Environmental enrichment for captive capuchin monkeys ( Sapajus spp . ) using natural material

Captive environments are relatively less complex as compared to wild, and consequently contain less stimuli for any animal within them, which may be stressful. One approach to mitigate stress in captive animals is the application of environmental enrichment techniques. Capuchin monkeys are among the most common primates in captivity, therefore, evaluating the efficacy of environmental enrichment techniques for these species is important. In this study, we evaluated the effects of natural material as environmental enrichment for Sapajus spp. We performed a comparative study of the behavior of seven individuals when their enclosure was non-enriched and enriched with bamboo, leaf-litter and bromeliads. Our results suggested that activities such as body care and affiliative interactions were less frequent when environmentally enriching objects were placed in their enclosure. They also spent more time in object manipulation, and monitoring behavior significantly reduced during these periods. Of the three materials used, we consider bamboo the most efficient enrichment material, since it remained in the enclosure longer and so attracted capuchin attention for a more extended period, and significantly modified their behavior. Key-words: welfare, captivity, Sapajus apella, Sapajus libidinosus, Sapajus nigritus. Resumo. Enriquecimento ambiental para macacos-prego (Sapajus spp.) usando materiais naturais. Ambientes de cativeiro podem ser estressantes aos animais uma vez que são relativamente menos complexos que ambientes naturais e contém menos estímulos. Uma das abordagens para reduzir o estresse de animais em cativeiro é a aplicação de técnicas de enriquecimento ambiental. Macacos-prego estão entre os primatas mais comumente encontrados em cativeiro. Portanto, é importante avaliar a efetividade de técnicas de enriquecimento ambiental para essas espécies. Neste estudo tivemos como objetivo estimar a eficiência de materiais naturais como enriquecimento ambiental para espécies do gênero Sapajus. Para tanto, comparamos o comportamento de sete indivíduos quando submetidos à adição de bambu, folhiço e bromélias em seus recintos. Diante dos enriquecimentos ambientais, os macacos-prego reduziram a frequência de atividades de cuidado corporal e interações afiliativas. Eles também aumentaram a frequência de manipulação de objetos e reduziram a frequência de monitoramento do ambiente durante os períodos com enriquecimento. Concluímos que o bambu foi a ferramenta mais eficaz de enriquecimento, pois permaneceu mais tempo intacto no recinto e com isso atraiu a atenção dos macacos-prego por mais tempo, modificando de forma significativa seu comportamento. Palavras-chave: bem-estar, cativeiro, Sapajus apella, Sapajus libidinosus, Sapajus nigritus. ISSN 1517-6770 Volume Especial Bem-estar Animal Revista Brasileira de Zoociências 19(2): 47-58. 2018


