Content uploaded by Gale A Kleven
Author content
All content in this area was uploaded by Gale A Kleven on Sep 29, 2020
Content may be subject to copyright.
Vol 55, No 2
March 2016
Pages 161–167
Journal of the American Association for Laboratory Animal Science
Copyright 2016
by the American Association for Laboratory Animal Science
161
The Hairless strain of guinea pigs (Cavia porcellus) was rst
discovered at the Institute Armand Frappier (IAF) in 1978. Un-
like most other nude rodents, Hairless guinea pigs are euthymic
and have a functioning immune system.2 This animal model
has been used extensively in dermatology research6,10,18,22 and,
more recently, to examine behavioral development.4,7 Despite
the strain’s availability for several decades, studies investigat-
ing the temperature preference or thermoregulatory behavior
of the Hairless strain are unavailable. This lack of information
is problematic for behavioral studies, because temperature is a
mitigating factor for most mammalian behavior.23
More importantly, ambient temperatures too divergent from
an optimal thermoneutral range can have long-term detri-
mental effects, particularly during development. For example,
pregnant Hartley guinea pigs housed at 15 °C exhibit a greater
stress response to handling than do controls housed at typical
temperatures (22 °C).16 In addition, offspring from these cold-
condition pregnancies show elevated stress responses 2 mo
after birth, demonstrating that the temperature-evoked stress
response is transgenerational and possibly persistent.17 In con-
trast, outcomes in offspring from elevated ambient temperatures
during gestation are not merely persistent but teratogenic. For
example, brains from the offspring of Hartley guinea pigs ex-
posed to a 44 °C environment for as little as 1 h while pregnant
show signs of apoptosis and arrested mitotic activity for 4 to 8 h
after exposure.24 Longer exposure times lead to even more del-
eterious effects, including abortion and gross malformation.8,9
Animals naturally seek to avoid extremes in temperature
through a process known as behavioral thermoregulation.23
Several studies have demonstrated that given a choice, Hartley
guinea pigs show a preference for an environment of approxi-
mately 30.6 °C, which is within the thermoneutral range for
guinea pigs (30 to 31 °C).13,16 Within this range, animals make
the most efcient use of energy expenditure, requiring the least
amount to maintain body temperature. Outside of this range,
lower temperatures cause a rise in metabolic rate, possibly lead-
ing to thermal stress, and higher temperatures result in elevated
body temperature or even a febrile state.14
The purpose of the current study was to determine the tem-
perature preference of Hairless guinea pigs as compared with
the Hartley strain from which it was derived. This aim was
achieved by taking advantage of thermoregulatory behavior
and allowing the guinea pigs to freely explore a temperature-
gradient environment. Because they lack fur, we hypothesized
that Hairless guinea pigs would prefer a slightly warmer
temperature (for example, 32 to 38 °C) within the gradient
than would Hartley guinea pigs. Although animals typically
are tested one at a time in temperature-preference studies,
we also observed cagemate pairs in our temperature-gradient
experiment. The rationale for this additional study condition
was that guinea pigs are naturally social animals. As such,
testing solitary subjects is atypical and less likely to evoke the
true thermoregulatory behavior of guinea pigs. In addition,
guinea pigs housed together tend to spend considerable time
in close proximity, often huddling together.1 We hypothesized
that when tested in pairs, the preferences of both strains would
shift to slightly cooler temperatures, because of the additional
thermoregulation provided by proximity to a cagemate.
Materials and Methods
Subjects. A total of 24 guinea pigs (Cavia porcellus) were used
in this study. IAF Hairless adult female (n = 8) and male (n =
8) guinea pigs were obtained from Charles River Laboratory
(Kingston, NY). Adult female NIH multicolored Hartley guinea
pigs (n = 8) were acquired from Elm Hill Labs (Chelmsford,
MA). Guinea pigs were guaranteed to be pathogen-free from
Temperature Preference in IAF Hairless and
Hartley GuineaPigs(Caviaporcellus)
Gale A Kleven* and Prianca Joshi†
The Hairless strain of guinea pigs (Cavia porcellus) is the result of a spontaneous recessive mutation rst identied at the
Institute Armand Frappier (IAF) in 1978. Despite the longstanding availability of this strain, little is known about its ther-
moregulatory behavior. The aim of this study was to determine temperature preference in Hartley and Hairless guinea pigs
by observing each strain in a ring-shaped apparatus containing a nonlinear temperature gradient. Temperatures were main-
tained by separately controlled heating mats lining the apparatus. Set point temperatures ranged from 24 to 38 °C. Guinea
pigs (Hartley female, Hairless female, and Hairless male guinea pigs; n = 8 each group) were placed either singly or in pairs at
1 of the 8 randomized starting points within the apparatus. Subjects were observed for 30 min and coded for location within
the temperature gradient by both frequency and duration. When placed singly in the apparatus, all 3 groups spent more time
in the 30 °C zones. However, when placed as pairs with a cagemate, Hartley female guinea pigs spent more time in the cooler
range of temperatures from 24 to 30 °C, whereas Hairless guinea pigs preferred a range of 30 to 38 °C. These results conrm
a temperature preference of 30 ± 2 °C for both Hartley and Hairless guinea pigs when singly housed. However, data from the
paired housing condition suggest that context plays an important role in thermoregulatory behavior.
