Brain Research 1054 (2005) 174 – 182

www.elsevier.com/locate/brainres

Research Report

Modest environmental enrichment: Effect on a radial maze validation

and well being of rats
Elsa Brillaud*, Delphine Morillion, René de Seze
Unité de Toxicologie Expérimentale, INERIS, Parc technologique ALATA, BP2, 60550 Verneuil-en-Halatte, France

Accepted 28 June 2005

Available online 10 August 2005

Abstract

Our 8-arm radial maze test was validated to demonstrate memory deficits in rats treated with the muscarinic antagonist scopolamine hydro
bromide (SHB, 0.1 mg/kg, i.p.). To improve quality of life, we enriched the environment of single housing rats. Enrichment procedures were
chosen to increase the animals’ well being without disturbing a lot the results of behavioural tests. It is modest, consisting of a plastic tube
and corn chips. Enriched environment (EE) and Non-enriched Environment (NE) animals’ performances were compared during the 8-arms
radial maze validation. Enrichment procedures were chosen to increase the animals’ well being without disturbing the results of behavioural
tests. The impact of our enrichment conditions was then evaluated on the general behaviour of rats, weight evolution and results of a plus
maze anxiety test. Results showed a deficit and a delay in learning for SHB-treated animals, and a general time-dependent learning effect,
validating our test. No effect of enrichment on negative control animals was observed. For SHB-treated animals, enrichment increased
performances during learning task and accentuated the deficits in test task. Exploratory behaviour of enriched animals seemed to be
increased. A general amelioration of well being for EE animals was found (stable weight). We conclude that our enrichment allows increasing
exploratory behaviour not modifying radial maze sensitivity using a simple modification of our protocol (limitation to 16 visits/trial). We
decided to generalise this enrichment to all our studies, given its simplicity and obtained benefits.
D 2005 Elsevier B.V. All rights reserved.

Theme: Neural basis of behavior

Topic: Motivation and emotion

Keywords: Enriched environment; Single housing; Well-being; Radial maze; Plus maze; Rat; Scopolamine