Introduction
Animals may be kept in captivity for a variety of reasons, which includes medical researches (Mittermeier et al., 1994;Bailey, 2005), entertainment, education (Hyson, 2004), rescue (Tribe & Brown, 2000), and conservation (Kleiman et al., 1986).The maintenance of wild animals in captive environments as "sanctuaries", breeding colonies and zoos is an important tool for both species conservation and human health (Bailey, 2005).However, it can compromise the welfare of the captive individuals, since the enclosures usually differ markedly from native habitat in structure and composition.Captive environments are always smaller than the areas animals would naturally occupy, and consequently contain fewer stimuli, and this may be stressful for animals (Ross et al., 2009).Many captive environments also contain other sources of stress such as noise or husbandry routines (e.g.handling, nursing, breeding) (Morgan & Tromborg, 2007;Dickens & Benley, 2014;Quadros et al., 2014).Stress induced by these factors can compel the species to modify their behavior and activity budgets to adapt into the environment (Young, 2003;De Azevedo, 2007;Novak et al., 2017;De Almeida et al., 2018).Among behavioral changes, stereotypies are key for eavesdropping animal welfare.Stereotypies are expressed as repetitive motor behaviors without an apparent purpose and are commonly considered to be indicators of stress (Mason & Rushen, 2008).
Institutions that keep captive animals are responsible for their well-being (Bloomsmith et al., 1991;Waza, 2005).As an effort to reduce negative impacts, a variety of forms of environmental enrichment have been attempted (Young, 2003).Environmental enrichment improves the environments of captive animals, with the aim of promote their mental and physical well-being.Such initiatives generally fall into such categories as "food-based", "structural", "sensory", and "cognitive" enrichment (Newberry, 1995;Clark, 2017).These activities aim to increase the possibilities that captive animals express their motor, exploratory and cognitive abilities in ways that most closely resemble their natural forms (Young, 2003).Such enrichment has great potential to promote physiological and mental well-being in captive animals as it can substantially reduce stress levels (Mellen & Macphee, 2001;De Almeida et al., 2018).Even though the application of environmental enrichment is now widely used, systematic testing of its effectiveness is rarely conducted (Mellen & Macphee, 2001).
Capuchin monkeys, primates from the genus Cebus and Sapajus, are commonly kept in captivity (Lynch-Alfaro et al., 2012, 2014).They are known by their high behavioral flexibility, intelligence, and memory (Fragaszy et al., 2004).They are very active and spend half of their days foraging and feeding (Ross, 1988;Fragaszy et al., 2004).Their social groups can be as large as 35 individuals (Ludwig, 2005), so they usually use large areas that may reach 800 ha (Spironello, 2001).Within this, they may travel as much as ~3 km daily (Edwards et al., 2017).Therefore, when such primates are kept in captivity, it is highly desirable to provide activities that simulate those they commonly perform in the wild (Fragaszy, 2005).One method is to introduce into the captive enclosure natural materials with which species would normally have contact in the wild.However, any such materials must be nontoxic, innocuous, and preferably low-cost (Bloomsmith et al., 1991;Boinski et al., 1999;Costa et al., 2018).
Bearing this in mind, the objective of the current study was to estimate the efficiency of three kinds of environmental enrichment 1) leaf-litter; 2) bromeliads; 3) bamboo, with the addition of earthworms in all treatments.We choose this material because capuchins are known to eat bromeliads and bamboo in the wild (Fragaszy et al., 2004), and descend to the ground to forage for animal resources (Torralvo et al., 2017).We tested if the provided material has induced any behavioral changes in capuchins by observing the animals before and after enrichment introduction.We predicted that during environmental enrichment the individuals would: 1) Increase foraging and feeding activities; 2) Reduce resting time; 3) Decrease in agonistic and increase in affiliative interactions; 4) Increase the behaviors "Monitoring" and "Object manipulation"; and 5) Reduce stereotyped behavior.By testing in this manner, we also aimed to generate basal guidelines for the choice of environmental enrichments suitable for captive capuchin monkeys that are natural, generate low waste of products and are cost-effective.

Study site
The study was carried out in the São Bráz Conservationist Sanctuary (IBAMA: 2/43/96/0001-0) located in the Santa Maria municipality of Rio Grande do Sul state, Brazil (29°41'50.89"S;53°55'13.57"W).The sanctuary is approximately 5 ha in area and consists of 75 enclosures which inhabit some 400 exotic and native animals.Visitation is prohibited to the general public but schools and college groups are allowed.

Subjects and housing
We studied seven adult individuals: one male and one female Sapajus libidinosus Spix 1823, two female S. apella Linnaeus 1758, two males and one female S. nigritus Goldfuss 1809.They were housed in the same enclosure (4.30 m x 6.20 x 3.5 m).Animals were fed every morning (09:00-10:00) with fruits and eggs, with drinking water provided ad libitum.The enclosure, placed in an open field, contained trunks, ropes, a shelter, and small bushes around it with which the capuchins could have physical contact.

Data collection and analysis
We recorded capuchins behavior for twenty five consecutive days, for a total of ~206 hours of observation between December of 2010 and February of 2011.Of these, some 55.5 hours constituted the control phase, before any environmental enrichment had been placed inside the enclosure.The remaining period was spent observing the capuchins' behavior following environmental enrichment.We used as material (1) leaf-litter, (2) bromeliads (Bromeliaceae), ( 3) bamboo (Poaceae) pieces with holes.Within each enrichment, some 25 earthworms (Eisenia andrei) were hidden daily.All of the materials were natural and were available in the sanctuary gardens, except the bromeliads that were purchased commercially (price ~US$ 3.63/unit), which resembled with those found in the animals' native forest habitat.Enrichment materials were inserted in the enclosure early in the morning before the beginning of behavioral observations, and were replaced daily, if necessary.Remnants were removed at the end of sampling, and replacement material subsequently introduced.