Abbreviation: IAF, Institute Armand Frappier
Received: 06 Mar 2015. Revision requested: 13 Apr 2015. Accepted: 19 Jun 2015.
Department of Psychology, Wright State University, Dayton, Ohio
*Corresponding author. Email: gale.kleven@wright.edu
†Current address: Brown University, Providence, Rhode Island
jaalas15000029.indd 161 2/25/2016 10:39:28 AM
162
Vol 55, No 2
Journal of the American Association for Laboratory Animal Science
March 2016
zones enabled the placement of subjects into the apparatus with
minimal heat loss or change in temperature.
To regulate each zone, a temperature probe was xed at the
top of the outer wall over the center of each of the 8 heating
mats. These individual temperature controllers turned each
of the heating mats on and off according to the set-point tem-
perature of the controllers. Controllers (catalog no. 11-463-47A,
Fisher Scientic, Waltham, MA) were programmed to regulate
the temperature of each zone to within 0.5 °C of the set point.
The set points ranged from room temperature for the Hairless
strain (24 °C) to body temperature (38 °C). The temperature set
points were arranged in 2 to 4 °C increments across the zones
to create a gradient of temperatures (Figure 1).
The apparatus was lined with synthetic eece sewn to the
exact shape of the ring dimensions. A fresh liner was used each
time a guinea pig was placed in the apparatus. Liners were
reused on successive days after laundering with enzymatic
detergent (Free and Clear, Seventh Generation, Burlington,
VT) and visual inspection under UV light (UVR-9000, BAYCO
Products, Wylie, TX). This process ensured the lack of residual
bodily uids that might inuence place preference through
olfactory cues.
Data collection and analysis. Data were collected across a 4-d
period for each of 3 different experiments: 1) apparatus control;
2) single subject; and 3) cagemate pair. The same subjects were
used in all 3 experiments, which were run in the order listed. The
purpose of the apparatus-control experiment was to determine
whether the guinea pigs displayed a preference regarding loca-
tion within the apparatus irrespective of temperature. For this
experiment, the heating mats were left in place but turned off,
the vendors, and no additional surveillance was performed in
the vivarium other than weekly weighing and daily monitoring
for healthy appearance. Guinea pigs were 8 to 10 mo old at the
time of testing and had been exposed to the temperature condi-
tions in the viviarum (Wright State University, Dayton, OH) for
at least 6 mo. At the time of testing, female Hairless guinea pigs
weighed 955.18 ± 41.47 g (mean ± 1 SD), and males weighed
1071.31 ± 71.82 g; female Hartley guinea pigs weighed 1041.56
± 61.28 g. Although the 2 strains were maintained in separate
vivaria, both were housed in standard polycarbonate laboratory
housing (76 cm × 56 cm) with a cagemate from the same strain.
Food (Teklad 7006 Guinea Pig Diet, Harlan, Indianapolis, IN)
and water were provided free choice. Housing areas were on a
12:12-h light:dark cycle (lights on, 0800). Cages were lined with
paper bedding (TEK-Fresh, Harlan) and maintained in humid-
ity-controlled rooms (46% to 56%). To ensure sufcient warmth
for the Hairless strain, the room temperature was increased to 24
°C,3 as compared with the standard room temperature of 23 °C
maintained for the Hartley strain. In addition, the home cages
of Hairless guinea pigs contained a tunnel blanket (length, 36
cm; diameter, 30 cm) constructed from synthetic eece (Polartec,
Monsanto, Creve Couer, MO). The tunnel blanket provided
a burrow-like environment that the guinea pigs could use to
obtain additional warmth. Hartley guinea pigs were given a
section of PVC tubing to act as a home-cage burrow.
In addition to the home-cage supplementation, guinea pigs
were exposed to an enrichment program. For approximately
1 h daily on each weekday (Monday through Friday), 8 to 10
same-sex guinea pigs were placed in an open-eld arena (140
cm × 70 cm) that was lined with synthetic eece (Monsanto).
Wire spheres containing Western timothy hay (Pheleum pratense
L.) were located at 2 ends of the arena, and water was available.
Guinea pigs had unlimited access to both the hay and water
throughout the time spent in the open-eld arena. This enrich-
ment program has been demonstrated to provide increased
social interaction and exercise for both Hartley and Hairless
guinea pigs.7
At all times, guinea pigs were cared for in accordance with the
Guide for the Care and Use of Laboratory Animals.15 All procedures,
husbandry, and care involving the guinea pigs in this study were
reviewed and approved by the Wright State University Animal
Care and Use Committee.
Temperature-gradient apparatus. The experimental apparatus
(Figure 1) was constructed of molded plastic in the shape of a
ring (diameter, 92 cm; circumference, 2.9 m). This design was
chosen to eliminate corners in which guinea pigs might choose
to burrow or huddle regardless of the local ambient temperature.
