The impact of moving to a novel environment on social networks, activity and wellbeing in two new world primates.

American Journal of Primatology 73:1–10 (2011)
RESEARCH ARTICLE
The Impact of Moving to a Novel Environment on Social Networks, Activity
and Wellbeing in Two New World Primates
V. DUFOUR1,2, C. SUEUR3,4, A. WHITEN1, AND H.M. BUCHANAN-SMITH5�
1Centre for Social Learning and Cognitive Evolution and Scottish Primate Research Group, School of Psychology,
University of St. Andrews, St. Andrews, Scotland
2Royal Zoological Society of Scotland, Corstorphine Road, Edinburgh, Scotland
3Unit of Social Ecology, Free University of Brussels, Brussels, Belgium
4Primate Research Institute, Kyoto University, Inuyama, Japan
5Scottish Primate Research Group, Psychology, School of Natural Sciences, University of Stirling, Stirling, Scotland
Among the stressors that can affect animal welfare in zoos, the immediate effect of relocation to a novel
environment is one that has received little attention in the literature. Here, we compare the social
network, daily activity and the expression of stress-related behavior in capuchins (Cebus apella) and
squirrel monkeys (Saimiri sciureus) before and just after they were relocated to a new enriched
enclosure. Results showed similar immediate responses to the move in the two species. Both showed a
substantial increase in the time spent resting and spent more time in the highest and ‘‘safest’’ part of
their enclosure after relocation. Both capuchins and squirrel monkeys spent significantly more time
in close proximity to other group members after relocation, compared to before. In squirrel monkeys,
the structure of the social network, which was initially correlated to affiliation, was no longer so after
the move. In capuchins, the network analysis showed that individuals regrouped by age, with the
youngsters who were potentially more affected by stress being in the center of the network. Social
network analysis helped to achieve a more complete picture of how individuals were affected by
relocation. We suggest that this type of analysis should be used alongside traditional methods of
observation and analysis to encompass the most complex aspects of animal behavior in times of stress
and to improve welfare. Am. J. Primatol. 73:1–10, 2011. r 2011 Wiley-Liss, Inc.
Key words: relocation; welfare; capuchin; Cebus; squirrel monkey; Saimiri
INTRODUCTION
During the lifetime of captive primates held in
zoological parks, individuals will generally move
between several different locations, both within and
between establishments. As an example, most brown
capuchin monkeys (Cebus a. apella) belonging to the
European zoos’ population have been relocated to
a different zoo at least once in their lifetime, with
up to five relocations for some individuals [Robert &
Quintar, European Studbook Reports, 2009]. Inter-
estingly, there are few well-documented reports on
how captive primates and other exotic species cope
with major changes in their environment due to
moving [Carlstead et al., 1993 in leopard cats (Felis
bengalensis); Smith et al., 1998; Schaffner & Smith,
2005 in black tufted-ear marmosets (Callithrix
kuhlii); Little & Sommer, 2002 in hanuman langurs
(Presbytis entellus); Clarke et al., 1983 in chimpan-
zees (Pan troglodytes); Goerke et al., 1987 in gorillas
(Gorilla g. gorilla); Matheson et al., 2005 in brown
capuchins]. Many institutions lack the resources to
carry out short or long-term monitoring on beha-
vioral adjustment to new conditions. Also, while
some obvious behaviors (like stereotypical behavior)
are easily recordable, other states of stress might be
less detectable: acute stress-like relocation may
strongly disrupt the social organization of a group
in social species of primates, the extent of which can
be revealed by looking at grouping patterns and
social cohesion disturbances. To evaluate better such
perturbations, traditional methodological tools such
as behavioral data recording can be easily comple-
mented with more recent tools such as social
network analysis. In this paper we combine both
approaches to investigate how behavioral adjustment
Published online in Wiley Online Library (wileyonlinelibrary.com).
DOI 10.1002/ajp.20943
Received 2 August 2010; revised 31 January 2011; revision
accepted 5 February 2011
Contract grant sponsors: Scottish Funding Council; Fyssen
Foundation; Franco-American Commission.
�Correspondence to: H.M. Buchanan-Smith, Behaviour and
Evolution Research Group and Scottish Primate Research
Group, Psychology, School of Natural Sciences, University of
Stirling, Stirling, FK9 4LA, Scotland.
E-mail: h.m.buchanan-smith@stir.ac.uk
r 2011 Wiley-Liss, Inc.

to relocation can be detected in social groups of
primates.
When recording behavioral effects of relocations,
determination of the cause of any changes is proble-
matic as many factors change simultaneously. Social
factors, individual life history events, health status,
capture and transport methods are some of the
variables that will likely contribute to how well the
primate copes with the degree of stress associated
with relocation, together with the scale of changes in
the physical and social environment and husbandry.
These factors can be expected to affect how fast and
how well animals respond and adapt to their new
conditions. Most behavioral activities (locomotion,
vigilance, resting, foraging and playing) are likely to
be affected in response to a stressful event [Poole,
1988; Shepherdson, 1989]. Determining whether
they reflect an enhanced or decreased wellbeing
requires the significant responses to be considered
in relation to each other [Badihi, 2006]. For example,
changes in locomotion in monkeys can result from
an increase in exploratory behaviors, indicating
improved well-being [Bayne, 1989; Buchanan-Smith
et al., 2004; Snowdon & Savage, 1989], but can also
result from abnormal hyperactivity, indicating a
decrease in wellbeing [Poole, 1988]. Several other
behavioral categories are commonly found to increase
in response to stress in several primate species, some
of which are self-directed behaviors, and are termed
displacement behaviors [Maestripieri et al., 1992].
