International Journal of Neuropsychopharmacology (2010), 13, 997–1009. Copyright f CINP 2010
doi:10.1017/S1461145710000039
ARTICLE
THEMATIC SECTION
A cognitive deficit induced in rats by chronic
intermittent cold stress is reversed by
chronic antidepressant treatment
Memory and Cognition
M. Danet S. Lapiz-Bluhm and David A. Morilak
Department of Pharmacology and Center for Biomedical Neuroscience, University of Texas Health Science Center, San Antonio,
TX, USA
Abstract
We have previously reported that 14-d chronic intermittent cold (CIC) stress induced a cognitive deficit in
reversal learning on the rat attentional set-shifting test. This effect may be related to dysregulation of 5-HT
function in orbitofrontal cortex, as a model of cognitive dysfunction in depression. To test the ability
of chronic antidepressant drug treatment to reverse the cognitive deficit induced by CIC, it was first
necessary to assess the temporal characteristics of the CIC-induced cognitive deficit. Thus, in the first
experiment, we assessed the duration of the cognitive deficit following 2-wk CIC stress. Replicating
previous experiments, CIC induced a reversal learning deficit tested 3 d after the last cold exposure.
However, cognitive performance of CIC-stressed rats was no different from unstressed controls when
tested 7, 14 or 21 d after termination of the stress treatment. We next compared behaviour 3 d after 2-wk
CIC to that seen 3 d after 5-wk CIC, and found similar deficits in reversal learning. Thus, in the final
experiment, antidepressant drug treatment was initiated after 2-wk CIC stress, and was maintained for
3 wk, concurrent with the continuation of CIC stress. Both chronic and acute treatment with the selective
serotonin reuptake inhibitor, citalopram, but not the norepinephrine reuptake blocker, desipramine,
reversed the cognitive deficit induced by CIC stress. Thus, this stress-induced cognitive deficit may be a
useful model for cognitive deficits related to prefrontal cortical hypoactivity in depression, and for investigating neurobiological mechanisms underlying the beneficial effects of chronic antidepressant drug
treatment.
Received 1 May 2009 ; Reviewed 13 June 2009 ; Revised 4 November 2009 ; Accepted 30 December 2009 ;
First published online 11 February 2010
Key words : Antidepressants, chronic stress, cognition, depression, prefrontal cortex.
Introduction
Chronic stress is a risk factor for depression and dysregulation of brain monoaminergic neurotransmitters
(Anisman & Zacharko, 1986). Monoamines play a role
in state-related modulation of executive function, and
disturbances of executive function have been implicated in depression (Beck, 1976 ; Beck et al. 1987).
Deficits in executive function, such as impaired cognitive set-shifting, behavioural inflexibility and perseveration (see Fossati et al. 1999), are consistent with
the perseverative negative ruminations and emotional
Address for correspondence : Dr D. A. Morilak, Department of
Pharmacology, MC 7764, University of Texas Health Science Center at
San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229-3900,
USA.
Tel. : 1-210-567-4174 Fax : 1-210-567-4303
Email : [email protected]
biases present in depression (Murphy et al. 2003,
1999), and are associated with prefrontal cortical
hypoactivity (Rogers et al. 2004 ; Sheline, 2003), which
resolves with successful antidepressant treatment
(Goldapple et al. 2004 ; Kennedy et al. 2001 ; Prasko et al.
2004).
In previous studies, we have used an attentional
set-shifting test (AST) for rats (Birrell & Brown, 2000),
to address modulatory effects of monoamines on
cognitive function in prefrontal cortex, and possible
mechanisms underlying chronic stress-induced cognitive dysfunction. In the AST, adapted from tests of
cognitive flexibility in non-human primates (Roberts
et al. 1992), and humans (Beats et al. 1996), rats learn a
series of contingencies in which cues within one of
two stimulus dimensions (odour or texture) signal
the location of a food reward. Once they master the
contingency on each successive task, the rules are
998
M. D. S. Lapiz-Bluhm and D. A. Morilak
changed, and they must suppress the previously successful strategy to learn the new rule and proceed
through the next task. The manner in which the
rules are changed allows an assessment of different
forms of cognitive flexibility, each dependent to varying degrees on the integrity of different subregions
of prefrontal cortex (see Birrell & Brown, 2000 ; Bondi
et al. 2008 ; Lapiz & Morilak, 2006 ; Lapiz et al. 2007 ;
Lapiz-Bluhm et al. 2009 ; McAlonan & Brown, 2003).
Lesions of medial prefrontal cortex (mPFC) caused
selective impairment of extra-dimensional (ED) cognitive set-shifting (Birrell & Brown, 2000), whereas
lesions of orbitofrontal cortex (OFC) resulted in selective impairment of reversal learning (McAlonan &
Brown, 2003). Previous studies also indicate that these
cognitive processes are modulated by monoaminergic
neurotransmission. Increasing noradrenergic transmission in mPFC facilitated ED set-shifting (Lapiz &
Morilak, 2006 ; Lapiz et al. 2007), and serotonergic
transmission in OFC selectively facilitated reversal
learning (Clarke et al. 2004, 2005, 2007).