1. Introduction Centrally, acetylcholine (ACh) plays a specific role in

The 8-arms radial maze is specific of learning and
memories, using different protocols [31,45]. An eight baited
arms protocol was followed to test working memory. To
validate our test, reference drugs were used: scopolamine
hydro bromide (SHB) and scopolamine methyl bromide
(SMB) [13,17]. SHB is a specific cholinergic antagonist,
able to cross the blood – brain barrier (BBB), involving
effects on peripheral nervous system (PNS) and central
nervous system (CNS). SMB is also a specific cholinergic
antagonist, but the methyl group prevents BBB crossing.
* Corresponding author. Fax: +33 3 44 55 68 00.
E-mail address: Elsa.Brillaud@ineris.fr (E. Brillaud).
0006-8993/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.brainres.2005.06.069
learning and memory processes [4]. Central antagonist
effects of scopolamine involve impairing working memory
in radial maze task by increasing the number of working
memory errors (visits of already visited arm) and delay to
enter in arm and decreasing motivation [5,24,38,47,55]. In
periphery, ACh is the neurotransmitter of parasympathetic
system, reducing energetic consumption. Peripheral effects
obtained [41] affect task results without impairing learning
and memory [23,54]. With the SMB, only peripheral effects
are obtained. This treatment was used as the SHB control
group to determine the possible observed effects resulting
from a non-specific impairment of learning or memory.
To control the weight during radial maze task (food
motivation), animals were singly housed. In accordance to
 E. Brillaud et al. / Brain Research 1054 (2005) 174 – 182
ethics and legislation [11,12,28,32], alleviating stress from
single housing and deprivation was imperative. To improve
well-being and reduce stress, we decided to enrich the
animal’s environment.
Housing environment has an impact on biological
mechanisms underlying animal behaviour, and slight
changes during an experiment can alter responses [1,25].
Laboratory animals are deprived of some natural behaviours
and that induces suffering and a decrease of well being
[9,26,51]. Environmental enrichment (EE) allows welfare
improvement, and stress level decreases by giving more
behavioural possibilities. Animals consider it as a recom-
pense [2,42,50]. Several studies looked at the effects of EE
at physiological and behavioural levels [10,39,52]. Enrich-
ment increases brain weight, cortex development, which has
beneficial effect on performance of rats in behavioural tasks.
Moreover, EE could reduce effects of various affections
(stress, non-handling, brain lesion, ischemia, gliosis) and
could be used as treatment for them [19,29,33,37,46]. The
term ‘‘enrichment’’ covers social and inanimate stimulation
[40], but these two kinds of enrichment have dissociated
effects [43,44,58]. Social stimulation consists in putting
animals together. Indeed, single housing affects physiolog-
ical parameters by increasing stress level, increasing
locomotion activity and decreasing exploration [14,22]. In
learning behavioural tasks, isolated animals have worse
performances than enriched and socialised ones [8,15,20].
Isolation effects seem to be strain dependent [52] and can be
reduced by adding inanimate enrichment in the cage [3].
That consists in putting objects or enlarging cage dimen-
sions and increasing the complexity of the environment and
of behavioural possibilities. Rats prefer large cages and
objects to chew and to nest [6,34,35]. Behavioural benefice
of enrichment has been observed in an open field task:
quality of exploration is increased, whereas locomotion
behaviour decreases [7,18,53]. These effects depend on the
enrichment complexity but not on duration of enrichment
[48,58]. Besides, only few elements are needed to obtain
results [56].
To compare our results with those of previously
published studies, EE had not modified too much the
radial maze task results. The enrichment was chosen to be
modest and to not complicate the animal facilities
organisation. We test its effect during the validation of
the radial maze task. To test the impact of our enrichment
on welfare, a plus maze test was then performed. This
behavioural test is widely used to measure anxiety-like
reactions and exploratory behaviour in various pharmaco-
logical studies with rodents [16,27,36]. Furthermore, as the
radial maze, the plus maze is constituted of unclosed
elevated arms. It was chosen in order to allow an easier
comparison between results of two tests. In these con-
ditions, animals were not deprived. Their weight was
measured daily as a well being index, and general attitude
(coat aspect, facility of contention, behaviour in the cage)
was observed.
175
The first aim of the present study was to validate in our
laboratory a working memory protocol in the radial maze
using drugs. The second aim was to compare enriched
animal’s performances to the controls during the validation
of the radial maze task and to measure the well-being
increase due to enrichment.
2. Materials and methods
2.1. Radial maze experiment
2.1.1. Animals
Thirty-two male Sprague –Dawley rats (OFA Iffa Credo,
France), weighing 160 –180 g on arrival in the laboratory (6
to 7 weeks old), were used. Environmental ambient
conditions were controlled: ambient temperature 22 T 1
-C; hygrometry 50% T 2%; pressure + 2 mm Hg; 12 h light/
dark cycle. Experimental design showing different habitu-
ation phases is detailed in Fig. 1. Animals were first housed
4 per cage with food and water ad libitum and left 5 days
without any manipulation to be accustomed to animal
facilities. Later on, study experimenters handled them 10
min a day during 8 days. After 3 days of experimenters’
habituation, rats were randomised and placed 1 per cage