Increases in the frequency of scratching [Troisi et al.,
1991], auto- and allogrooming [Maestripieri et al.,
1992; Schino et al., 1988; Troisi & Schino, 1987], or
scent marking behaviors [Schwartz & Rosenblum,
1980] are often seen in response to stressful events.
New environments can also introduce new
perceived threats and monkeys’ responses to these
can be explored in terms of disruptions of grouping
patterns, use of space and changes in the duration of
vigilance. If there is a perception of threat, monkeys
will increase their level of vigilance and/or seek to
regroup [Dehn, 1990; Lima, 1995; Powell, 1974; but
also see Caine & Mara 1988; Hirsch, 2002; Roberts,
1996]. In normal circumstances, grouping patterns
are not random and depend on the social organiza-
tion, dominance style and number of affiliates [Hinde,
1976; Thierry, 2004]. Social network analysis can be
used to measure social cohesion and grouping
patterns in primates [Kasper & Voelkl, 2009; Sueur
et al., this issue]. In macaques (Macaca spp.), for
example, the more tolerant the dominance style of the
species, the less clustered the grouping pattern. This
holds true during transitory states of the group such
as collective moves [Sueur & Petit, 2008a,b; Sueur
et al., 2009] and illustrates the stability of
the system. Thus, social network analysis can be
implemented to detect the extent to which the
stability of a social group has been disrupted. There
are few reports on the effects of relocation in captive
primates and none, to our knowledge, use social
network analysis to investigate patterns of change.
When looking at group responses, Little and
Sommer [2002] showed that, in the long term,
relocated hanuman langurs increased inter-
individual distances and locomotory behavior, while
decreasing aggression. In common squirrel monkeys
(Saimiri sciureus), increasing space in a previously
familiar enclosure also led to higher inter-individual
distances and an increase in exploratory behavior
[Marriott & Meyers, 2005]. Another study showed
that a familiar pair of black tufted-ear marmosets
[Schaffner & Smith, 2005] were still affected by
relocation several weeks after the event and tended to
seek their partner’s proximity more. These behaviors
are similar to those produced after a predator has
been detected. While organization of individuals is
influenced by affiliative relationships, dominance or
kinship during spontaneous collective movements,
this influence disappears for movements induced by
the presence of a predator. In this case, animals move
faster and in a more cohesive way [Meunier et al.,
2006; Petit et al., 2009; Sueur et al., 2009]. Compara-
tive analyses of inter-individual distances and social
networks can play an important role in understanding
species differences in response to change and help
predict how relocation will disrupt social organization.
In terms of welfare, most studies mainly report
positive results of relocation to more enriched enclo-
sures [Clarke et al., 1983; Goerke et al., 1987; but see
also Howell et al., 2002; for adverse effects of such
relocation], but the very first contact with a new
environment is probably critical in determining how
animals will adapt to it. Surprisingly, behavioral data
collected in the crucial time immediately after a move
are lacking. Here, we recorded the behavior of a group
of 8 brown capuchin monkeys and of a group of 12
common squirrel monkeys that were initially housed in
similar enclosures at the Royal Zoological Society of
Scotland’s Edinburgh Zoo and were then moved
several months later into the new ‘‘Living Links to
Human Evolution’’ Research Centre [Leonardi et al.,
2010; MacDonald & Whiten, 2011], also located in
Edinburgh Zoo. There, both species were housed in
similar conditions with comparable working routines,
diets and familiar keepers. This provided the potential
to study the immediate impact of relocation on welfare
and social networks when only brief transport was
involved and when all other disturbances were kept
to a minimum. The new environment was highly
enriched with live plants and trees forming a canopy.
We recorded how individuals of each species adapted to
this new environment by comparing several behaviors
thought to provide objective measures of adjustment to
the move. Among these behaviors, inter-individual
distances were analyzed using social network analysis
prior to and immediately after the move. We investi-
gated whether the social network became more
cohesive and the influence of social factors.
Am. J. Primatol.
2 / Dufour et al.

METHODS
Subjects
We followed one group of 12 common squirrel
monkeys and one group of 8 brown capuchin
monkeys housed in Edinburgh Zoo in Scotland,
UK. Age and sex for both groups are reported in
Table I. The squirrel monkey group was composed of
eight adults (one male and seven females), and four
juveniles (three males and one female). In this
species, female–female relationships are organized
following hierarchical ranking [Mitchell, 1994;
Mitchell et al., 1991] but males’ integration in the
group remains unclear [Boinski, 1999]. In most
squirrel monkey species, juvenile males emigrate
from the maternal group at adulthood. There is one
resident adult male who can either be central to the
group in the reproductive season or merely tolerated
by the females out of the breeding season [Boinski,
1999]. In common squirrel monkeys, however, males
can also reside in their natal troop while females may
emigrate [Boinski, 1999]. The capuchin monkey
group was initially composed of five adults (two
males and three females), two juveniles (two males)
and one infant (female) born after the initial data
collection and still on the back of her mother at the
time of the move. Due to the low level of interaction
initiated by the infant, she was not included in the
data collection. Capuchin groups are multi-male
multi-female groups with males migrating at adult-
hood [Fragaszy et al., 2004]. In our group the older
female was the mother of all other individuals except
for the alpha male.
General Procedure
The baseline data on both groups were recorded
several months before their move. Their behavior was
recorded again on the first two days of the move to the
Living Links to Human Evolution Research Centre,
also located at Edinburgh Zoo. For reasons not
relevant to this study, two moves took place for the
squirrel monkeys. The first move (October 4, 2008)
occurred just after the initial data collection, the group
being moved to a smaller enclosure for 6 months
before being moved again into the Living Links
Centre. Data analysis reported here includes only
prerelocation data and data collected at Living Links.