We have used the AST previously to study chronic
stress-induced changes in cognitive flexibility relevant
to depression. Chronic unpredictable stress (CUS), a
psychogenic stressor, impaired ED set-shifting and
reversal learning, and these impairments were prevented by concurrent treatment with desipramine
(Des), a selective norepinephrine (NE) reuptake
blocker (Bondi et al. 2008). By contrast, chronic intermittent cold (CIC) stress, a metabolic stressor, induced
a reproducible and selective impairment in reversal
learning that was mimicked by central serotonin
(5-HT) depletion (Lapiz-Bluhm et al. 2009). Extracellular levels of 5-HT, measured by microdialysis in
the OFC during testing, were reduced in CIC-stressed
rats, and the CIC stress-induced reversal learning
deficit was attenuated by elevating serotonergic neurotransmission with an acute administration of citalopram (Cit), a selective serotonin reuptake inhibitor
(SSRI) (Lapiz-Bluhm et al. 2009). The effect of chronic
SSRI treatment has not yet been tested.
In these previous studies, reuptake blockers were
administered either acutely, immediately prior to
testing, or chronically by osmotic minipump using a
preventive strategy, in which drug treatment was
initiated prior to the beginning of stress, and was maintained during the course of chronic stress treatment
until testing. However, it would be more relevant to
the human clinical situation, and would allow a more
informative investigation of adaptive mechanisms
underlying chronic antidepressant drug effects, if
treatment could be initiated after the targeted behavioural effect of stress had been established, in order to
test its ability to reverse rather than prevent the stressinduced cognitive deficit. Towards this end, the purpose of the first experiment in this paper was to
characterize the duration of the reversal learning
deficit induced after 2-wk CIC stress. This experiment
replicated the selective deficit in reversal learning
shown previously when tested 3 d, but not 7, 14 or
21 d after the final cold stress exposure, indicating
that the duration of effect was insufficient for testing
the efficacy of 3-wk antidepressant drug treatment
initiated after the termination of stress. Thus, in the
second experiment, an alternate approach was tested,
extending CIC stress treatment to 5 wk, which would
allow a 3-wk drug treatment to be initiated after the
cognitive deficit had been established at 2 wk. In this
experiment, a selective reversal learning deficit was
seen after 5-wk CIC stress, similar to that after 2 wk.
Thus, the 5-wk CIC treatment protocol was used in the
third experiment to test the ability of chronic treatment
with a SSRI, Cit, or a selective NE reuptake inhibitor,
Des, initiated after 2-wk CIC stress and maintained
while stress treatment was continued for an additional
3 wk, to reverse the cognitive deficit induced by CIC
stress. Portions of this work have been presented in
abstract form (Lapiz-Bluhm et al. 2008).
Methods
Animals
A total of 221 adult male Sprague–Dawley rats,
weighing 220–240 g on arrival, were initially grouphoused (3 per cage) in 25r45r15 cm plastic cages, on
a 12-h light/dark cycle (lights on 07:00 hours), with
food and water available ad libitum. After acclimatizing
for at least 7 d, rats were housed individually prior
to any experimental manipulation. Experiments were
conducted during the light phase of the cycle, between
09:00 and 17:00 hours. All procedures were reviewed
and approved by the Institutional Animal Care and
Use Committee of the University of Texas Health
Science Center at San Antonio, and were consistent
with NIH guidelines for the care and use of laboratory
animals. All efforts were made to minimize pain, discomfort, suffering, and the number of rats used.
CIC stress
The procedure for CIC stress exposure was as
described previously (Lapiz-Bluhm et al. 2009 ; Ma &
Morilak, 2005). Rats were randomly assigned to control or CIC stress conditions. Rats in the CIC stress
group were transported in their home cages, with
food, water and bedding, into a cold room (4 xC) for
Antidepressants and cognitive effects of stress
999
Table 1. Behavioural protocol for the attentional set-shifting test (AST)
Dimensions
Example combinations
Discrimination stage
Relevant
Irrelevant
SD
CD
Odour
Odour
Medium
Reversal 1 (R1)
Odour
Medium
ID shift
Odour
Medium
Reversal 2 (R2)
Odour
Medium
ED shift
Medium
Odour
Reversal 3 (R3)
Medium
Odour
(+)
(x)
Clove/Sawdust
Clove/Raffia
Clove/Metallic Filler
Nutmeg/Raffia
Nutmeg/Metallic Filler
Rosemary/Wood balls
Rosemary/Plastic beads
Cinnamon/Plastic beads
Cinnamon/Wood balls
Velvet/Citronella
Velvet/Thyme
Crepe/Thyme
Crepe/Citronella
Nutmeg/Sawdust
Nutmeg/Metallic Filler
Nutmeg/Raffia
Clove/Metallic Filler
Clove/Raffia
Cinnamon/Plastic beads
Cinnamon/Wood balls
Rosemary/Wood balls
Rosemary/Plastic beads
Crepe/Thyme
Crepe/Citronella
Velvet/Citronella
Velvet/Thyme
CD, Compound discrimination ; ED, extra-dimensional ; ID, intra-dimensional ; SD, simple discrimination.