Behavior category Description
Feeding Individuals introduce food items into their mouth and ingested them.
Foraging Individuals moved on the trunks, branches, leaves, structures in the enclosure and ground in an apparent search for food sources (such as insects).
Locomotion Individuals moved either on the ground or on aerial structure (trunks and fences) by jumping, running, climbing up or down.
Resting Individuals remained still seated, lying or sleeping.
Body care An activity where animals clean their own pelage (autogrooming) or other body part (nose, ears, anus).

Affiliative interactions
Social interaction directed at other individuals with "friendly" intent, such as allogrooming, playing, copulating.

Agonistic interactions
Social interaction directed to other individuals with an aggressive approach such as piloerection, teeth threat display, chasing or attacking.

Alert
Individuals interrupted its activity and remains motionless with an erect posture and eyes fixed at a point for a few seconds.
Object manipulation Individuals hold or put in the mouth objects without apparent intent of feeding.

Monitoring
Individuals stare at a given item in the environment for a few seconds.
.51 1).In order to compare capuchin behavior with different kinds of environmental enrichment and control period, we calculated the frequency of each behavioral category.Each category was analyzed separately.The duration of observation periods for each environmental enrichment category were not identical (control: 55.5 hours; leaf-litter: 51.4 hours; bromeliads: 51.3 hours; and bamboo: 48.3 hours), so we parameterized the data by dividing the frequency of each behavior by the total number of scans for each period of observation (morning or afternoon).As our data was non-parametric, we used a Kruskal-Wallis to compare all treatments and Dunn post hoc test compare pairs of categories.All analysis were run on the R 3.4.0program (R Core Team, 2016).The study was non-invasive and complied with appropriate Brazilian laws.

Results
Overall, capuchin monkeys manipulated all of the inserted environmental enrichment, thus the treatments were efficient in stimulating the animals.The frequency of most behavioral categories did not differ between pre-and post-enrichment, and this was true for all enrichment types (Tab.2).We also did not corroborate most of our hypotheses.However, we find that body care decreased in the presence of leaf-lit- Monitoring was also smaller during bamboo enrichment compared to bromeliads (Dunn test: H=2.3666; p=0.0090).Following the provision of all three enrichment types object manipulation increased (leaf-litter: H=3.8403; p=0.0001; bromeliads: H=2.7962; p=0.0026; and bamboo: H= -4.9231; p<0.0001) (Fig. 1).Surprisingly, the category "others" was lower in control (Dunn test: H=2.28766; p=0.0111) and bamboo (Dunn test: H=3.1894; p=0.0007) when compared to its frequency with leaf-litter.

Stereotyped behavior
Repetitive movement without apparent function (e.g.moving in circles, head and body twirl).
Out of sight Animals were in the shelter and the observer could not record the behavior during the scan.

Others
Activities that did not fit any of the other categories.This included such activities as defecating and scratching.