Prerelocation data collection: baseline activity
Prior to the beginning of this study, each group
had been housed for more than a year in adjacent
enclosures of equal size in view of each other. For
each species, the enclosure was composed of one
indoor room connected to an outdoor compartment
(3� 4.5� 6 m high) by a 1.5-m long meshed tunnel.
Monkeys were not visible when occupying this
tunnel. Both indoor and outdoor enclosures were
otherwise visible to members of the public. Data
collection was conducted from the outside enclosure
as neither species regularly used their indoor
enclosure, spending most of their time in the tunnel
and in their outside enclosure. Monkeys were free to
use any part of their enclosure at all times.
Data collection was conducted using individual
focal sampling for 3 min. The observer (VD) stood in
front of the outdoor enclosure and recorded beha-
viors on a voice recorder. The behaviors recorded are
described and defined in Table II. Whenever one
individual remained out of sight (i.e. in the inside
tunnel or inside enclosure), the focal was extended
for up to two minutes in order to obtain three
minutes of data. If this criterion was not met, the
focal was discarded and the failed focal repeated the
day after at a similar time of the day. Data were
collected so that each individual was sampled 18
times spread evenly within early and late morning,
and early and late afternoon starting from 9.45 hr to
16.30 hr. Observations were completed over two
weeks, totaling 10 hr 48 m for the squirrel monkeys
(216 focal samples), and 6 hr 18 m for the capuchin
monkeys (126 focal sampling). The order that
individuals were sampled within each period of the
day (early morning and afternoon and late morning
and afternoon) was randomly determined. Each focal
was then transcribed to the software, ‘‘The Observer
Pro’’ (Noldus), to calculate the duration and fre-
quency of behavior and proximity.
Move into Living Links
The capuchins were moved to Living Links
on March 13, 2008, and were released in their
inside enclosure at 10.40 hr. Squirrel monkeys were
TABLE I. Group Composition for Squirrel Monkeys
and Capuchin Monkeys
Sex Born Mother
Squirrel monkeys
Gerda F 1999
Georgette F 1999
Rio (alpha) M 2000
Harlette F 2000
Jasmine F 2001 Gerda’s sister
Lucienne F 2004 Georgette
Lea F 2004 Gerda
Dego M 2006 Georgette
Chavez M 2006 Gerda
Matisse M 2006 Harlette
Toomi F 2006 Jasmine
Capuchin monkeys
Maurice (alpha) M 1971
Lana F 1995
Popeye M 2001 Lana
Santi F 2002 Lana
Sylvie F 2003 Lana
Toka M 2004 Lana
Figo M 2006 Lana
Pedro M 2008 Lana
Am. J. Primatol.
Novelty, Social Networks and Behavior / 3

re-homed a few weeks later on April 24, 2008, at
10.10 hr. The groups were restricted to their indoor
enclosure for several weeks before letting them
explore their outside enclosure. Squirrel monkeys
were housed in a 5.5� 4.5� 6 m high indoor enclosure.
Capuchin monkeys were housed in a 7� 4.5� 6 m
high indoor enclosure. For each species, this main
inside enclosure was connected to a smaller room
(out of sight of the public view). The inside enclosure
was larger and more complex than the earlier
‘‘baseline’’ enclosure. There were several live plants
at ground level, a live tree and dead tree trunks
bolted to the floor, which were connected by natural
branches, vines and rubber lianas, thereby giving
access to the whole volume of the enclosure [see
Leonardi et al., 2010 for further details of housing
and husbandry]. On the day of the move, focal
sampling observations began 15 min after the release
of the first individual. Individual focal samples of
3 min were carried out every 45–50 min totaling 18
individual focal samples spread evenly on the two
first days of the move. There was a total of 216
samples (10 hr 48 m) for the squirrel monkeys and
126 samples (6 hr 18 m) for the capuchins.
Data Analysis
Durations of activity of space occupation and
frequencies of stress-related behavior were tested for
normality using Kolmogorov–Smirnov tests and, as
neither the data nor their log transform were
normally distributed, nonparametric statistics are
used. Data on duration of activities are expressed as
a percentage, as all activities were mutually exclu-
sive, and time out of sight of the observer was thus
set aside. A similar analysis was conducted on the
time spent high in their enclosure (above 4 m high).
Stress-related behaviors were examined by grouping
together scratching, urine washing, defecation,
repeated tail scratching and head tilting in both
species. Scent marking behaviors (anogenital rub-
bing, face rubbing with hands, face rubbing on
substrate) were only seen in squirrel monkeys and
were analyzed separately from the other stress-
related behaviors. The frequency of stress-related
behaviors per minute was also corrected for the time
spent out of sight from the observer. Comparisons of
pre- and post-relocation were examined using the
nonparametric Wilcoxon signed-rank test.