with water ad libitum but controlled food quantity. To
reinforce their motivation, food was reduced to obtain a
weight decrease of 10 to 15%. Animals were weighted daily
at the same hour, and food quantity was adapted according
to the theoretical weight curve (IFFA CREDO documents).
Pellets (20 mg; Phymep SARL, France) were given as
reinforcement food.
2.1.2. Experimental groups and treatment
2.1.2.1. Housing conditions. Half the animals (n = 16)
were housed in standard plastic laboratory cages, in Non-
enriched environment (NE) condition (Type III H, 425 
266  185 mm; sawdust + water + food) and the other half
in enriched cages, EE condition. This enrichment consisted
in adding corn chips (2 cm thickness) and a translucent
plastic tube (12 cm length; 7 cm diameter).
Fig. 1. Experimental design of Experiment 1.
176 E. Brillaud et al. / Brain Research 1054 (2005) 174 – 182
2.1.2.2. Treatment. Animals were divided in 6 groups, 3
with EE and 3 with NE. Three different treatments were
used for each environmental condition. Rats were injected
i.p. (1 ml/kg) with: (i) sodium chloride solution at 0.9%
(NaCl; Sigma Aldrich) as control group (EE and NE_NaCl;
n = 4 each); (ii) SHB 0.1 mg/kg (Sigma Aldrich) as positive
control group (EE and NE_SHB, n = 8 each); (iii) SMB 0.3
mg/kg (Sigma Aldrich) as negative control group (EE and
NE_SMB; n = 4 each).
Treatments started on the first learning day. However, in
order to limit stress, we also accustomed the animals to be
treated by injecting them with sodium chloride solution
during radial habituation.
2.1.3. Apparatus
The apparatus is a custom-made automated elevated
radial arm maze. Eight horizontal arms (57  11 cm) are
placed radially at a 360/8- angle around an elevated central
platform 80 cm above the floor. Automated doors (20 cm
high) are located at the entrance of each arm. Arms, central
platform and doors are made in grey opaque Plexiglas.
Partial walls (17 to 2 cm high; 15 cm long) are made in
transparent Plexiglas and placed at the beginning of each
arm. They prevent going from one arm to another one
without coming back to the central platform. Each arm is
equipped with three infrared diodes (2.5 cm above the
maze floor) located: (i) at the beginning; (ii) in the middle;
and (iii) 5 cm from the end of the arm. A feeding bowl
equipped with an infrared sensor is present at the end of
each arm. It can be baited with pellets by automated rods.
Infrared sensors are connected to a computer where the
animal location is recorded during the task. Automation of
the radial arm maze enables (i) the control of doors
movements with a specific program; (ii) the visualisation
of the animal location on a computer screen; and (iii) the
food consumption by animals during the test. A camera
fixed at the ceiling is connected to a black and white
monitor and a video recorder. It allows observing animal in
the maze.
The apparatus was isolated from the experimenter by two
large opaque curtains (same colour as walls). A halogen
lamp illuminated the ceiling of the room and induced 50 lx
lighting. To equilibrate external noise (air conditioning,
computer, experimenter and air compressor), a radio was
placed in the box at the opposite side from the experimenter.
Walls were clear, plain and naked except the following set of
elements: (i) on the first wall, a big white air conditioner, the
halogen and mural plugs; (ii) on the second one, a sink and
mural plugs; a coloured cue was placed in complement. Two
distinguishable posters were placed in the centre of each
curtain.
2.1.4. Behavioural protocol
The radial maze protocol consists of three consecutive
phases: habituation (3 days), the learning task (15 days
maximal) and the test task (6 days). Rats were trained every
day, once per day. Each rat was treated and was placed 30
min later on the central platform with all doors closed. After
30 s, all doors were opened, and the maze could be visited
during 10 min. Each rat was placed in the radial maze in a
random order changed every day. Radial maze was cleaned
between each animal with water and absorbing paper to
minimise olfactory intra-maze cues.
Animal course was recorded on computer, and
experimenter notes completed it. Data considered were
(i) arm entries and their order; (ii) total trial time; (iii)
first entry latency. With those data, the number of
working memory (WM) errors was counted. Every entry
in an already visited arm was considered as a WM error.
A non-visited arm was also considered as an error.
Animals were classified according to their performances:
first, by decreasing value of the number of visited arms
(8 to 0); secondly, by ascending value of the number of
WM errors (0 to x); third, by descending value of the
first error rank (9 to 0, with 9 equivalent to 8 arms
without error); and finally, by total trial time. A rank was
then allocated to each rat (1 to 31). Statistics were
performed on three parameters: (i) total trial time, (ii)
total number of errors (WM errors + number of non-
visited arms), (iii) rank.
2.1.4.1. Habituation. This habituation allowed accustom-
ing the rats to the maze and to eat pellets at the end of arms.
All rats were sham treated (NaCl 0.09% i.p. 1 ml/kg). All
arms were maintained baited during all the trial time. On the
first day, some pellets were placed along the maze to invite
animal to go to the end of the arms. Doors and automated
rods were regularly activated to accustom rat to movements
and induced noise.
2.1.4.2. Learning task. Our protocol was an Olton’s
procedure [31,45] with each arm baited only one time.
Each arm had to be visited only once within a trial. A trial
was finished when: (i) 10 min past or (ii) animal visited the
8 arms at least one time. The learning task was stopped
after 15 trial days or when the following criterion was