Representative example of stimulus pairs, and the progression through stages of the AST protocol. In this example, odour was
the initial discriminative stimulus dimension, shifting to digging medium in the ED stage. For each stage, the positive
stimulus is in bold, and is paired randomly across trials with the two stimuli from the irrelevant dimension.
6 h, then returned to the housing room, every day for
14 consecutive days (expts 1 and 2), or for 5 wk (expts
2 and 3). Control rats remained undisturbed in the
housing facility during this time. One week before
testing, rats were restricted to 14 g food per day, with
water freely available. After the CIC treatment period,
rats were taken through the 3-d AST protocol.
AST
The AST was conducted as described previously
(Lapiz & Morilak, 2006), in a rectangular wooden
arena (lengthrwidthrheight: 71r40r20 cm) painted
white on all surfaces. A removable divider separated
one-third of the arena, forming a start box which also
served as a holding area following each trial. To begin
a trial, a rat was placed in the start box, and the divider
lifted. A white Plexiglas panel divided the opposite
third of the arena into two sections. At testing, a terracotta digging bowl (internal rim diameter 7 cm,
depth 6 cm) was placed in each section. Each bowl was
defined by a pair of cues along two stimulus dimensions, the material with which the pot was filled and
an odour (see Table 1). To mark each pot with a distinct odour, two drops (20 ml) of scented aromatic oil
(Frontier Natural Brands, USA) was applied to
the inner rim 5 d before use, and 3–5 ml reapplied the
day before use. The bait, a 1/4 Honey Nut Cheerio
(General Mills Cereals, USA), was buried 2 cm below
the surface of the digging medium in the ‘ positive ’
pot. At the beginning of each task, a small quantity of
powdered Cheerio was sprinkled onto the medium in
both pots to prevent rats from locating the bait by
smell rather than learning the discrimination.
Digging was defined as vigorous displacement of
the medium to retrieve the reward. Simply investigating the rim or surface of the medium with paws
or snout without displacing material was not scored
as a ‘dig’. Thus, rats could access tactile, visual and
olfactory cues.
The procedure entailed 3 d.
Day 1 : Habituation. Rats were trained to dig in the pots
for food reward. Two unscented baited pots were
placed in the home cage for a series of three exposures,
5 min each, in which the bait was covered with
increasing amounts of sawdust. Once the rat was
digging reliably, it was placed in the test arena for
three trials to retrieve reward from both sawdust-filled
pots.
Day 2 : Training. Rats were trained on a series of
simple discriminations (SDs), to a criterion of six
consecutive correct trials. They first had to learn to
associate the food reward with an odour cue (lemon
vs. rosewood, both pots filled with sawdust). After
1000
M. D. S. Lapiz-Bluhm and D. A. Morilak
reaching criterion, they then had to discriminate by
the digging media (felt strips vs. shredded paper,
no odour). All rats were trained using the same stimuli
in the same order. Training stimuli were not used
again in testing trials.
Day 3 : Testing. Rats were tested on a series of discrimination tasks (see Table 1). When the rat reached
criterion of six consecutive correct trials, testing
proceeded to the next stage. The first stage was an
SD, involving only one stimulus dimension. Half the
rats in each group were required at this stage to
discriminate between two odours, only one of which
was associated with reward, with both pots filled with
sawdust. The other half began by discriminating the
digging media, with no odours applied to the pots (for
the sake of clarity, the remainder of this description
considers only the example beginning with odour
discrimination). The second task was a compound
discrimination (CD), where the second, irrelevant
stimulus was introduced. Only one odour was associated with reward, as in the SD, but two different
digging media were paired randomly with the odours.
The third stage was a reversal of this discrimination
(R1), in which the same odours and media were
used, and odour was still the relevant dimension, but
the negative odour from the previous stage was
now positive (i.e. associated with the reward), and
the positive odour from the previous stage was now
negative. The fourth stage was an intra-dimensional
(ID) shift, wherein odour was still the relevant
dimension but all new stimuli were introduced. The
fifth stage was a reversal of this discrimination (R2),
in which the previously negative odour was now
positive, as in R1. The sixth stage required an ED setshift, in which new stimuli were again introduced,
but this time the relevant dimension was also
changed, e.g. the digging medium became the relevant dimension and odour was now irrelevant.
Finally, the seventh stage was another reversal (R3),
where the previously negative medium was now
positive. Table 1 outlines the progression through
these stages, and provides examples of the cue combinations used. The dependent measure was trials
to criterion (TTC), the number of trials to reach
criterion of six consecutive correct responses at each
test stage.
Elevated plus maze (EPM) test
In expt 2, the EPM test was conducted as described
previously (Bondi et al. 2008), the day after the AST.