A social network analysis was used to investigate
whether the move affected social proximity (duration
of time spent in contact, or in proximity (within
50 cm) of other individuals) in the groups [Chepko-
Sade et al., 1989; Krause et al., 2007; Sueur & Petit,
2008b; Wey et al., 2008; Whitehead, 2008]. The ana-
lysis was conducted using the software SocProg2.3
[Sueur & Petit, 2008b; Whitehead, 1997, 2007]. We
constructed a social network of preferred proximities
and investigated its general properties, as well as the
role of individuals in the network. We visualized the
network using Netdraw in Ucinet 6.0 [Borgatti et al.,
2002]. Centrality of each individual was calculated
using the eigenvector centrality coefficient, in which
TABLE II. Definition of Behavioral Categories
Recorded, Adapted From Leonardi et al. [2010]
Behavior Definition
Activity
Resting The monkey is either sleeping or in a
state of calmness, with the body
relaxed in a stationary position. Eyes
may be closed or open, but not actively
scanning the environment
Vigilant Sitting or standing, with eyes actively
scanning the surroundings
Slow locomotion Moving, usually walking, with no jumping
or running
Agitated
locomotion
Monkey is moving in relation to its
surroundings: movements are made at
a rapid pace, that is, running speed and
also include jumping and leaping when
there is more than one leap/jump made,
that is, a succession. Is not scored when
playing
Foraging Searching for food, including ground
digging, scanning the environment for
insects or pieces of food, and eating
Playing Monkey engages in high activity
interaction (e.g. chase, rough and
tumble, mock wrestling) with other
individuals. This can include
nonaggressive physical contact, or
occur at a distance, for example,
hopping and running, steep leaps
(almost vertical jumps with minimal
forward locomotion) or swinging by the
feet, while visually checking/
coordinating with play partners
Potential stress indicators
Scratch The monkey repeatedly moves hand or
foot using nails to scrape the skin and/
or fur
Head tilt The monkey angles his/her skull to one
side repeatedly, sometimes alternating
sides
Urinate or
defecation
Self explanatory
Urine washing The monkey urinates on legs or palms of
hand and then rubs urine on lower
body parts and soles of feet
Scent marking Slowly rubbing the anogenital region
or face on substrate, or rubbing the
face with both hands; often follows
sneezing
Others
Proximity Recorded in two categories of (1) being in
physical contact with another
individual or (2) o50 cm from another
individual but not in contact
Space use Time spent above 4 m
Am. J. Primatol.
4 / Dufour et al.

nodes with high eigenvector centrality are either
connected to many other nodes and/or are connected
to nodes that are highly connected themselves.
We then used the modularity method to identify
sub-groups in the social network [Newman, 2004].
We also investigated whether the proximity network
correlated with kinship and age (i.e. the age
differences within a dyad). We then compared the
proximity matrices and the individual centralities
before and after the move to understand how this
change affected social behaviors of individuals. This
was done using Dietz’R matrix correlation tests as
implemented in Socprog2.3 [Whitehead, 1997, 2009].
For each correlation test we performed 10,000
permutations [de Vries, 1993; Hemelrijk, 1990].
The research was approved by the University of
Stirling Psychology Department Ethics Committee,
followed the Living Links Centre research protocols
and adhered to the legal requirements in the United
Kingdom.
RESULTS
Effect of the Relocation on Activity Budgets
Activity changes in squirrel monkeys
In order to assess the effect of the move on the
general activity we compared the duration of
relevant categories of behavior collected prereloca-
tion to similar categories collected on the first two
days postrelocation. In relation to the activity budget
(Fig. 1), the time spent resting increased signifi-
cantly after the move compared to before (from 29.4
to 52.7%: Wilcoxon signed-ranks test, Z5 3.06,
P5 0.002, N5 12). Vigilance did not differ signifi-
cantly (from 26.9 to 21.6%: Z5 1.8, P5 0.07, N5 12),
but slow locomotion had a significantly lower dura-
tion (from 11.2 to 7.8%: Z5 4.48, Po0.008, N5 12)
after the move. Agitated locomotion was not sig-
nificantly affected by relocation (from 4.9 to 4.3%:
Z5 0.86, P5 0.4, N5 12). A significant decrease in
the time spent foraging was observed (from 26.7 to
12.9%: Z5 3.06, P5 0.002, N5 12). The move had no
significant effect on the time spent playing (Z5 0.88,
P5 0.37, N5 12), which was very low both prior
(mean playing duration5 1.0%) and after relocation
(mean playing duration5 0.7%). More time was
spent at a height of 44 m after relocation compared
to before relocation (from 51.6 to 70.9%: Z5 2.75,
P5 0.006, N5 12). Scent marking increased signifi-
cantly (from 0.05/min to 0.12/min: Z5 2.35, P5 0.02,
N5 12). Stress-related behavior rates (Z5 0.88,
P5 0.37, N5 12) remained stable. This remained
the case when considering the scratching rates
separately (from 0.64/min to 0.53/min: Z5 1.02,
P5 0.31, N5 12).
Activity changes in capuchin monkeys
A similar analysis was run for capuchin monkeys
(Fig. 1). The time spent resting increased signifi-
cantly (from 8.6 to 48.4%: Z5 2.37, P5 0.02, N5 7).
There was no significant effect of relocation on the
time spent being vigilant (from 36.8 to 33.5%:
Z5 0.67, P5 0.49, N5 7). This was also the case
for the duration spent in slow locomotion (from 9.6
to 9.1%: Z5 0, P5 1, N5 7). Agitated locomotion had
a significantly shortened duration (from 4.3 to 0.8%:
Z5 2.37, P5 0.02, N5 7). A significant decrease was
also found in the time spent foraging (from 24.1 to
6.5%: Z5 2.37, P5 0.02, N5 7). This was also true
for the time spent playing (from 16.6 to 1.7%:
Z5 2.37, P5 0.02, N5 7). More time was spent
at the highest possible level after relocation, 44 m
compared to prior to relocation (from 39.0 to 89.6%:
Z5 2.37, P5 0.02, N5 7). Stress-related behaviors
were not significantly affected by relocation (from
0.32/min to 0.17/min: Z5 1.86, P5 0.06, N5 7). This
remained the case when considering the scratching
rates separately (from 0.28/min to 0.13/min: Z51.86,
P50.06, N57).