reached. Criterion consisted in doing either no error for 8
entries or at maximum 1 error for 9 entries, during 3
consecutive days.
2.1.4.3. Test task. This protocol aimed at testing the
memorising of the task and the impact of SHB treatment.
The day after the end of learning task, NaCl and SMB
groups were treated with SHB, and SHB groups were
treated with NaCl, excluding animals that reached the
criterion. This protocol allowed attesting the efficiency of
treatment: SHB animals that had not learned should be able
to reach the criterion under NaCl treatment.
2.1.5. Statistical analysis
All statistical analyses were performed with SPSS 11.5
software.
 E. Brillaud et al. / Brain Research 1054 (2005) 174 – 182
Animals did not have the same number of learning trial
days (15 or less if criterion was reached). Statistics of
learning task were then performed on results of each two
first days (D1 and D2) and each last 3 days (Dn-2, Dn-1,
Dn). To analyse test task, the last day of learning task (Dn)
was compared to the last day of test task (Te6).
A one-way ANOVA was first performed on control
groups (EE_NaCl, NE_NaCl, EE_SMB, NE_SMB; n = 4).
According to the results, control animals were grouped by
treatment conditions (NaCl, SMB; n = 8). A two-way
ANOVA (‘‘day’’ and ‘‘treatment’’ factors) was performed on
control and SHB groups. A one-way ANOVA was carried
out to compare control and SHB groups day by day. For
each ANOVA, the beta error was calculated (according to a
risk of 5%, in percentage), and a Bonferroni post hoc test
was performed if necessary.
One animal of the NE_SHB group was not considered
because he died (unknown cause after autopsy) during the
night after the 12th training day.
2.2. Plus maze and weight evolution
2.2.1. Animals
Twenty four male Sprague – Dawley rats (Iffa Credo,
France), weighing 160 –180 g (6 to 7 weeks old) on arrival
to the laboratory, were used. Environmental ambient
conditions were described previously.
2.2.2. Experimental groups and treatment
2.2.2.1. Housing. Half (n = 12) were housed in EE
conditions and others in standard cages. Housing conditions
were described previously.
2.2.2.2. Treatment. Animals were divided in 4 groups (n =
6), 2 with EE and 2 with NE. Two different treatments
were used. Rats were injected i.p. 30 min before the test
with: (i) sodium chloride solution at 0.9% (1 ml/kg; NaCl;
Sigma Aldrich) as control group (EE and NE); (ii) SHB
0.1 mg/kg (1 ml/kg; Sigma Aldrich) as test group (EE and
NE).
2.2.3. Apparatus
The plus-maze apparatus is composed of four arms (50 
10 cm) placed in cross at right angle around an elevated
central platform (10  10 cm, 70 cm above the floor). 40 cm
height walls border two arms facing. The two others are
open, without lateral border. The apparatus was isolated
from the experimenter in same conditions than radial maze.
Animals were observed in the maze through a camera fixed
at the ceiling.
2.2.4. Behavioural protocol
After 5 habituation days to animal facilities without
manipulation, animals were habituated to an experimenter
at day 5. On day 6, each animal was placed at the centre
177
of the plus maze, facing the open arm on the opposite of
experimenter position and was allowed to freely explore
the maze during 8 min. The experimenter recorded
behavioural data (latency and activity). Non-anxiety index
was linked to the total time spent in open arms and at
the extremity of the open arms (segment 3). The number
of runs in all arms was also recorded to assess
exploratory level. The maze was cleaned between each
animal with water and absorbing paper to minimise
olfactory impact.
General behaviour was observed. The weight of rats was
measured on days 5 and 6 as a well being index.
2.2.5. Statistical analysis
All statistical analyses were performed with SPSS 11.5
software.
General effects of ‘‘treatment’’ and ‘‘environment’’
factors were measured using a two-way ANOVA. A one-
way ANOVA was then done to compare each group. The
beta error was calculated, and a Bonferroni post hoc test was
performed if necessary.
2.3. Results
2.3.1. Learning task
Concerning control groups, no difference was found for
the three parameters (total trial time, total number of errors,
rank; P > 0.5; b < 10%) between both EE and NE groups.
Control animals could be then grouped by treatment (NaCl,
SMB; n = 8).
Comparison of control and SHB groups, enriched or not,
revealed a ‘‘treatment’’ effect on the three parameters (P <
0.001, b > 95%). A ‘‘day’’ effect was observed only on the
total number of errors (P = 0.007, b = 88%), and a tendency
was observed over time (P = 0.102; b = 58%).
Two by two comparison showed a general difference
between EE _SHB and control groups for the three
parameters (P < 0.02 to 0.001). For the NE_SHB groups,
differences with control groups were found for time and
rank (P < 0.003 and 0.0001), but concerning the number of
errors, it was only different from the EE_SHB group (P =
0.016). Moreover, a significant difference was obtained
between days D2 and Dn (P = 0.003) concerning the
number of errors.
The one-way ANOVA revealed a ‘‘group’’ effect on time
at D1 (P = 0.031; b = 71%) and D2 (P = 0.007; b = 87%),
on the number of errors at D2 (P = 0.017; b = 78%) and on
rank for all the days (P < 0.05 to 0.0001; b = 65 to 99%).
The observation in Fig. 2 completed by the Bonferroni test
results showed that rats of SHB groups spent more time to
do the task (A) (D1 and D2: EE_SHB # NaCl, P < 0.05),
did more errors during trials and the number of errors
decreased with training days (B) and had worse ranks than
control groups (C) (D1: SHB # controls, P < 0.05 to 0.001;
D2 and Dn: NE_SHB # controls, P < 0.05; Dn-1 and Dn-2:
NE_SHB # SMB, P < 0.05). Graphs showed the tendency
178 E. Brillaud et al. / Brain Research 1054 (2005) 174 – 182