Rats were transported to the testing room and allowed
to acclimate for 15 min. The EPM had four white
plastic arms, 10r50 cm, oriented in the shape of a
cross, intersecting at a 10r10 cm platform. Two closed
arms situated opposite each other were enclosed by
48 cm walls, and the remaining two open arms had
no walls. The open arms were fitted with a 0.5 cm
clear plastic rim around the edges to prevent animals
from falling off. The maze was elevated 100 cm from
the floor, and testing took place under normal ambient
overhead lighting (200 lx in the open arms) with 60 dB
background white noise. To begin the 5-min test,
a rat was placed in the centre platform facing the
junction of a closed and open arm. Activity was
tracked using ANYmaze software (Stoelting Co.,
USA). Measures included time, number of entries,
and distance travelled in each area of the maze. Open
to total ratios (OTR) for time and entries (open/
open+closed) were calculated as indices of open-arm
exploration. Total distance travelled was analysed as
a measure of non-specific changes in locomotion independent of OTR. The maze was cleaned with 70 %
ethanol, then water, and dried completely before the
next test.
Expt 1 : duration of the cognitive effect induced by
14-d CIC stress exposure
Eighty-three rats were randomly assigned to two
groups : control or 14-d CIC stress. Seven days before
testing, food was restricted to 14 g/d. Separate groups
of control and CIC-stressed rats underwent habituation, training and testing as above, at 3, 7, 14 or 21 d
after the last cold exposure (n=8–12 per group at each
time-point).
Expt 2 : effects of 2-wk vs. 5-wk CIC stress exposure
To compare the effects of 2-wk and 5-wk CIC stress,
23 rats were randomly assigned to two groups : CIC
stress or control, then further subdivided into 2- or
5-wk treatment groups. CIC stress was conducted as
above. Seven days before testing, food was restricted
to 14 g/d. Following the last day of cold exposure,
rats were habituated and trained as above, and testing
was conducted 3 d after the last cold exposure. The
beginning of chronic stress treatment was staggered
such that animals from all groups were tested together. As there were no differences between the
two control groups, these were pooled into a single
unstressed control group, resulting in n=7–8 rats
in each of three groups (unstressed controls, 2-wk CIC,
5-wk CIC). The day following the AST, rats were
tested on the EPM.
Antidepressants and cognitive effects of stress
Expt 3 : effects of chronic antidepressant treatment on
the CIC stress-induced reversal learning deficit
Fifty-nine rats in two independent cohorts (n=30 and
n=29, respectively) were assigned to two treatment
groups, control or 5-wk CIC stress. These were each
further subdivided into two chronic drug treatments,
either Cit and its vehicle control group (n=7–9/
group), or Des and its vehicle control group (n=7–8/
group). Rats were exposed to 2-wk CIC stress as
above. On day 15, they were anaesthetized (43 mg/ml
ketamine, 1.4 mg/ml acepromazine, 8.6 mg/ml xylazine, in 1.0 ml/kg i.m.). Osmotic minipumps (model
2ML4, Alzet Corp., USA), containing citalopram hydrobromide (Shanco International, USA), desipramine
hydrochloride (Sigma, USA), or vehicle (10 % ethanol
in saline), were implanted intraperitoneally (i.p.). Cit
was delivered at a dose of 20 mg/kg.d free base, and
Des at 7.5 mg/kg.d free base. Rats were given antibiotic (penicillin G, 300 000 IU/ml/kg). Three days
after surgery, CIC stress treatment resumed for the
remainder of the 5-wk period. Seven days before testing, food was restricted to 14 g/d. Rats underwent
habituation and training as above, with testing conducted 3 d after the last cold exposure. Chronic stress
and drug treatments were staggered such that rats
from all groups were tested together. Each drug group
was compared to its respective vehicle control.
Because the high drug concentrations required for
chronic delivery by minipump exceeded their solubility in saline, 10 % ethanol was used as vehicle.
A pilot comparison showed no significant difference
in performance on the AST of rats treated chronically
by minipump with saline vs. 10 % ethanol (F1,5=0.82,
p=0.41).
A separate set of 56 rats were used to test acute
drug effects. Two independent cohorts (n=32 and
n=24, respectively) were assigned to two treatment
groups, control or CIC stress, then subdivided into
two acute drug treatments, either Cit and its vehicle
control (n=8 per group), or Des and its vehicle control
(n=6 per group). Because the cognitive deficits after
2- and 5-wk CIC stress were comparable, acute drug
treatments were tested only in rats exposed to 2-wk
CIC. After stress or control treatments, rats were
habituated and trained as above. On the test day, they
were taken through the SD and CD test stages, then
received an i.p. injection of either Cit (5.0 mg/kg free
base) or saline vehicle (1 ml/kg). Our previous study
showed that this dose of Cit acutely improved reversal
learning performance of CIC-stressed rats (LapizBluhm et al. 2009). Testing resumed on the reversal
stage 30 min after injection. Other rats, treated and
1001
tested in the same manner, received either Des
(5.0 mg/kg) or saline vehicle (1 ml/kg i.p.), 30 min
before reversal testing. This acute dose of Des, slightly
lower than doses we have used previously, was
selected based on pilot results indicating that doses of
o7.5 mg/kg hindered performance of CIC-stressed
rats on the AST. Control rats given 7.5 mg/kg Des
were mildly sedated and took longer to perform, but
generally completed all stages of the test, while a
number of CIC-stressed rats did not, sometimes
refusing to dig for >4 h.