Effect of Relocation on Social Proximity,
Measured Through Social Network Analysis
Social proximity in squirrel monkeys
Proximity matrices were not correlated before
and after relocation (Dietz R-test, R5 0.21; P5 0.07,
Fig. 1. Percentage of time (7SE) spent in each activity for squirrel monkeys and capuchin monkeys prior and after relocation.
Am. J. Primatol.
Novelty, Social Networks and Behavior / 5

N5 12) when considering the mean duration spent
in proximity to others (Fig. 2). This indicates that
relationships between individuals differed after the
move compared to before. Before the move, there was
a significant correlation between proximities and
kinship matrices (Dietz R-test, R5 0.52; Po0.001,
N5 12) that was not observed after the move (Dietz
R-test, R5 0.12; P5 0.17, N5 12). The mean dura-
tion of proximity with another individual was
significantly higher after relocation compared to
before (from 2 sec/min to 4 sec/min; Wilcoxon,
Z5�5.19; Po0.0001, N5 12). This indicates that
all individuals sought proximity with others more
after the move than before. The centrality for
individuals was correlated (Spearman, RS5 0.6;
P5 0.02, N5 12) between before and after reloca-
tion, indicating that those individuals that were the
most ‘‘popular’’ before relocation were still so after-
wards (Fig. 3). Difference in centralities between
individuals did not vary significantly (range5 0.26
prior to the relocation; range5 0.24 after). However,
two individuals among those having average or high
centrality prior to the move decreased their central-
ity after (Rio, the alpha male and Tumi). Five
individuals showed noticeably the reverse pattern.
Associations before or after the move were not a
function of individual age, since age matrices did not
correlate with proximity matrices whether before
(Dietz R-test, R5 0.14; P5 0.1, N5 12) or after the
move (Dietz R-test, R5 0.03; P5 0.4, N5 12). No
clusters could be identified before relocation.
Social proximity in capuchin monkeys
A similar analysis was carried out in capuchin
monkeys (Fig. 2). The matrices of proximity were
correlated between before and after relocation (Dietz
R-test, R5 0.42, P5 0.036, N5 7), indicating that
individuals had similar affiliative relationships with
their conspecifics before and after relocation. Still,
individuals spent more time in proximity of others of
the same age group after the move (Dietz R-test,
R5 0.43, P5 0.02, N5 7) compared to before (Dietz
R-test, R5 0.21; P5 0.17, N5 7). Mean duration of
Fig. 2. Illustration of the social network in squirrel monkeys (above) and capuchin monkeys (below). In squirrel monkeys, asterix
indicate kinship between individuals (�Georgette’s matriline; ��Gerda’s matriline; ���Harlette’s matriline). Individual proximity to kin
in squirrel monkeys is affected by relocation. Also note the distance of the alpha male Rio from the rest of the group after relocation that
also coincides temporally with a nonreproductive period. In capuchin monkeys, all subjects are related apart from the alpha male
Maurice, who is central in the group. Juveniles (Toka and Figo) are marked with an ‘‘a’’.
Am. J. Primatol.
6 / Dufour et al.

proximity was significantly higher after the relocation
compared to before (from 4 sec/min to 21 sec/min;
Wilcoxon, Z5�5.64, Po0.001, N57), showing that
individuals spent more time in close proximity after
the move. Centrality for individuals was correlated
(Spearman, RS50.81, P50.03, N57), meaning that
those individuals who were popular prior to relocation
were still so afterwards (Fig. 3). Difference of
centrality between individuals seems to be a little
lower after the relocation (range50.17) than before
(range50.25). One of the two individuals with lowest
centrality prior to the move increased it after the
relocation (Sylvie), the other individual (Popeye,
second oldest male in the group) remained peripheral
after relocation. The four individuals with the highest
centrality prior to the move decreased it after the
relocation. The alpha male (Maurice) had the highest
centrality after the move. No clusters could be
identified before or after relocation.
DISCUSSION
Relocation impacted on both squirrel and capu-
chin monkeys’ activity budget and behavior. Social
network analysis allowed us to measure how strongly
the grouping patterns were affected in both species.
In squirrel monkeys, the time initially spent in
activities like foraging and slow locomotion became
allocated to resting after the move. Vigilance and
agitated locomotion were expected to increase if the
relocation was stressful, but were not affected. The
monkeys did perform other behaviors we interpret as
functioning to cope with the relocation—they spent
most time at the highest levels of their enclosure,
scent marked more and sought proximity more with
other group members. Capuchin monkeys showed
almost similar behavioral adjustments in response to
the move. Time spent resting increased to an even
higher proportion than in squirrel monkeys while
foraging and playing decreased. There was no
increase in the frequency of stress-related behaviors.
Even more than for squirrel monkeys, capuchins
sought each others’ proximity spending more time
close to each other.
Upon arrival in their new enclosure, both
species faced a large, wooded and canopy-like
environment with lianas and moving platforms. This
new enclosure could itself incorporate unknown
threats to the animals. Unexpectedly, both species
spent most of their time resting. One would at least
expect the monkeys to maintain some level of
vigilance as a result of potential threats. In our
study, vigilance did not increase in either species but
monkeys sought proximity with each other. Either
vigilance was needed less since individuals regrouped
[see Roberts, 1996 for discussion of group size effects
and behavior] or the novelty of the enclosure was not
a major stressor compared to the acute stress of the
catching and transportation in the zoo and there was
no need for more vigilance from the monkeys (‘‘the
danger has passed’’). That few other stress-related
indicators increased suggests the latter and suggests
also that behavioral strategies were turned toward
recuperating from the stress.