In addition, the round design affords no directional cues for test
subjects. To further reduce the inuence of external cues, the
apparatus was placed in the center of a 20-m2 room and tightly
enclosed with a ring of curtains (height, 2 m) as a blind. The
apparatus was positioned in the room in such a way that only
blank ceiling was visible from the apparatus.
The apparatus was covered with 8 overlapping heating mats
(Figure 1, dotted lines; width, 30 cm; length, 60 cm) to create 8
temperature zones. The mats were placed in U-shaped patterns
so that each one covered a continuous section of the outer wall
(height, 23 cm), apparatus oor (width, 16 cm), and inner wall
(height, 23 cm). Eight clear acrylic panels, each cut precisely
to the trapezoidal shape of a zone, were placed on top of the
walls directly over each heating mat. These acrylic panels al-
lowed unobstructed visibility of guinea pig behavior during
the experiments and prevented the subjects from exploring
outside the apparatus. In addition, the individual lids over the
Figure 1. Schematic of the temperature-gradient apparatus. The ap-
paratus was divided into 8 equal zones. Each zone was lined with a
heating mat (dotted lines), which covered the apparatus oor (16 cm
wide) and both inner and outer walls (height, 23 cm) in a continu-
ous U-shaped pattern. Zones were tted with a trapezoidal acrylic lid,
placed directly over each heating mat. Having individual lids enabled
the placement of subjects in the apparatus with minimal heat loss.
Above the mat at the top of the outer wall, a temperature probe (T)
recorded the air temperature for each of the 8 zones. Individual tem-
perature controllers (not shown) were set to the temperature specied
for the zone. For example, the controller for zone 34a was set to 34 °C.
These controllers turned the heating mats on or off according to the
reading of the temperature probe (within 0.5 °C). This arrangement of
equally spaced controllers created a temperature gradient that ranged
from 24 °C (room temperature) to 38 °C (guinea pig body tempera-
ture). In addition, the apparatus was lined with a tted synthetic polar
eece that was changed for each subject.
jaalas15000029.indd 162 2/25/2016 10:39:28 AM
163
Temperature preference of guinea pigs
multivariate test criteria and Statistica software (version 10,
StatSoft, Tulsa, OK). Zones were treated as within-subject
repeated measures, with the 3 guinea pig groups (male Hair-
less, female Hairless, female Hartley) and pairs (subject and
cagemate) treated as between-subjects factors. For all analyses,
differences were considered signicant when the P value was
less than 0.05, and effect sizes were calculated by using ηp
2.
Posthoc pairwise comparisons were performed by using the
Fisher protected least-square difference only on results with
statistically signicant overall analyses.
Results
Zone temperatures recorded just prior to the placement of
each test subject remained within 1 °C of the designated set point
temperatures (Table 1) and did not vary signicantly between
single-subject and cagemate-pair experiments (F1, 46 = 1.400, P =
0.235, ηp
2 = 0.031). Although a main effect of zone was detected
(F7, 40 = 1590.601, P < 0.001, ηp
2 = 0.979), there was no interaction
with experiment (F7, 322 = 1.400, P = 0.213, ηp
2 = 0.029).
Experiment 1: apparatus control. After placement, guinea pigs
actively explored the apparatus, visiting all zones repeatedly
(Figure 2 A). No signicant difference in the frequency of zone
crossing was detected across the 8 zones (F7, 15 = 1.842, P = 0.152,
ηp
2 = 0.073) nor was there a zone × group interaction (F14, 32 =
0.765, P = 0.696, ηp
2 = 0.136). There was, however, a main effect of
group (F2, 21 = 5.407, P = 0.013, ηp
2 = 0.340), with Hairless females
making more frequent visits to most zones than did the Hart-
ley females or Hairless males. Posthoc pairwise comparisons
conrmed differences between Hairless and Hartley females
for all 8 zones (P < 0.05) and signicantly elevated frequencies
for Hairless females compared with males for all zones except
26a, 30a, and 34a (P < 0.05).
In addition, guinea pigs spent similar amounts of time in each
zone. (Figure 2 B). Analysis of the total time spent in each zone
(Figure 2 B) revealed no signicant effects of zone (F7, 15 = 1.842,
P = 0.152, ηp
2 = 0.070), group (F2, 21 =1.617, P = 0.222, ηp
2 = 0.133),
or zone × group interaction (F14, 32 = 0.811, P = 0.652, ηp
2 = 0.094).
Experiment 2: single subject. Similar to the ndings from
the apparatus-control experiment, guinea pigs placed in the
temperature gradient as single subjects actively explored the
environment, visiting all zones frequently (Figure 3 A). A main
effect of zone was revealed (F7, 15 = 5.342, P = 0.003, ηp
2 = 0.191).
However, no differences in rates of exploration were detected
between the 3 groups (F2, 21 = 2.394, P = 0.116, ηp
2 = 0.186), nor
was there a zone × group interaction (F14, 32 = 0.778, P = 0.684,
ηp
2 = 0.055). Posthoc analyses determined that most differences
were between the coolest zone (24 °C) and the warmest 2 zones,
34b and 38 (P < 0.05).
