. Rijksinstituut voor
Natuurbeheer te Leersum.
Wells, D.L. (2005). A note on the influence of visitors on the behaviour and welfare of zoo-housed gorillas. Applied Animal Behaviour Science, 93, 13-17.