The way one experiences stress and recovers
from it differs between individuals, which affects in
turn the social structure of the group. Since most
primates are social species, it is likely that stressful
events have repercussions on how, when and for how
long individuals interact with each other. Social
network analysis is increasingly used in the study of
animal behavior [Croft et al., 2008; Sueur et al., this
issue; Wey et al., 2008] to describe the social
structure of a species [guppies (Poecilia reticulata):
Croft et al., 2004; spider monkeys (Ateles geoffroyi):
Ramos-Ferna´ndez et al., 2009] to compare the social
organization of closely related species [Sueur &
Petit, 2008a,b; Sueur et al., 2010; Sundaresan
et al., 2007], or to investigate animal well-being
[McCowan et al., 2008]. In our study, grouping
pattern was affected in capuchin monkeys, who
spent most of their time in proximity to another
Fig. 3. Eigenvector centrality coefficient of (A) squirrel monkeys and (B) capuchins before (rhombus) and after (square) the relocation.
Individuals are ranked according to their centrality pre-relocation (In squirrel monkeys: 1, Harlette; 2, Gerda; 3, Rio; 4, Georgette;
5, Lea; 6, Mathis; 7, Tumi; 8, Ina; 9, Lucienne; 10, Jasmine; 11, Chavez; 12, Dego; In capuchins: 1, Sylvie; 2, Popeye; 3, Santi; 4, Toka;
5, Lana; 6, Maurice; 7, Figo). Lines represent the relation between individual and centrality and the change after relocation.
Am. J. Primatol.
Novelty, Social Networks and Behavior / 7

individual after the move. The alpha male held a
strongly networked position in the center of this
group. Even if central individuals remained so after
the move, the difference of centrality between
individuals decreased, showing a redistribution of
social behaviors between all partners. Regrouping
after a move has also been described in one study
where a bonded pair of back tufted-ear marmosets
spent an increased amount of time in contact after
relocation compare to before [Schaffner & Smith,
2005]. Stress might encourage individuals to regroup
in response to potential threats, thus limiting the
need for vigilance [Lima, 1995; Robinson, 1981;
Treves, 2000]. Regrouping is also a well-known
strategy to cope with anxiety in many contexts
[Hennessey, 1984; Jordan et al., 1985]. This is one
of the reasons why individuals of social species
should wherever possible be housed with conspecifics
in a captive environment.
In squirrel monkeys, the social network also
became tighter. However, there were two main
differences between the squirrel monkeys and the
capuchins. First, the alpha male became peripher-
alized (while in the capuchins the alpha male was
central). This reflects differences in social structure.
Second, squirrel monkey grouping was less a func-
tion of kinship after the move compared to before,
while in capuchins, individuals regrouped by age
more after the move compared to before. However,
the change in the squirrel monkey grouping pattern
may not be solely due to the stress of the move. The
move into Living Links was the second they
experienced in a year, and the first move might have
had longer term effects on reactivity to stressful
events than was expected. In addition, some of these
effects might also come from the younger individuals
who, more than 6 months after initial data collection,
had become independent from their mothers, thus
having weaker bonds. Since the initial data collection
they may also have increased the quality of their
relationship with other group members. In squirrel
monkeys, however, females are philopatric and kin-
ship bonds remain a determinant of social proximity
[Mitchell et al., 1991]. Random associations, as seems
to be the case here, would support a ‘‘safety in
numbers’’ [Mitchell, 1994] explanation for the dis-
ruption observed in the grouping patterns. In
capuchins, younger individuals became more central
to the group than they were initially, suggesting that
stress may impact more on younger individuals
compared to the older ones, which will affect social
patterns. This study confirms, coupled with classical
measures of stress, the use of social network analysis
as a powerful tool to assess group responses to major
changes in their environment in relation to social
variables.
Network structure in a group may vary over
time [Wey et al., 2008] and may also vary following
the behavioral activity involved: to measure social
influence on the diffusion of a new foraging techni-
que, for example, one may need to build networks
using data on foraging parties instead of resting
parties [Hoppitt et al., 2010]. Social network analysis
can encompass most social situations and therefore
reveal the diversity of strategies available to social
animals. It could also be used to monitor group
adjustment when artificially modifying group struc-
tures (removing or adding new individuals in a pre-
existent social group for example). Our results
promote the generalization of social network analysis
as a tool for monitoring social animal welfare in
captive or non-captive species, especially if data and
recording techniques are shared between institu-
tions. Although this could not be achieved in this
study, long-term recording would benefit long-lived
species like primates and allow receiving institutions
to prepare optimally for newly arrived animals. More
generally, our understanding would gain from
developing the use of this tool further in future
research on animal welfare.
ACKNOWLEDGMENTS
We thank Charlotte MacDonald, Head Keeper
and Research Liaison officer, and her staff at the
Living Links to Human Evolution Research Centre
for careful husbandry of the monkeys before, during
and after the move. The Living Links Centre was
funded by a Strategic Research and Development
Grant from the Scottish Funding Council, awarded to
Professor Andrew Whiten, University of St. Andrews.
Cedric Sueur was funded by the Fyssen Foundation
and the Franco-American commission. The research
was approved by the University of Stirling Psychology
Department Ethics Committee, adhered to legal
requirements in the United Kingdom, and adhered
to the American Society of Primatologists principles
for the ethical treatment of primates.
REFERENCES
Badihi I. 2006. The effects of complexity, choice and control on
the behaviour and the welfare of captive common marmo-
sets (Callithrix jacchus) [Thesis], University of Stirling,
411p. Available from: www.https://dspace.stir.ac.uk/dspace/
index.jsp
Bayne KAL. 1989. Environmental enrichment alternatives
for laboratory nonhuman primates. In: Driscoll JW, editor.