The average time spent in the 30 °C zones was signicantly
greater than those in other temperature zones (Figure 3 B). Analy-
ses revealed only a main effect of zone (F7, 15 = 6.968, P < 0.001,
ηp
2 = 0.174), with no effect of group (F2, 21 = 0.662, P = 0.526, ηp
2 =
0.059) or zone × group interaction (F14, 32 = 0.610, P = 0.837, ηp
2 =
0.073). Posthoc pairwise comparisons conrmed increased mean
durations in the 30a zone compared with zones 24, 26a, 26b, and
34a (P < 0.05) for Hartley female guinea pigs and in the 30b zone
compared with all other zones (P < 0.05) for Hairless females.
Elevated rates for Hairless male guinea pigs were conrmed for
both zones 30a and 30b relative to other zones (P < 0.05).
Similar ndings were obtained for the total time spent in each
zone (Figure 3 C). Signicant effects were revealed for zone (F7,
15 = 31.043, P < 0.001, ηp
2 = 0.215) and group (F2, 21 = 14.250, P <
0.001, ηp
2 = 0.576) but not zone × group (F14, 32 = 1.045, P = 0.438,
ηp
2 = 0.303). Posthoc pairwise comparisons conrmed elevated
and all zones in the apparatus were room temperature (24 °C).
Each guinea pig in a group was randomly assigned to 1 of the
8 zones for initial placement. In the single-subject experiment,
the zone temperatures were set and maintained as depicted in
Figure 1; guinea pigs again were randomly assigned to a starting
zone. The cagemate-pair experiment used the same temperature
set points as the single-subject experiment. However because
guinea pigs are social creatures that spend much of their time
huddled with cagemates, guinea pigs were placed as pairs
within the apparatus in this third experiment, to ascertain if
different temperature zones would be traversed than those ob-
served in the rst 2 experiments. Pairs were cagemates that had
also served as subjects in the rst 2 experiments. Consequently,
all guinea pigs were equally familiar with the apparatus. Each
subject was tested: 1) once in the apparatus-control study; 2)
once in the single-subject study; and 3) twice in the cagemate-
pair study. Cagemate pairs were run twice, once on each of 2 d,
because cagemates served as subjects on one run, and subjects
served as cagemates on the other. The initial starting zone for the
subject guinea pig was randomized, with the cagemate placed
in the zone opposite from that of the subject. For example, when
the subject was placed in zone 34a, the cagemate was placed
in zone 26b.
For all experiments, data were collected during the light phase
of the light:dark cycle. Zone temperatures were recorded just
prior to subject placement in the apparatus. Guinea pigs were
transferred from the adjacent colony room (4 to 5 m away)
in an opaque container (15 cm × 25 cm × 15 cm) immediately
before being placed in the apparatus. Sequestered transfer was
preformed to mask any room or orientation cues. Upon enter-
ing the experimental room, the container with the guinea pig
was placed next to the apparatus inside the curtain barrier. The
experimenter then reached between the curtains and placed the
guinea pig in the randomly assigned zone.
While in the temperature-gradient apparatus, the guinea
pigs were monitored through a remote-controlled digital
camcorder (3MOS HD, Panasonic, Osaka, Japan) placed above
the apparatus. Guinea pigs were videotaped for 30 min in the
apparatus and then returned to the colony room inside the
covered container. Video les were transferred to a computer
and rendered for scoring into MPEG2 format by using Premier
Pro CS4 software (Adobe, San Jose, CA).
Coding for the location of the guinea pigs by zone within the
apparatus was achieved through play back of the video les
by using the JWatcher event recorder.5 Although coders were
easily able to determine the zone boundaries created by the
acrylic covers, they were blinded to the actual designation and
temperature of the zones. Guinea pigs were coded as being in
a particular zone when the center of the body rst moved into
that zone, regardless of whether other body parts (for example,
head, limbs) were in an adjacent zone. This system of coding
allowed both frequencies and durations of time (for example,
mean duration, total time) to be calculated for each zone. In
the experiment involving cagemate pairs, each guinea pig was
scored on a separate pass. The les were then combined for
subsequent analysis of frequencies and durations, including
quantication of when cagemates shared adjacent zones with
the subject. Adjacent was dened as when the cagemate was in
the same zone, or one to either side of the subject. For example,
when the subject guinea pig was in zone 30a (Figure 1), the
cagemate was considered to be adjacent when located in zone
26a, 30a, or 34a.