(P < 0.01, b > 82%). A tendency was observed at day Te6

on time (P = 0.072, b = 57.3%).
The observation in Fig. 3 completed by the Bonferroni
test results showed the reversing of rats’ performances
with treatment change. SHB and controls results were
opposed. SHB animals then treated with NaCl decreased
their time spent by trial (A) (Te6: tendency EE_SHB #
EE_control, P = 0.060), their number of errors (B) (Te6:
EE_SHB # EE_control, P = 0.032) and their rank (C) (Dn:
NE_SHB # NE_control, P = 0.014, tendency NE_SHB #
EE_control, P = 0.059; Te6: EE_SHB # EE_control, P =
0.006, tendency EE_SHB # NE_control, P = 0.071) at day
Te6 compared to Dn. Inversely for the controls animals
treated with SHB.

2.3.3. Plus maze and weigh evolution

No significant ‘‘treatment’’ effect was shown. General
tendencies were observed on time spent in open arms (P =
0.116, b = 34.7%), time spent in segment 3 (P = 0.102, b =
37.1%) and total number of visits (P = 0.131, b = 32.3%).
Even if Fig. 4 seems to show that EE animals spent more
time in open arms and segment 3, no ‘‘environment’’ effect

Fig. 2. Learning task: performances of rats (mean T SEM) grouped by

treatment (n = 8) represented for first 2 days and last 3 days of training; (A)
total time spent to do the trial; (B) number of errors; (C) rank.

that NE_SHB group spent more time to do the task, did

more error the first days (D2: NE_SHB # EE_SHB, P =
0.028) and have fairer ranks the last days than EE_SHB
animals.