Data analysis
Investigators were blind to the treatment conditions
of rats being tested. Mean trials to criterion on the
SD task on the training day were first compared
by ANOVA, to ensure that acquisition and general
performance capability were equivalent between
groups.
For expt 1, data collected at each time-point
following termination of CIC stress were analysed
by two-way ANOVA (stressrtask), with repeated
measures over task. For expt 2, data were analysed
by two-way ANOVA (stressrtask), with repeated
measures. Data from the EPM were analysed by oneway ANOVA. In expt 3, effects of Cit and Des treatment were analysed separately by three-way ANOVA
(stressrdrugrtask), comparing each drug group to
their respective vehicle controls. For acute injections,
data collected for SD and CD tasks prior to the injections were analysed by two-way ANOVA (stressr
task). Performance on the reversal task (R1) following
acute injections was then analysed using two-way
ANOVA (stressrdrug). For all analyses, where significant main effects or interactions were indicated,
post-hoc comparisons were performed using the
Newman–Keuls test. Significance was determined at
p<0.05.
In all experiments, rats were weighed before
beginning CIC stress or the corresponding control
period, and again on the day of testing. Mean body
weight gain was analysed by one-way ANOVA in
experiments with stress as the between-group factor,
or by two-way ANOVA with stress and drug as
between-group factors.
Only rats completing all stages of the AST were
included in the final analyses. A rat was eliminated if
it stopped responding, or failed to complete all test
stages within 5.5 h of testing, as continuing to test
beyond that time approached the transition into the
dark phase of the light cycle. This resulted in elimination of three rats from expt 1 ; three rats from expt 2 ;
1002
M. D. S. Lapiz-Bluhm and D. A. Morilak
Table 2. Mean body weight gain at testing (weight at testingxstarting weight)
Experiment
Expt 1
3 d post-CIC
7 d post-CIC
14 d post-CIC
21 d post-CIC
Expt 2
2-wk CIC
5-wk CIC
Expt 3
Chronic Cit
Chronic Des
Acute Cit
Acute Des
Group
Body weight gain
(mean¡S.E.M.), g
Control
CIC
Control
CIC
Control
CIC
Control
CIC
18.67¡5.80
14.33¡3.77
22.36¡6.55
18.33¡9.02
26.25¡2.92
22.80¡2.81
36.63¡2.72
31.50¡1.63
Control
CIC
Control
CIC
18.75¡14.48
11.63¡10.51
91.50¡10.74
88.86¡9.75
Control-Veh
Control-Cit
CIC-Veh
CIC-Cit
Control-Veh
Control-Cit
CIC-Veh
CIC-Cit
Control
CIC
Control
CIC
65.33¡4.27
59.71¡8.56
52.71¡7.58
56.86¡4.57
61.25¡5.45
64.86¡4.48
57.00¡6.58
55.29¡4.48
14.63¡3.43
5.63¡3.86
6.00¡5.21
3.33¡3.47
F value
(* p<0.05)
F1,22=4.70*
F1,21=1.48
F1,16=0.80
F1,16=3.22
F1,10=0.18
F1,9=0.04
F1,26=0.02 (drug)
F1,26=1.74 (stress)
F1,25=0.04 (drug)
F1,25=1.93 (stress)
F1,30=4.33*
F1,22=0.20
CIC, Chronic intermittent cold ; Cit, citalopram ; Des, desipramine ; Veh, vehicle.
* p<0.05 compared to the corresponding control group.
seven rats from the chronic drug treatment study in
expt 3, and three rats from the acute drug treatment
study. Animals so excluded were not included in the
reported number of rats used.
Results
Table 2 shows mean body weight gain in each experiment. In general, CIC stress slowed body weight gain
moderately, but this was significant in only a few
cases : in expt 1, the 3-d post-CIC group ; and in expt 3,
the acute Cit study.
Expt 1 : duration of the cognitive effect induced by
14-d CIC stress exposure
CIC stress had no significant effect on training (F1,22=
0.21, p=0.65), indicating that chronic stress did not
impair acquisition or the ability to perform on the
test in general. This was true for all groups, regardless
of the time at which testing was conducted after the
last cold exposure.
Replicating previous results (Lapiz-Bluhm et al.
2009), 14-d CIC stress significantly impaired reversal
learning 3 d following the last cold exposure (Fig. 1).