Data for each experiment were analyzed separately with
repeated-measures ANOVA by using Pillai–Bartlett trace
jaalas15000029.indd 163 2/25/2016 10:39:29 AM
164
Vol 55, No 2
Journal of the American Association for Laboratory Animal Science
March 2016
P = 0.017, ηp
2 = 0.096). Specically, Hartley female guinea pigs
and their cagemates spent more time in zones 30 °C and cooler
(Figure 5 D) than in the 3 warmest zones (P < 0.05). In contrast,
the Hairless female guinea pigs spent signicantly more time in
zones 26 °C and warmer (Figure 5 E) than in cooler zones, with
the greatest amount of time in zone 30a (P < 0.05). Male Hairless
subjects spent more time in the 30 °C zones, whereas cagemates
spent signicantly more time in zones 26b and 34b compared
with zones 34a and 38. The remaining analyses revealed no main
effect of pair (F1, 42 = 2,547, P = 0.118, ηp
2 = 0.057) or interaction
of group × pair (F2, 42 = 0.002, P = 0.998, ηp
2 < 0.001), zone × pair
(F7, 36 = 0.560, P = 0.783, ηp
2 = 0.016), or zone × group × pair
(F14, 74 = 0.761, P = 0.706, ηp
2 = 0.032).
To better quantify the spatial relationship between subject and
cagemate, the frequency and total time that paired guinea pigs
spent in proximity to one another were analyzed. Proximity was
dened as when the cagemate was in the same or an adjacent
zone as the subject. For example, when the subject was in zone
34a, the cagemate was adjacent when it was in zone 30a, 34a,
or 38 (Figure 1). The adjacent analysis for zone frequency re-
vealed a main effect of group (F2, 21 = 11.392, P < 0.001, ηp
2 = 0.520),
with increased rates of proximity for Hartley female guinea
pigs compared with the other groups, but no effect of zone
(F7, 147 = 0.828, P = 0.565, ηp
2 = 0.038) or zone × group interac-
tion (F14, 147 = 1.608, P = 0.083, ηp
2 = 0.133; Figure 6 A). Posthoc
analyses conrmed signicantly increased rates of proximity
for Hartley female guinea pigs, relative to both other groups,
at all zones except 30a, 30b, and 38 (P < 0.05).
Similar to those for zone frequency, analyses of the total
time animals spent adjacent in each zone revealed main ef-
fects of group (F2, 21 = 3.695, P = 0.042, ηp
2 = 0.260) and zone
(F7, 15 = 3.535, P = 0.019, ηp
2 = 0.259) but no zone × group interac-
tion (F14, 32 = 1.509, P = 0.164, ηp
2 = 0.109; Figure 6 B). However, in
this analysis of total time adjacent, the main effects of group and
Figure 2. Frequency and total time (mean ± SEM) for the apparatus-
control experiment. (A) Frequency of zone crossings in the room tem-
perature (24 °C) apparatus for Hartley female and both male and fe-
male Hairless guinea pigs. Hairless females had signicantly higher
frequencies of zone crossings than Hartley females for all zones (P <
0.05). Asterisks denote the zones for which Hairless females had high-
er crossing rates than did males (P < 0.05). (B) Total time that guinea
pigs spent in each zone of the room temperature apparatus during the
30-min observation period. There was no difference in time spent per
zone for any of the 3 guinea pig groups.
Table 1. Mean apparatus temperatures and set points during experi-
ments
Zone
Set point
(°C)
Mean temperature (°C), 95% condence interval
Single subject Cagemate pair
24 24 23.995 23.683
23.766–24.226 23.453–23.914
26a 26 26.408 26.775
26.136–26.680 26.503–27.047
30a 30 30.229 30.050
29.934–30.525 29.755–30.345
34a 34 34.521 34.621
34.135–34.906 34.235–35.006
38 38 38.425 38.717
38.023–38.827 38.315–39.119
34b 34 34.379 34.521
34.050–34.709 34.191–34.850
30b 30 30.067 30.296
29.683–30.450 29.913–30.679
26b 26 26.200 26.563
26.004–26.396 26.367–26.758
*see Figure 1 for physical arrangement of zones
mean durations for Hartley female guinea pigs in the 30a zone
compared with zones 24, 26a, 26b, and 34a (P < 0.05) and for
Hairless females in the 30b zone compared with all other zones
(P < 0.05). Elevated rates for Hairless males were conrmed as
well, but for both zones 30a and 30b as compared with other
temperature zones (P < 0.05).
Experiment 3: cagemate pairs. Similar to ndings from both
experiments 1 and 2, guinea pigs actively explored the apparatus
in the cagemate experiment. Overall 3-way ANOVA for activity
by zone revealed elevated frequency rates for Hartley female
guinea pigs, as compared with both Hairless groups, when
placed in the temperature apparatus with a cagemate (Figure 4).
Although a signicant main effect of group emerged (F2, 42 =
19.514, P < 0.001, ηp
2 = 0.482), no signicant differences were
detected for zone (F7, 36 = 1.443, P = 0.219, ηp
2 = 0.009), pair
(F1, 42 = 0.047, P = 0.830, ηp
2 = 0.001) zone × group (F14, 74 = 1.726,
P = 0.068, ηp
2 = 0.087), zone × pair (F7, 36 = 0.680, P = 0.688, ηp
2 =
0.013) or zone × group × pair (F14, 74 = 1.193, P = 0.299, ηp
2 = 0.028).