2.3.2. Test task

After two by two comparison as described before, no
difference was found (P > 0.05; b < 20%), and control
animals were grouped by environmental condition (EE_con-
trol, NE_control).
The two-ways ANOVA revealed a ‘‘treatment by day’’
interaction on the number of errors and rank (P = 0.003
and < 0.001, b > 90%) and a tendency on time (P = 0.078,
b = 56.8%). A ‘‘day’’ effect was observed on the number
of errors (P = 0.049, b = 50.8%), and a tendency of a
‘‘treatment’’ effect was noted on the number of errors and
rank (P = 0.061 and 0.055, b > 60%). Fig. 3. Test task: evolution of performances (mean T SEM) of rats grouped
The one-way ANOVA revealed a ‘‘group’’ effect for the 2 by environment (n = 6 to 8) comparing last day of training and last day of
days on the number of errors (P < 0.05, b > 66%) and rank test; (A) total time spent to do the trial; (B) number of errors; (C) rank.
 E. Brillaud et al. / Brain Research 1054 (2005) 174 – 182
Fig. 4. Results of the plus maze test (mean T SEM); (A) total number of
visits; (B) total time spent in open arms and segment 3 of these.
was found (P > 0.43, b < 12%) except a tendency on the
number of falls (P = 0.078, b = 42.5%). As shown in Fig.
4, rats of SHB groups seem to do more visits and spend
more time in open arms and in the third segment in
comparison to NaCl animals. An ANOVA on weight
evolution (see Table 1) showed an ‘‘environment’’ effect at
day 6 (P = 0.012, b = 80.2%).
2.4. Discussion
2.4.1. Treatment effects and validation of the radial
maze task
2.4.1.1. Learning task. The training evolution, as expected,
differed according to treatments. A small effect of SMB
treatment could be noted on the total trial due to the
peripheral disturbances (motor disorganisation). No clear
difference was detected between the two control groups
(SMB and NaCl). Peripheral scopolamine effects did not
significantly influence performances in the test. Central level
action of scopolamine (SHB treatment) involved a more
significant number of errors, longer test duration and
superior ranks in comparison to control animals. The
decreasing number of errors in the ‘‘day’’ effect has shown
a time course learning effect for each treatment. As expected,
a low dose of SHB decreases working memory performances
but does not affect spatial memory [21]. These results attest
the validity of our test: it allows characterising a learning
effect and also a difference of training results between two
groups (in case of learning delay).
179
2.4.1.2. Test task. All groups’ data were reversed between
the last day of training and the last day of test (strong
interaction between ‘‘treatment by day’’ for all parameters).
Animals first treated with SHB and then with NaCl
improved quickly their results and could reach the criterion.
For control groups (NaCl and SMB) then treated with SHB,
performances decreased. Effects observed (learning delay,
lapse) were effectively due to SHB treatment and central
cholinergic effects of scopolamine. Our test was able to
show a delay of learning and a memorisation deficit after
training during a test task.
2.4.1.3. Plus maze. In the plus maze task, no ‘‘treatment’’
effect was clearly found on anxiety or exploratory behav-
iour. Tendencies and graphs suggested a difference between
SHB and NaCl treated animals. It was not significant due to
the intra-group variations (b > 20%). The low SHB dose
(0.1 mg/kg) used did not disturb our plus-maze results.
Radial maze performances of SHB-treated animals were due
to a deficit learning effect and not to an anxiety effect.
SHB treatment seemed to increase exploratory behaviour
(tendency on total number of visits). Anxiety level appeared
to be decreased (tendency on time spent in open arms and
third segments). This could be explain by an increase of the
exploratory behaviour: SHB-treated animals had a tendency
to visit more arms. Animals did not visit more open arms;
they did more complete visits, going to the end of arms
(third segments).
2.4.2. EE effects
EE conditions were chosen according to our future
studies. In some toxicological studies, animals are restrained
in Plexiglas rockets, which consist of a cylinder (7 cm
diameter, 15 cm length) and a truncated cone (3 cm length)
where animals place their head. These restraint conditions
are stressful for animals and can lead to erroneous results in
biochemical and behavioural studies. To decrease the stress
level, animals should be accustomed to the rocket before
exposure. Placing a plastic tube in the cage facilitates the
restraint habituation.
2.4.2.1. Learning task. Concerning the environmental
enrichment effect, no difference was found between control
groups (NaCl and SMB). A slight difference was noticed
between EE and NE animals of SHB-treated groups.
EE_SHB animals seemed to spend more time and to make
more errors in first days of the task, but this observation
Table 1
Animal’s weight evolution (mean T SEM) during the 2 days of plus-maze
experiment (weight range on arrival day was a supplier data)
Weight evolution (g) Arrival day Day 5 Day 6
Mean T SEM Mean T SEM
Enriched (n = 18) 160 – 180 211.4 5.6 214.1 4.1
Non-enriched (n = 12) 215.0 7.4 226.5 8.1
180 E. Brillaud et al. / Brain Research 1054 (2005) 174 – 182
(significant on day 2 for number of errors) was reversed on
the last 3 days of training. This difference could be
explained by anxiety behaviour. In the presence of new
environmental conditions, animals could be anxious, much
more if it is an unclosed high space as the radial maze is
[30,49]. Anxiety is related to novelty, so this effect
decreases with trials.
Enrichment can also increase exploratory behaviour
[7,58]. EE_SHB animals made more visits (frequently more
than 16/trial, limits not exceeded by other groups) involving
a greater number of errors (due to treatment). Visits were
more frequently complete (until the end of the arm) than
NE_SHB animals. Having visited more could explain
quicker understanding of task: 3 animals/8 of EE_SHB
group reached criterion in 15 days against none of NE_SHB
group.
2.4.2.2. Test task. As regards to environmental condition,
evolution of SHB then NaCl animals did not depend on the
environment. Both groups were similar (equivalent curve
slopes) in spite of a non-equivalent starting level. For
control animals then treated with SHB, EE ones increased
their deficits compared to NE (slopes of time and errors
curves higher). Still, this difference came from that enriched