Animal care and use in behavioral research: regulations,
issues and applications. Beltsville, MD: United States
Department of Agriculture. p 91–102.
Boinski S. 1999. The social organization of squirrel monkeys:
implications for ecological models of social evolution.
Evolutionary Anthropology 8:101–112.
Borgatti SP, Everett MG, Freeman LC. 2002. UCINET for
Windows: Software for Social Network Analysis. Harvard,
MA: Analytic Technologies, http://www.analytictech.com
Buchanan-Smith HM, Prescott MJ, Cross NJ. 2004. What
factors should determine cage sizes for primates in the
laboratory? Animal Welfare 13:S197–S201.
Caine NG, Mara SL. 1988. Vigilance and social organization in
two species of primates. Animal Behaviour 36:897–904.
Am. J. Primatol.
8 / Dufour et al.

Carlstead K, Brown JL, Seidensticker J. 1993. Behavioral
and adrenocortical responses to environmental changes
in leopard cats (Felis bengalensis). Zoo Biology 12:321–331.
Chepko-Sade BD, Reitz PR, Sade DS. 1989. Sociometrics of
Macaca mulatta IV: network analysis of social structure of a
pre-fission group. Social Networks 11:293–314.
Clarke SA, Juno CJ, Maple TL. 1983. Behavioral effects of
a change in the physical environment: a pilot study of
captive chimpanzees. Zoo Biology 1:371–380.
Croft DP, Krause J, James R. 2004. Social networks in the
guppy (Poecilia reticulata). Proceedings of the Royal Society B:
Biological Sciences 271:516–519.
Croft DP, James R, Krause J. 2008. Exploring animal social
networks. Princeton: Princeton University Press.
de Vries H. 1993. The rowwise correlation between two
proximity matrices and the partial rowwise correlation.
Psychometrika 58:53–69.
Dehn MM. 1990. Vigilance for predators: detection and
dilution effects. Behavioral Ecology and Sociobiology 26:
337–342.
Fragaszy DM, Visalberghi E, Fedigan LM. 2004. The
complete capuchin. In: Fragaszy D, Perry S, editors.
The biology of the genus Cebus. Cambridge: Cambridge
University Press.
Goerke B, Fleming L, Creel M. 1987. Behavioral changes of
a juvenile gorilla after a transfer to a more naturalistic
environment. Zoo Biology 6:283–295.
Hemelrijk CK. 1990. Models of, and tests for, reciprocity,
unidirectionality and other social interaction patterns at
a group level. Animal Behavior 39:1013–1029.
Hennessey MB. 1984. Presence of a companion moderates
arousal of monkeys with restricted social experience.
Physiology and Behavior 33:693–698.
Hinde RA. 1976. Interactions, relationships and social struc-
ture. Man 11:1–17.
Hirsch BT. 2002. Social monitoring and vigilance behaviour
in brown capuchin monkeys (Cebus apella). Behavioral
Ecology and Sociobiology 52:458–464.
Hoppitt W, Boogert NJ, Laland KN. 2010. Detecting social
transmission in networks. Journal of Theoretical Biology
263:544–555.
Howell S, Schwandt M, Fritz J, Walker S. 2002. From
laboratory to more natural enclosures: maintaining the
well-being of captive chimpanzees (Pan troglodytes).
Laboratory Primate Newsletter 41:5–9.
Jordan TC, Hennessey MB, Gonzalez CA, Levine S. 1985.
Social and environmental factors influencing mother-infant
separation-reunion in squirrel monkeys. Physiology and
Behavior 34:489–493.
Kasper C, Voelkl B. 2009. A social network analysis of primate
groups. Primates 50:343–356.
Krause J, Croft DP, James R. 2007. Social network theory in
the behavioral sciences: potential applications. Behavioral
Ecology Sociobiology 62:15–27.
Leonardi R, Buchanan-Smith HM, Dufour V, MacDonald C,
Whiten A. 2010. Living together: behavior and welfare in
single and mixed species groups of capuchin (Cebus apella)
and squirrel monkeys (Saimiri sciureus). American Journal
of Primatology 71:1–15.
Lima SL. 1995. Back to the basics of anti-predatory vigilance:
the group size effect. Animal Behaviour 49:11–20.
Little KA, Sommer V. 2002. Change of enclosure in langur
monkeys: implications for the evaluation of environmental
enrichment. Zoo Biology 21:549–559.
MacDonald C, Whiten A. 2011. The ‘‘Living Links to Human
Evolution’’ Research Centre in Edinburgh Zoo: A new
endeavour in collaboration. International Zoo Yearbook,
in press.
Maestripieri D, Schino G, Aureli F, Troisi A. 1992. A modest
proposal: displacement activities as an indicator of emotions
in primates. Animal Behaviour 44:967–979.
Marriott BM, Meyers DM. 2005. Effects of altered enclosure
size and substrates on squirrel monkeys (Saimiri sciureus
sciureus) behavior. Contemporary Topics 44:15–19.
Matheson MD, Fragaszy DM, Johnson-Pynn JS. 2005.
Response to novel housing in two groups of captive tufted
capuchin monkeys (Cebus apella). Primates 46:235–240.
McCowan B, Anderson K, Heagarty A, Cameron A. 2008.
Utility of social network analysis for primate behavioral
management and well-being. Applied Animal Behaviour
Science 109:396–405.
Meunier H, Leca JB, Deneubourg JL, Petit O. 2006. Group
movement decisions in capuchin monkeys: the utility of an
experimental study and a mathematical model to explore
the relations between individual and collective behaviours.