ANOVA indicated no overall main effect of stress
(F1,22=2.99, p=0.09), but a significant effect of task
(F1,22=11.31, p<0.001), and a significant stressrtask
interaction (F1,22=2.38, p=0.03). Post-hoc analysis at
day 3 showed that CIC-stressed rats required significantly more trials to reach criterion selectively on the
R1 reversal stage compared to unstressed controls
(p<0.01). However, similar analyses performed on
data collected at days 7, 14 and 21 following the last
cold exposure showed no significant differences in
performance between controls and CIC-stressed rats
on any task (Fig. 1).
Antidepressants and cognitive effects of stress
(a)
(b)
25
25
Day 3
*
15
10
5
0
Day 7
20
Trials to criterion
Trials to criterion
20
15
10
5
SD
CD
R1
ID
R2
ED
0
R3
SD
CD
R1
Task
ID
R2
ED
R3
R2
ED
R3
Task
(c)
(d)
25
25
Day 14
15
10
5
0
Day 21
20
Trials to criterion
20
Trials to criterion
1003
15
10
5
SD
CD
R1
ID
R2
ED
R3
Task
0
SD
CD
R1
ID
Task
Fig. 1. Cognitive effects of 14-d chronic intermittent cold (CIC) stress, tested at 3, 7, 14 and 21 d after termination of cold
stress exposure. CIC induced a reversal learning deficit tested 3 d after the last stress exposure (a). However, this effect
dissipated at 7, 14, and 21 d after cold exposure (b–d). At day 3, CIC-stressed rats required significantly more trials to reach
criterion on the R1 reversal task compared to non-stressed controls (* p<0.01 compared to unstressed control rats on the
same task ; mean¡S.E.M., n=8–12 per group). &, CIC ; %, control. CD, Compound discrimination ; ED, extra-dimensional ;
ID, intra-dimensional ; R1, R2, R3, reversal tasks 1, 2 and 3 ; SD, simple discrimination.
Expt 2 : effects of 2-wk vs. 5-wk CIC stress exposure
There was no difference in training of CIC-stressed
rats and controls (F3,19=1.92, p=0.16). Figure 2 shows
the test-day performance of 2- and 5-wk CIC-stressed
rats compared to unstressed controls. ANOVA revealed significant main effects of stress (F2,20=6.49,
p<0.01) and task (F6,120=18.22, p<0.001), and a
stressrtask interaction (F12,120=2.45, p<0.01). Post-hoc
analyses again showed a significant deficit on the
R1 reversal task, in both 2-wk (p<0.001), and 5-wk
(p<0.01) CIC-stressed rats. Although in this experiment, the trials to criterion on the R1 task were
significantly higher in the 2-wk stress group than in
the 5-wk stress group (p<0.05), the reversal deficit
at 5 wk was comparable to that after 2 wk in expt 1,
and in previous studies (Lapiz-Bluhm et al. 2009).
On the EPM, CIC stress induced no anxiety-like
changes in open-arm exploration (Fig. 3), analysed
as OTR for time (F2,20=1.42, p=0.27), or entries
(F2,20=0.77, p=0.48). There were also no non-specific
locomotor effects (F2,20=1.29, p=0.30 for total distance
travelled, not shown).
Expt 3 : effects of chronic and acute antidepressant
treatment on the CIC stress-induced reversal
learning deficit
Neither stress (F1,53=0.42, p=0.51) nor drug treatment
(F2,53=0.70, p=0.50) affected training. Figure 4 shows
performance on the AST following 5-wk CIC stress
exposure, with concomitant drug treatment during
the final 3 wk. In the Cit experiment, ANOVA revealed significant effects of stress (F1,26=4.42, p=
0.045), drug (F1,26=9.82, p<0.01) and task (F 6,156=
10.43, p<0.001), with significant interactions between
stressrdrug (F1,26=12.84, p<0.01), and stressr
drugrtask (F6,156=3.05, p<0.01). As in the preceding
M. D. S. Lapiz-Bluhm and D. A. Morilak
1004
30
(a)
*+
Control
(b)
0.8
0.8
0.6
0.6
25
OTR (Time)
5 week CIC
Trials to criterion
*
20
15
0.2
Control 2-wk
CIC
5-wk
CIC
0.4
0.2
0
Control 2-wk
CIC
5-wk
CIC
Fig. 3. There were no anxiety-like changes in open-arm
exploratory behaviour on the elevated plus-maze after
chronic intermittent cold (CIC) stress exposure for 2 or 5 wk,
measured either as open to total ratios (OTR) for (a) time
or (b) entries (mean¡S.E.M., n=7–8 per group).
5
0
0.4
0
10
OTR (Entries)
2 week CIC
SD
CD
R1
ID
R2
ED
R3
Task
Fig. 2. Exposure to chronic intermittent cold (CIC) stress for
either 2 or 5 wk induced a selective and significant
impairment in reversal learning compared to unstressed
controls, tested 3 d after the termination of chronic cold stress
exposure (* p<0.01 compared to unstressed control rats
on the same task ; +p<0.05 for 2-wk compared to 5-wk
CIC-stressed rats on the same task ; mean¡S.E.M., n=7–8
per group). For abbreviations see Fig. 1 legend.
experiments, post-hoc analyses again revealed a CIC
stress-induced deficit in reversal learning, with a significant increase in trials to criterion on the R1 task
compared to vehicle-treated unstressed controls (p<
0.001). Chronic Cit treatment significantly reversed
the CIC-induced impairment on the R1 task (p<0.001,
Fig. 4 a).