Quantication of mean durations (Figure 5 A through C) re-
vealed main effects of zone (F7, 36 = 2.657, P = 0.025, ηp
2 = 0.059)
and group (F2, 42 = 3.276, P = 0.048, ηp
2 = 0.135) but not pair
(F1, 42 = 0.039, P = 0.844, ηp
2 = 0.135). Similarly, no interaction was
detected for zone × group, (F14, 74 = 1.650, P = 0.086, ηp
2 = 0.067),
zone × pair (F7, 36 = 0.722, P = 0.654, ηp
2 = 0.010), or zone × group
× pair (F14, 74 = 0.533, P = 0.906, ηp
2 = 0.019). Posthoc comparisons
revealed no differences in mean duration in a zone between
Hartley pairs (Figure 5 A) but elevated mean durations were
detected for Hairless female guinea pigs in zone 30a (P < 0.05),
with subject means further increased over those of cagemates
(Figure 5 B). For the Hairless male guinea pigs, signicant
elevation in zone duration were detected for cagemates only
relative to subject means (Figure 5 C). Elevations in mean du-
ration were detected for zone 30a relative to 38; for zone 30b
relative to 26a, 34a, and 38; and for zone 34b relative to both
34a and 38 (P < 0.05).
A different pattern emerged for the total amount of time each
group of guinea pigs spent in the various zones (Figure 5, D
through F), as evidenced by a main effect of zone (F7, 36 = 5.179, P
< 0.001, ηp
2 = 0.065) and a zone × group interaction (F14, 74 = 2.171,
jaalas15000029.indd 164 2/25/2016 10:39:29 AM
165
Temperature preference of guinea pigs
the Hartley strain may reect higher rates of social interaction
between the cagemate pairs. In a prior study examining the
behavioral differences of the 2 guinea pigs strains, we found
increased levels of social interaction between Hartley compared
with Hairless female guinea pigs in an open-eld environment.7
Consequently, the variation noted in the current study might
similarly reveal social differences between these 2 strains.
Further support for this conclusion is found in the increased
frequency of proximity for Hartley cagemates relative to those
for both Hairless male and female guinea pigs (Figure 6 A).
Our guinea pigs had very different distributions for time
spent across various zones in the single-subject compared with
cagemate-pair experiments. When placed individually in the
temperature gradient, subjects from all 3 groups spent greater
durations and longer total time in the 30 °C zones. This nding
is similar to prior studies of behavioral thermoregulation for
the Hartley strain.13,16 However, when placed within the ap-
paratus with a cagemate, guinea pigs from all 3 groups spent
more time across a much wider range of zones (Figure 5). This
result differed by strain, with Hartley females spending more
time in the cooler zones of 24 to 30 °C (Figure 5 D) and both
male and female Hairless guinea pigs spending more time in
the warmer zones at 30 to 38 °C (Figure 5 E and F).
The greater distribution across zones between the subjects in
the cagemate-pair experiment was not likely the result of space
constraint, because neither of the female groups spent signicant
zone occurred in Hairless male guinea pigs instead of Hartley
females, as occurred for the frequency analysis. Posthoc com-
parisons conrmed that Hairless males spent increased time in
zones 30a and 30b compared with zones 26b and higher.
Discussion
Guinea pigs from all groups actively explored the apparatus,
crossing frequently to all zones. In the apparatus-control experi-
ment, neither the frequency of crossing nor total time differed
between zones (Figure 2), suggesting that the guinea pigs had no
particular preference for place within the apparatus. Similarly,
the frequency of exploration in the single-subject and cagemate-
pair experiments did not differ across the various zones (Figures
3 A and 4). However, female guinea pigs of both strains showed
increased zone crossings compared with male Hairless guinea
pigs, and Hairless females showed higher rates of zone cross-
ings in the apparatus control than did Hartley females (Figure
2 A). In comparison, Hartley female guinea pigs had higher
frequencies of zone crossings in the cagemate-pair experiment
than did the other groups (Figure 4). These increased rates for
Figure 3. Frequency, duration, and total time per zone (mean ± SEM)
for the single-subject experiment. Although guinea pigs actively ex-
plored all zones within the temperature-gradient apparatus, they
spent signicantly longer times in the 30 °C zones. (A) Frequency of
crossings per zone for the 3 guinea pig groups. (B) Duration within
each of the 8 zones. Signicantly longer durations were detected for
Hartley female guinea pigs in zone 30a (30 °C), for Hairless females in
zone 30b (30 °C), and for males in both of the zones maintained at 30
°C (P < 0.05). (C) Total time spent by the guinea pig groups in each of
the temperature zones. Guinea pigs spent more than half the 30-min
observation period in one of the 30 °C zones. Similar to the duration
results, Hartley female guinea pigs spent signicantly more time in
zone 30a, Hairless females spent more time in 30b, and Hairless males
spent increased time in both of the 30 °C zones (P < 0.05).