animals visited many more arms per trial without succeed-
ing more quickly than non-enriched ones.
2.4.2.3. Plus maze and weight evolution. In this test, no
‘‘environment’’ effect was clearly found on anxiety or
exploratory behaviour. EE animals seemed to be less
anxious than NE ones. They past more time in open arms
and segment 3. Fact that only some NE animals (3/12) fell
from the maze could show a better agility of EE animals due
to elements placed in cages. The number of visits was not
increased significantly in that case.
Enriched animals’ weight increase was less consequent
than NE ones, with a smaller intra-group variation. Animals
were more active and less stressed.
2.4.3. General discussion
The learning task, testing short time memory, and the test
task in a radial maze are validated in our laboratory. In
future studies, effects on learning and/or memorising can be
characterised as specific ones. Bringing the enriched
environment presented, even if modest, improved perform-
ances during the training task of animals having a learning
deficit. On the other hand, amnesic deficits during test task
were accentuated. Limiting the number of visits per trial (16
commonly allowed) can attenuate these effects. Indeed, as
expected, the small increase of the environmental complex-
ity did not affect behaviour of powerful animals (control
ones) in this radial maze task. It will be possible to compare
our future results to published ones.
Plus maze results and weigh evolution show that animals
were less stressed. These conclusions were comforted by the
general observation of rats: coat aspect, facility of con-
tention and behaviour during i.p. injections. Even if
animal’s wellness is difficult to evaluate, benefits seemed
important. This modest enrichment will be used systemati-
cally in our future neurotoxicological studies to decrease the
stress level due to restraint and to increase the well being of
single housing condition.
One animal (NE-SHB group) was excluded because he
died at the 12th day of the learning task. His first results
were in concordance with those obtained in his group (no
criterion at day 12). This animal would not have modified
results obtained. Exclusion was performed because of
statistical problem (no complete values). It would have
been possible to evaluate missing data using average of
others animals.
In conclusion, the aim of this enrichment, chosen in
accordance to other laboratory studies, was to improve
comfort and to decrease stress of animals with the least
disturbance of the radial maze test results. These conditions
seem to be reached, involving only a small correction to our
protocol (16 per trial limit). EE is a good and simple
solution to improve life quality of laboratory animals. It is
especially important concerning long test periods as
toxicological chronic studies. With small non-expensive
elements, we are able to decrease the stress of animals
without modifying sensitivity of behavioural tests. This
study shows that it is easy to create an enriched environment
without modifying the maintenance of the animals’ facili-
ties. This objective is in accordance with current general
ethic of concerns on welfare of animals [57].
Acknowledgments
The authors want to acknowledge Frédéric BOIS, Celine
BROCHOT and Cheick DIACK for English support. This
research was supported by the European project RAMP
2001 (CE n- QLK4-CT-2001-00463) and the French
ministry of Ecology and Sustainable Development, (BCRD
n- CV02000013).
References
[1] N. Barnard, S. Hou, Inherent stress: the tough life in lab routine, Lab.
Anim. 17 (1988) 21 – 27.
[2] B.V. Beaver, Environmental enrichment for laboratory animals, ILAR
News 31 (1989) 5 – 11.
[3] E.E. Belz, J.S. Kennell, R.K. Czambel, R.T. Rubin, M.E. Rhodes,
Environmental enrichment lowers stress-responsive hormones in
singly housed male and female rats, Pharmacol. Biochem. Behav. 76
(2003) 481 – 486.
[4] A. Blokland, Acetylcholine: a neurotransmitter for learning and
memory? Brain Res. Rev. 21 (1995) 285 – 300.
[5] O. Buresova, J. Bures, Radial maze as a tool for assessing the effect of
drugs on the working memory of rats, Psychopharmacology (Berlin)
77 (1982) 268 – 271.
[6] D.J. Chmiel Jr., M. Noonan, Preference of laboratory rats for
potentially enriching stimulus objects, Lab. Anim. 30 (1996) 97 – 101.
 E. Brillaud et al. / Brain Research 1054 (2005) 174 – 182
[7] P. Clausing, H.K. Mottles, B. Opitz, S. Kormann, Differential effects
of communal rearing and preweaning handling on open-field
behavior and hot-plate latencies in mice, Behav. Brain Res. 82
(1997) 179 – 184.
[8] J.M. Daniel, S.L. Roberts, G.P. Dohanich, Effects of ovarian hormones
and environment on radial maze and water maze performance of
female rats, Physiol. Behav. 66 (1999) 11 – 20.
[9] M.S. Dawkins, Behavioural deprivation: a central problem in animal
welfare, Appl. Anim. Behav. Sci. 20 (1988) 209 – 225.
[10] M.C. Diamond, Response of the brain to enrichment, An. Acad. Bras.
Cienc. 73 (2001) 211 – 220.
[11] Directive 86/609/EEC, European Union, O.J. L 358 (1986).
[12] Directive 2004/73/EC, European Community, O.J. L152 2004
(2004).
[13] D.A. Eckerman, W.A. Gordon, J.D. Edwards, R.C. MacPhail, M.I.
Gage, Effects of scopolamine, pentobarbital, and amphetamine on
radial arm maze performance in the rat, Pharmacol. Biochem. Behav.
12 (1980) 595 – 602.
[14] D. Einon, Environmental influences and exploration in the rat, Appl.
Anim. Ethol. 6 (1980) 384.
[15] D. Einon, Spatial memory and response strategies in rats: age, sex and
rearing differences in performance, Q. J. Exp. Psychol. 32 (1980)
473 – 489.
[16] S.E. File, The interplay of learning and anxiety in the elevated plus-
maze, Behav. Brain Res. 58 (1993) 199 – 202.
[17] R.E. Fitzgerald, M. Berres, U. Schaeppi, Validation of a radial maze
test for assessing learning and memory in rats, Toxicology 49 (1988)
425 – 432.
[18] E.B. Gardner, J.J. Boitano, N.S. Mancino, D.P. D’Amico, Environ-
mental enrichment and deprivation: effects on learning, memory and
exploration, Physiol. Behav. 14 (1975) 321 – 327.
[19] O.L. Gobbo, S.M. O’Mara, Impact of enriched-environment housing
on brain-derived neurotrophic factor and on cognitive performance
after a transient global ischemia, Behav. Brain Res. 152 (2004)
231 – 241.
[20] W.T. Greenough, T.C. Madden, T.B. Fleischmann, Effects of isolation,
daily handling, and enriched rearing on maze learning, Psychon. Sci.