Discussion
The capuchin monkeys were attracted to all the materials used for environmental enrichment, since they are very curious and explorative species.Though, most of our hypotheses were not corroborated.The capuchin monkeys did not spend much time in foraging and feeding activities, as we expected because of the addition of earthworms (animal protein) in the enrichment.There are two possible reasons for this: 1) the number of added earthworms was too low (25/day) or 2) capuchins did not consider earthworms as relevant food sources.We recorded multiple times that capuchins manipulated the earthworms but did not eat them.This result is not surprising, as capuchins have rarely been recorded eating earthworms in the wild (but see Barnett et al., 2002).Also, whenever in contact with novel food capuchins usually explore it but remain cautious about ingesting it (Sabbatini et al., 2007).
We expected that resting would decrease during enrichment due to the insertion of elements inside the enclosure which elicited attention and object manipulation.However, this hypothesis was not corroborated.They spent less time resting during the control period as well, which may be related to the typically active behavior of capuchins (Ross, 1988;Zhang, 1995).Even if the enclosure size may affect locomotion rates (Webb et al.,2018), we believe that this result is mostly related to the fact that wild capuchins have high foraging and traveling rates, as well as low rates of resting and social interactions (Rímoli et al., 2008).Even in captivity capuchins can spend much of their energy in vertical (~45 m/hr) and horizontal (541 m/hr) travel throughout the day (Edwards et al., 2017).We also expected a decrease in agonistic and an increase in affiliative interactions.Although we saw the opposite trend for affiliative interactions, and no difference was found in the frequency of agonistic interaction.Besides that, we believe it is unlikely that it would produce any remarkable change in the already established social system of the group.Such findings would also be positive, once the enrichment stimulate the animals without disrupting their established social system (Westergaard & Fragaszy, 1985).
Capuchins possess social learning and show object manipulation both in nature and in captivity (Ottoni & Izar, 2004).Indeed, as expected, object manipulation increased during enrichments and the animals often manipulated the introduced material.Our results coincide with those found by Westergaard & Fragaszy (1985) who reported an increase in object manipulation by capuchins when their enclosure was enriched.Such activities can be helpful in buffering stress related hormones like corticosterone (Boinski, 1999).Boinski et al. (1999) found that Sapajus apella decrease body care activities during enrichment.Likewise, macaws are known to reduce body care (preening) during environmental enrichment (De Almeida et al., 2018).
Studies show that capuchin monkeys in captivity can invest 22 % of their time budget in monitoring the environment (Ferreira, 2017).Accordingly, we expected that frequency monitoring behavior would increase, once new items were inserted into the enclosure, and capuchins tend to be curious with novelty in its environment.However, we found a decrease in this activity, particularly during bamboo enrichment.This is probably linked to the fact that, during bamboo enrichment, the animals were more likely to be engaged in object manipulation (an activity that increased during this phase).
Stereotyped behavior occurred at very low frequency (1.946 %), even during the control phase (without enrichment).This is probably why we detected no changes in the general pattern of stereotyped behavior over the course of the study.There is good and widespread evidence that behaviors potentially indicator of stress occur in captive animals in general (Mason & Rushen, 2008).Such behaviors are often related to individual personality in capuchins (Ferreira et al., 2016).In our study only one animal, a female S. libidinosus, showed any form of stereotyped behavior.
Of the three enrichments deployed, leaf-litter, bromeliads and bamboo, the third appeared to be the most effective, since it induced changes in all of the four behavioral categories mentioned above.Additionally, such material remained in the enclosure for much longer than bromeliads and leaf-litter, materials that were both quickly destroyed or discarded by capuchin monkeys.Bamboo also had positive effects for enriching enclosures for a number of Old World primates, including gibbons, mona monkeys and brown lemurs (Costa et al., 2018).Our findings reinforce such results by showing that bamboo is also an effective enrichment for Neotropical primates.However, we caution that bamboo may be a dangerous material if the levels of aggressive behavior in the group is high, in this case leaf-litter and bromeliads enrichments would be more appropriate.
In general, dry leaves, bamboo and bromeliads with the addition of small invertebrates provide effective environment enrichment for Sapajus monkeys.The introduced material induced motor activities of captive animals that, in the wild, are extremely active.Thus, we strongly recommend the use of such items as environmental enrichment for captive capuchin monkeys.We also found that bamboo was the most efficient enrichment tool since it remained intact for longer in the enclosure and induced significant positive changes in the capuchins behavior.

Figure 1 .
Figure 1.Behavior rate (frequency of behaviors/number of scans) of A) Body care; B) Affiliative interaction; C) Monitoring and; D) Object manipulation by seven captive Sapajus spp.individuals, when provided with different kinds of environmental enrichment: Bromeliads, Leaf-Litter, Bamboo and Control (this final category indicating an absence of environmental enrichment).

Table 2 .
Results from Kruskall-Wallis analysis for each behavioral category.
* significant p-values