Figure 4. Zone frequency (mean ± SEM) for the cagemate-pair experi-
ment. Bars depict the mean activity level in each of the temperature
zones for subjects (white) and cagemates (black). Increased frequen-
cies of zone crossings were observed for Hartley female guinea pigs
(A, P < 0.05) relative to Hairless females (B) and males (C). No differ-
ences were detected between subjects and cagemates for any guinea
pig group in any zone.
jaalas15000029.indd 165 2/25/2016 10:39:30 AM
166
Vol 55, No 2
Journal of the American Association for Laboratory Animal Science
March 2016
than did the Hartley strain and, when placed with a cagemate,
the Hairless guinea pigs actually spent more time in warmer
zones instead of demonstrating the shift to cooler tempera-
tures that we predicted. Only the data from the Hartley strain
cagemates supported our hypothesis of a shift to cooler tem-
perature zones (Figure 5 D through F). Regardless, the Hairless
strain’s preference for the same ambient temperature as Hart-
ley guinea pigs, a strain with much more thermoregulatory
resources in the form of fur insulation, requires additional
discussion.
Three explanations for this singularity in preference are pos-
sible. The rst is that the Hairless strain has had insufcient
evolutionary time to develop new thermoregulatory adaptations
to being hairless and therefore seeks the same thermoneutral
zone temperature as does the Hartley strain.23 This explana-
tion might suggest that Hairless guinea pigs are incapable of
adapting its thermoregulatory behavior to a new environment
and thus simply selects a preprogrammed temperature. Al-
ternatively, Hairless guinea pigs may have adapted a higher
metabolism and therefore do not require a warmer ambient
environment to maintain body temperature. For both these
explanations, studies recording physiologic responses of Hair-
less guinea pigs to different ambient temperatures may help
answer these questions.
time in proximity. Only the Hairless male guinea pigs showed
elevated times in proximity to one another and even then only
in the 30 °C zones (Figure 6 B). This nding of limited time spent
in proximity stands in sharp contrast to typically observed,
social behavior of guinea pigs1,7,19,20 and likely results from
placement in a novel environment (that is, different from the
homecage). Such a conclusion raises the question of whether
temperature-gradient testing outside the homecage environ-
ment is generalizable beyond the apparatus used. Additional
studies comparing homecage temperature preferences with
those observed in an apparatus may help answer this question.
In addition, the atypical apparatus used in the current study
raises the question of whether a testing period of 30 min is
sufcient to determine temperature preference. A prior study
comparing the behavior of Hairless and Hartley guinea pigs
found that most differences compared with homecage behavior
were observed in the rst 15 min after placement in a familiar
apparatus.7 These ndings suggest that the 30-min observation
period used in the current study may have been sufcient for the
guinea pigs to acclimate to the apparatus. However additional
testing with longer observation times would provide conrma-
tion of the temperature preferences reported in this study.
Taken together, the data do not fully support our 2 hypoth-
eses in that the Hairless strain did not seek a warmer zone
Figure 5. Duration and total time per zone for the cagemate-pair experiment. (A–C) Duration and (D–F) total time of subjects (white) and cage-
mates (black) in the 3 guinea pig groups: Hartley female guinea pigs (A and D), Hairless females (B and E), and Hairless males (C and F). There
was no difference in duration or total time between the Hartley subject and cagemate for any zone (A and D). Although there was no difference
in duration for any zone (A), Hartley females spent a greater amount of the total time in zone temperatures of 24 to 30 °C (P < 0.05) compared
with the warmest 3 zones (D). In contrast, Hairless females stayed for longer periods in one of the 30 °C zones (B, P < 0.05) and spent more time in
the warmer zones (E, P < 0.05). Similarly, Hairless males stayed for longer periods in the 30 °C zones, with longer mean durations for cagemates
(C, P < 0.05) and greater total time per zone for subjects (F, P < 0.05). In addition, cagemates showed increased durations (C) and total time (F) in
one of the 34 °C zones and greater total time in one of the 26 °C zones (F, P < 0.05).
jaalas15000029.indd 166 2/25/2016 10:39:31 AM
167
Temperature preference of guinea pigs
References
1. Allen-Charters JB. 1904. The associative processes of the guinea
pig. J Neurol Psychol 14:294–361.
2. Balk MW. 1987. Emerging models in the USA: swine, woodchucks,
and the Hairless guinea pig. Prog Clin Biol Res 229:311–326.
3. Banks R. 1989. The guinea pig: biology, care, identication, no-
menclature, breeding, and genetics. USAMRID Seminar Series, 17
February 1989.
4. Bellinger SA, Lucas D, Kleven GA. 2015. An ecologically relevant
guinea pig model of fetal behavior. Behav Brain Res 283:175–183.
5. Blumstein DT, Daniel JC, Evans CS. 2000. JWatcher video user’s
guide. Macquarie University (Australia): Animal Behavior Labora-
tory.
6. Bolognia JL, Murray MS, Pawelek JM. 1990. Hairless pigmented
guinea pigs: a new model for the study of mammalian pigmenta-
tion. Pigment Cell Res 3:150–156.
7. Brewer JS, Bellinger SA, Joshi P, Kleven GA. 2014. Enriched
open eld facilitates exercise and social interaction in 2 strains of
guinea pig (Cavia porcellus). J Am Assoc Lab Anim Sci 53:344–355.
8. Edwards MJ. 1967. Congenital defects in guinea pigs following
induced hyperthermia during gestation. Arch Pathol 84:42–48.