27 (1972) 279 – 280.
[21] R. Grette Lydon, Shinshu Nakajima, Differential effects of scopol-
amine on working and reference memory depend upon level of
training, Pharmacol. Biochem. Behav. 43 (1992) 645 – 650.
[22] F.S. Hall, Social deprivation of neonatal, adolescent, and adult rats has
distinct neurochemical and behavioral consequences, Crit. Rev.
Neurobiol. 12 (1998) 129 – 162.
[23] A.H.J. Herremans, T.H. Hijzen, B. Olivier, J.L. Slangen, Cholinergic
drug effects on a delayed conditional discrimination task in the rat,
Behav. Neurosci. 109 (1995) 426 – 435.
[24] A. Higashida, N. Ogawa, Differences in the acquisition process and
the effect of scopolamine on radial maze performance in three strains
of rats, Pharmacol. Biochem. Behav. 27 (1987) 483 – 489.
[25] M. Jain, A.L. Baldwin, Are laboratory animals stressed by their
housing environment and are investigators aware that this stress can
affect physiological data? Med. Hypotheses 60 (2003) 284 – 289.
[26] P. Jensen, F.M. Toates, Who needs Fbehavioural needs_? Motivational
aspects of the needs of animals, Appl. Anim. Behav. Sci. 37 (1993)
161 – 181.
[27] R.G. Lister, The use of a plus-maze to measure anxiety in the mouse,
Psychopharmacology (Berlin) 92 (1987) 180 – 185.
[28] L. Martini, R.N. Lorenzini, S. Cinotti, M. Fini, G. Giavaresi, R.
Giardino, Evaluation of pain and stress levels of animals used in
experimental research, J. Surg. Res. 88 (2000) 114 – 119.
[29] F. Moncek, R. Duncko, B.B. Johansson, D. Jezova, Effect of
environmental enrichment on stress related systems in rats, J. Neuro-
endocrinol. 16 (2004) 423 – 431.
[30] K.C. Montgomery, The relation between fear induced by novel
stimulation and exploratory drive, J. Comp. Physiol. Psychol. 48
(1955) 254 – 260.
181
[31] D.S. Olton, The radial arm maze as a tool in behavioral pharmacology,
Physiol. Behav. 40 (1987) 793 – 797.
[32] Order in council 2001-464, French Agriculture Ministry, O.J. 31 May,
2001 (2001).
[33] M.J. Passineau, E.J. Green, W.D. Dietrich, Therapeutic effects of
environmental enrichment on cognitive function and tissue integrity
following severe traumatic brain injury in rats, Exp. Neurol. 168
(2001) 373 – 384.
[34] E.G. Patterson-Kane, Cage size preference in rats in the laboratory,
J. Appl. Anim. Welf. Sci. 5 (2002) 63 – 72.
[35] E.G. Patterson-Kane, D.N. Harper, M. Hunt, The cage preferences of
laboratory rats, Lab. Anim. 35 (2001) 74 – 79.
[36] S. Pellow, P. Chopin, S.E. File, M. Briley, Validation of open: closed
arm entries in an elevated plus-maze as a measure of anxiety in the rat,
J. Neurosci. Methods 14 (1985) 149 – 167.
[37] T.M. Pham, S. Soderstrom, B. Winblad, A.H. Mohammed, Effects
of environmental enrichment on cognitive function and hippo-
campal NGF in the non-handled rats, Behav. Brain Res. 103 (1999)
63 – 70.
[38] J.J. Pilcher, G.R. Sessions, S.A. Mcbride, Scopolamine impairs
spatial working memory in the radial maze: an analysis by error
type and arm choice 1, Pharmacol. Biochem. Behav. 58 (1997)
449 – 459.
[39] T.E. Reed, Effect of enriched (complex) environment on nerve
conduction velocity: new data and review of implications for
the speed of information processing, Intelligence 17 (1993)
533 – 540.
[40] M.R. Rosenzweig, E.L. Bennett, M. Hebert, H. Morimoto, Social
grouping cannot account for cerebral effects of enriched environments,
Brain Res. 153 (1978) 563 – 576.
[41] P.J. Rowsey, C.J. Gordon, A peripheral mechanism of fever: differ-
ential sensitivity to the antipyretic action of methyl scopolamine,
Auton. Neurosci. 85 (2000) 148 – 155.
[42] W. Scharmann, Improved housing of mice, rats and guinea-pigs: a
contribution to the refinement of animal experiments, Altern. Lab.
Anim. 19 (1991) 108 – 114.
[43] N.C.A. Schrijver, N.I. Bahr, I.C. Weiss, H. Wurbel, Dissociable effects
of isolation rearing and environmental enrichment on exploration,
spatial learning and HPA activity in adult rats, Pharmacol. Biochem.
Behav. 73 (2002) 209 – 224.
[44] N.C.A. Schrijver, P.N. Pallier, V.J. Brown, H. Wurbel, Double
dissociation of social and environmental stimulation on spatial
learning and reversal learning in rats, Behav. Brain Res. 152 (2004)
307 – 314.
[45] J.L. Slangen, B. Earley, R. Jaffard, M. Richelle, D.S. Olton,
Behavioral models of memory and amnesia, Pharmacopsychiatry 23
(Suppl. 2) (1990) 81 – 83 (discussion 84).
[46] M. Soffie, K. Hahn, E. Terao, F. Eclancher, Behavioural and glial
changes in old rats following environmental enrichment, Behav. Brain
Res. 101 (1999) 37 – 49.
[47] R. Stevens, Scopolamine impairs spatial maze performance in rats,
Physiol. Behav. 27 (1981) 385 – 386.
[48] D.R. Studelska, E.D. Kemble, Effects of briefly experienced environ-
mental complexity on open-field behavior in rats, Behav. Neural Biol.
26 (1979) 492 – 496.
[49] D. Treit, J. Menard, C. Royan, Anxiogenic stimuli in the elevated plus-
maze, Pharmacol. Biochem. Behav. 44 (1993) 463 – 469.
[50] J.E. Van der Harst, A.-M. Baars, B.M. Spruijt, Standard housed rats
are more sensitive to rewards than enriched housed rats as reflected
by their anticipatory behaviour, Behav. Brain Res. 142 (2003)
151 – 156.
[51] J.E. Van der Harst, P.C.J. Fermont, A.E. Bilstra, B.M. Spruijt, Access
to enriched housing is rewarding to rats as reflected by their
anticipatory behaviour, Anim. Behav. 66 (2003) 493 – 504.
[52] H. Van Praag, G. Kempermann, F.H. Gage, Neural consequences
of environmental enrichment, Nat. Rev. Neurosci. 1 (2000)
191 – 198.
182 E. Brillaud et al. / Brain Research 1054 (2005) 174 – 182

[53] M. Van Waas, M. Soffie, Differential environmental modulations on
locomotor activity, exploration and spatial behaviour in young and old
rats, Physiol. Behav. 59 (1996) 265 – 271.
[54] A.P. Viscardi, G.A. Heise, Effects of scopolamine on components of
delayed response performance in the rat, Pharmacol. Biochem. Behav.
25 (1986) 633 – 639.
[55] J. Watts, R. Stevens, C. Robinson, Effects of scopolamine on radial
maze performance in rats, Physiol. Behav. 26 (1981) 845 – 851.
[56] D.R. Widman, R.A. Rosellini, Restricted daily exposure to environ-
mental enrichment increases the diversity of exploration, Physiol.
Behav. 47 (1990) 57 – 62.
[57] H. Wurbel, Ideal homes? Housing effects on rodent brain and
behaviour, Trends Neurosci. 24 (2001) 207 – 211.
[58] A. Zimmermann, M. Stauffacher, W. Langhans, H. Wurbel, Enrich-
ment-dependent differences in novelty exploration in rats can be
explained by habituation, Behav. Brain Res. 121 (2001) 11 – 20.