Behaviour 143:1511–1527.
Mitchell CL. 1994. Migration alliances and coalitions among
adult male South American squirrel monkeys (Saimiri
sciureus). Behaviour 130:169–190.
Mitchell C, Boinski S, van Schaik CP. 1991. Competitive
regimes and female bonding in two species of squirrel
monkeys (Saimiri oerstedii and S. sciureus). Behavioral
Ecology and Sociobiology 28:55–60.
Newman MEJ. 2004. Analysis of weighted networks. Physical
Review E 70:56131.
Petit O, Gautrais J, Leca JB, Theraulaz G, Deneubourg JL.
2009. Collective decision-making in white-faced capuchins.
Proceedings of The Royal Society of London B 276:
3495–3503.
Poole TB. 1988. Behaviour, housing and welfare of non-human
primates. In: Beynen AC, Solleveld HA, editors. New
developments in biosciences: their implications for laboratory
animal science. Dordrecht: Martinus Nijhoff Publishers.
Powell GVN. 1974. Experimental value of the social flocking by
starlings (Sturnus vulgaris) in relation to predation and
foraging. Animal Behaviour 22:501–505.
Ramos-Ferna´ndez G, Boyer D, Aureli F, Vick LG. 2009.
Association networks in spider monkeys (Ateles geoffroyi).
Behavioral Ecology and Sociobiology 63:999–1013.
Roberts G. 1996. Why individual vigilance declines as group
size increases. Animal Behaviour 51:1077–1086.
Robert JY, Quintar B. 2009. European studbook reports for
Cebus apella. France: Museum of Besancon.
Robinson JG. 1981. Spatial structure in foraging groups of
wedge-capped capuchin monkeys (Cebus nigrivittatus).
Animal Behaviour 29:1036–1056.
Schaffner CM, Smith TE. 2005. Familiarity may buffer the
adverse effects of relocation on marmosets (Callithrix
kuhlii): preliminary evidence. Zoo Biology 24:93–100.
Schwartz GG, Rosenblum LA. 1980. Novelty, arousal, and
nasal marking in the squirrel monkey. Behavioral and
Neural Biology 28:116–122.
Shepherdson DJ. 1989. Environmental enrichment. Ratel 16:
4–9.
Schino G, Ssucchi S, Maestripieri D, Turillazi PG. 1988.
Allogrooming as a tension reduction mechanism: a beha-
vioral approach. American Journal of Primatology 16:43–50.
Smith TE, McGreer-Whitworth B, French JA. 1998. Close
proximity of the heterosexual partner reduces the physio-
logical and behavioral consequences of novel-cage housing
in black tufted-ear marmosets (Callithrix kuhlii). Hormones
and Behavior 34:211–222.
Snowdon CT, Savage A. 1989. Psychological well-being of
captive primates: general considerations and examples from
callitrichids. In: Segal EF, editor. Housing, care and
psychological well-being of captive and laboratory primates.
New Jersey: Noyes publications. p 75–88.
Sueur C, Petit O. 2008a. Shared or unshared consensus
decision in macaques? Behavioural Processes 78:84–92.
Sueur C, Petit O. 2008b. Organization of group members
at departure is driven by social structure in Macaca.
International Journal of Primatology 29:1085–1098.
Am. J. Primatol.
Novelty, Social Networks and Behavior / 9

Sueur C, Deneubourg JL, Petit O. 2009. Selective mimetism at
departure in collective movements of the Tonkean macaque
(Macaca tonkeana): a theoretical and experimental approach.
Animal Behaviour 78:1087–1095.
Sueur C, Petit O, Deneubourg JL 2010. Short-term group
fission processes in macaques: a social network analysis. The
Journal of Experimental Biology 213:1338–1346.
Sueur C, Jacobs A, Petit O, Amblard F, Kin AJ. in press.
How can social network analysis improve the study of
primate behaviour? American Journal of Primatology,
special issue: Social networks in primates. doi: 10.1002/
ajp.20915
Sundaresan SR, Fischhoff IR, Dushoff J, Rubenstein D. 2007.
Network metrics reveal differences in social organization
between two fission-fusion species, Grevy’s zebra and
onager. Oecologia 151:140–149.
Thierry B. 2004. Social epigenesis. In: Thierry B, Singh M,
Kaumanns W, editors. Macaque societies: a model for
the study of social organization. Cambridge: Cambridge
University Press. p 267–290.
Treves A. 2000. Theory and methods in study of vigilance and
aggregation. Animal Behaviour 60:711–722.
Troisi A, Schino G. 1987. Environmental and social influences
on autogrooming behaviour in a captive group of Java
monkeys. Behaviour 100:292–302.
Troisi A, Schino G, D’antoni M, Pandolfi N, Aureli F, D’Amato F.
1991. Scratching as a behavioural index of anxiety in macaques
mothers. Behavioral and Neural Biology 56:307–313.
Wey T, Blumstein DT, Shen W, Jorda´n F. 2008. Social network
analysis of animal behaviour: a promising tool for the study
of sociality. Animal Behaviour 75:333–344.
Whitehead H. 1997. Analysing animal social structure. Animal
Behaviour 53:1053–1067.
Whitehead H. 2007. Programs for analyzing social structure.
http://myweb.dal.ca/hwhitehe/manual.htm
Whitehead H. 2008. Analyzing animal societies: quantitative
methods for vertebrate social analysis. Chicago: The University
of Chicago Press.
Whitehead H. 2009. SOCPROG programs: analysing animal social
structures. Behavioral Ecology and Sociobiology 63:765–778.
Am. J. Primatol.
10 / Dufour et al.