9. Edwards MJ. 1969. Congenital defects in guinea pigs: fetal
resorptions, abortions, and malformations following induced
hyperthermia during early gestation. Teratology 2:313–328.
10. Fox MA, McNichols RJ, Gowda A, Motamedi M. 2004. The use
of the hairless guinea pig in tattoo research. Cont Top Lab Anim
Sci 43:35–38.
11. Gaskill BN, Gordon CJ, Pajor EA, Lucas JR, Davis JK, Garner JP.
2012. Heat or insulation: behavioral titration of mouse preference
for warmth or access to a nest. PLoS One 7:e32799.
12. Gaskill BN, Rohr SA, Pajor EA, Lucas JR, Garner JP. 2009. Some
like it hot: mouse temperature preferences in laboratory housing.
Appl Anim Behav Sci 116:279–285.
13. Gordon CJ. 1986. Relationship between behavioral and autonomic
thermoregulation in the guinea pig. Physiol Behav 38:827–831.
14. Gordon CJ. 2005. Temperature and toxicology. Boca Raton (FL):
CRC Press.
15. Institute for Laboratory Animal Research. 2010. Guide for the care
and use of laboratory animals, 8th ed. Washington (DC): National
Academy Press.
16. Laughter JS Jr, Blatteis CM. 1985. A system for the study of
behavioral thermoregulation of small animals. Physiol Behav
35:993–997.
17. Michel CL, Chastel O, Bonnet X. 2011. Ambient temperature
and pregnancy inuence cortisol levels in female guinea pigs and
entail long-term effects on the stress response of their offspring.
Gen Comp Endocrinol 171:275–282.
18. Miyauchi H, Horio T. 1992. A new animal model for contact
dermatitis: the Hairless guinea pig. J Dermatol 19:140–145.
19. Sachser N. 1986. Different forms of social organization at high and
low population densities in guinea pigs. Behaviour 97:253–272.
20. Sachser N. 1998. Of domestic and wild guinea pigs: studies in
sociophysiology, domestication, and social evolution. Naturwis-
senschaften 85:307–317.
21. Speakman JR, Keijer J. 2012. Not so hot: optimal housing tem-
peratures for mice to mimic the thermal environment of humans.
Mol Metab 2:5–9.
22. Sueki H, Gammal C, Kudoh K, Kligman AM. 2000. Hairless
guinea pig skin: anatomical basis for studies of cutaneous biology.
Eur J Dermatol 10:357–364.
23. Terrien J, Perret M, Aujard F. 2011. Behavioral thermoregu-
lation in mammals: a review. Front Biosci (Landmark Ed) 16:
1428–1444.
24. Upfold JB, Smith MS, Edwards MJ. 1989. Quantitative study of
the effects of maternal hyperthermia on cell death and prolifera-
tion in the guinea pig brain on day 21 of pregnancy. Teratology
39:173–179.
Finally, there may be a factor involving discrimination within
the temperature gradient. Because the temperature difference
between the zones surrounding 30a and 30b was 4 °C (Figure
1), guinea pigs may have been seeking temperatures slightly
above or below the set point for the zone. For example, within
the 30 °C zones, the ambient temperature likely ranged from
28 to 32 °C. Within such a wide range, the Hairless guinea pigs
could have been seeking a slightly higher temperature within
the zone compared with that preferred by the Hartley strain.
Future studies that use a narrower range of temperatures (for
example, 28 to 32 °C) across the entire apparatus may reveal
differences not detected in the current set of experiments.
Furthermore, caution should be used in extrapolating our
current results to ambient temperatures in other contexts such
as housing. Studies with other rodents have demonstrated
that the temperature an animal prefers may not be the optimal
housing temperature.11,12,21 As an initial study, however, the
experiments presented here provide much needed data on
the temperature preference of Hairless guinea pigs. In addi-
tion, this paper reports for the rst time the thermoregulatory
behavior of cagemate pairs for both laboratory strains. Results
from this study also highlight the role that context may play in
thermoregulatory behaviors.
Acknowledgments
We thank Lindsey Keene, Jacob Brewer, and Alex Arnett for their work
with video rendering and behavioral scoring. We also are grateful to the
Wright State University Laboratory of Animal Resource technicians for
the excellent care of our guinea pigs, with particular thanks to Emily
Dudley for her veterinary expertise.
Figure 6. Frequency and total time for proximity between subject and
cagemate. (A) Frequency and (B) total time per zone for the 3 guinea
pig groups when the cagemate was in proximity to the subject. Prox-
imity was dened as when the cagemate was in the same zone as or an
adjacent zone to that of the subject. For example, when the subject was
in zone 30a, the cagemate was proximate when it was in zone 26a, 30a,
or 34a (Figure 1). Hartley female cagemates were frequently adjacent
to the subject in most temperature zones (A, P < 0.05). No differences
in frequency of proximity were observed for Hairless females or males.
In contrast, Hairless males spent a greater amount of time in proximity
for the 30 °C zones relative to zones 26a and warmer (P < 0.05).
jaalas15000029.indd 167 2/25/2016 10:39:31 AM