In the chronic Des experiment, ANOVA revealed
significant main effects of stress (F1,25=8.71, p<0.01)
and task (F6,150=9.21, p<0.01), but no effect of drug
(F1,25=0.03, p=0.86). There was a significant stressr
task interaction (F6,150=5.90, p<0.01). As above, posthoc analysis showed that CIC stress significantly
increased trials to criterion on the R1 reversal task
(p<0.001, Fig. 4 b). In this experiment (but not in any
others), there was a significant effect of CIC stress on
the ED set-shifting task (p<0.05). Neither of the CIC
stress-induced deficits were reversed by chronic Des
treatment (Fig. 4 b).
In the acute drug study, there were no pre-existing
differences in performance on the training day between rats in the drug and vehicle groups (Cit :
F2,29=0.96, p=0.59 ; Des : F2,21=0.81, p=0.11). On the
test day, there were also no differences between either
of the drug groups and their respective vehicle control
groups in performance on the SD and CD tasks before
drug administration (Cit : F1,30=0.005, p=0.99 ; Des :
F1,22=0.24, p=0.62). Analysis of the R1 reversal task
after acute Cit treatment showed significant main
effects of stress (F1,28=8.93, p<0.01) and drug (F1,28=
6.64, p<0.05), but no interaction (F1,28=2.60, p=0.12).
Post-hoc comparisons showed that CIC stress increased trials to criterion on the R1 task, and acute Cit
significantly reduced trials to criterion on the reversal
task in CIC-stressed rats, to a level comparable to
that of unstressed controls (Fig. 5 a), replicating our
previous results (Lapiz-Bluhm et al. 2009). By contrast,
acute Des treatment had no such effect (Fig. 5 b).
ANOVA showed significant effects of stress (F1,20=
9.91, p<0.01), but no effect of drug (F1,20=0.46,
p=0.51), nor a stressrdrug interaction (F1,20=2.27,
p=0.15).
Discussion
Executive dysfunction related to prefrontal cortical
dysregulation, specifically perseveration, or the failure
to alter behaviour in response to environmental feedback, is prominent in depression (Channon, 1996).
Cognitive behavioural approaches (Beck, 1976, 2005)
presume that depression is associated with ruminative
perseverations within a set of negative perceptions
and beliefs about the self, the world and the future.
The intent is to help patients challenge such biases,
establishing a more flexible, realistic and adaptive
cognitive schema based on hypothesis-testing and
use of objective evidence to modify thoughts and
behaviours. An animal model that reflects cognitive
functioning of prefrontal cortex, is sensitive to stress,
and is modulated by monoaminergic neurotransmitters and antidepressants, would be a valuable tool
Antidepressants and cognitive effects of stress
(a)
(a)
15
Control-vehicle
Control-Cit
CIC-vehicle
CIC-Cit
30
*
25
10
+
5
0
SD
CD
R1
ID
R2
ED
R3
Trials to criterion (R1)
Trials to criterion
20
*
20
15
10
0
(b)
20
15
Control-vehicle
Control-Des
CIC-vehicle
CIC-Des
*
†
+
5
Task
Control- Controlvehicle
Cit
CICvehicle
CICCit
(b)
30
*
10
25
5
0
SD
CD
R1
ID
R2
ED
R3
Task
Trials to criterion (R1)
Trials to criterion
1005
*
20
15
10
Fig. 4. Effects of chronic antidepressant drug treatment,
initiated after 2-wk chronic intermittent cold (CIC) stress
and maintained during continuation of CIC stress exposure
for 3 more weeks. (a) Chronic treatment with citalopram
(20 mg/kg.d) reversed the CIC stress-induced cognitive
deficit (* p<0.001 compared to non-stressed control rats
on the same task, + p<0.001 compared to vehicle-treated
CIC-stressed rats on the same task ; mean¡S.E.M., n=7–9
per group). (b) Chronic treatment with desipramine
(7.5 mg/kg.d) had no beneficial effect on the elevated number
of trials to criterion on the R1 task after CIC stress,
comparable to vehicle-treated CIC-stressed rats. In this
experiment (but not in any others), CIC stress also increased
trials to criterion on the extra-dimensional cognitive setshifting task, and desipramine also did not reverse this effect.
* p<0.001 compared to vehicle-treated, non-stressed control
rats on the same task ; # p<0.05 compared to vehicle-treated,
non-stressed control rats on the same task ; mean¡S.E.M.,
n=7–8 per group). For abbreviations see Fig. 1 legend.
Fig. 5. Effects of acute antidepressant drug treatment,
given 30 min prior to the R1 reversal task. (a) Acute treatment
with citalopram (5 mg/kg i.p.) improved performance of
chronic intermittent cold (CIC) stress-stressed rats on the
R1 reversal task (* p<0.01 compared to non-stressed control
rats on the same task ; + p<0.01 compared to vehicle-treated
CIC-stressed rats on the same task ; mean¡S.E.M., n=8 per
group). (b) Acute treatment with desipramine (5 mg/kg i.p.)
had no beneficial effect on the increased number of trials
required to reach criterion on the R1 task after CIC stress,
comparable to vehicle-treated CIC-stressed rats (* p<0.05
compared to vehicle-treated, non-stressed control rats ;
mean¡S.E.M., n=6 per group).
for investigating mechanisms involved in the dysregulation and restoration of cognitive flexibility in
depression.
Results of the present study replicated our previous
finding that CIC stress induces a selective impairment
in reversal learning in rats, without interfering with
acquisition of new contingencies. There were no differences in training, SD, CD or ID set-shift tasks.
Moreover, as reported previously, CIC stress did not
generally affect ED cognitive set-shifting (except in
one experiment, see below), despite the rats having
greater difficulty mastering the reversal task earlier in
the test sequence, further demonstrating the relative
selectivity of the CIC stress effect.
The deficit in reversal learning after 2-wk CIC
treatment was transient. It was evident when tested
3 d after the last cold exposure, but had dissipated at
time-points further past the last cold exposure, i.e.
5
0
Control- Controlvehicle
Des
CICvehicle
CICDes
1006
M. D. S. Lapiz-Bluhm and D. A. Morilak
at 7, 14, or 21 d. This was not of sufficient duration to
investigate the effects of chronic antidepressant drug
treatment initiated after the stress-induced cognitive
deficit had been established, as in the clinical treatment of depression. Thus, in expt 2, the feasibility of
continuing chronic stress treatment for a time that
would be sufficient to allow chronic drug intervention
was investigated. Five-week CIC stress induced a reversal learning deficit comparable to that after 2 wk.
Thus, the 5-wk stress protocol was used in expt 3,
with chronic antidepressant drug treatment beginning
after 2 wk, when the reversal deficit had been established, and maintained for 3 wk thereafter, while
CIC stress continued. This design is more relevant to
the clinical treatment of depression, allowing chronic
drug treatment to be initiated after the pathological
behaviour has been established, but also because
treatment is maintained in the presence of continuing
stress, as depressed patients undergoing antidepressant therapy will undoubtedly continue to experience
many of the same life stressors that existed prior to
the initiation of treatment. In this experiment, chronic
treatment with the SSRI, Cit, but not the selective NE
reuptake blocker, Des, effectively reversed the CIC
stress-induced cognitive deficit.
Reversal learning, wherein a previously positive
discriminative stimulus becomes negative, and a previously negative stimulus becomes positive, has been
studied as a measure of cognitive flexibility in humans
(Fellows & Farah, 2003 ; Hornak et al. 2004 ; Murphy
et al. 2002 ; Rogers et al. 2000 ; Rolls et al. 1994), nonhuman primates (Clarke et al. 2004, 2005, 2007 ; Dias
et al. 1996 ; Lee et al. 2007), and rats (Birrell & Brown,
2000 ; Boulougouris et al. 2007, 2008 ; Idris et al. 2005 ;
McAlonan & Brown, 2003 ; van der Meulen et al.
2007). Reversal learning requires integrity of the OFC
(McAlonan & Brown, 2003). Thus, the present results
suggesting that reversal learning is sensitive to chronic
stress may provide insight into how chronic stress
alters prefrontal cortical function in ways that are
relevant to the cognitive deficits in depression. Human
studies have shown that acute cold exposure degrades cognitive performance, including vigilance,
reaction time, reasoning and short-term memory (see
Coleshaw et al. 1983 ; Mahoney et al. 2007 ; Patil et al.
1995 ; Shurtleff et al. 1993). Cold stress impaired performance on four-choice reaction time and delayed
match-to-sample tasks, and increased measures of
tension, confusion and depression, as well as ‘total
mood disturbance ’ scores (Mahoney et al. 2007).
Limited studies of chronic cold exposure also suggest
adverse effects on both cognition and mood (Makinen,
2007).
Food deprivation and changes in body weight
alone have been reported to influence cognition in rats,
although the evidence is equivocal. Food restriction
for 3 months promoted long-term recovery of learning
and memory following global ischaemia (Roberge
et al. 2008a, b). However, 1-yr food restriction had
negative effects (Yanai et al. 2004), and 40 % caloric
restriction for 6 months had no effects on cognition
(Martin et al. 2007). In the present study, the relatively
mild CIC stress treatment only moderately slowed
body weight gain. CIC-stressed rats and controls had
comparable performance during training, and on all
cognitive tasks except the reversal task, and there were
no differences in the effect of CIC stress on weight gain
that could explain the beneficial effects of SSRI treatment. Thus, it is unlikely that differences in body
weight alone could account for either the detrimental
effects of CIC stress on reversal learning, or the
beneficial effects of SSRI treatment.
We and others have also shown that reversal learning is specifically modulated by 5-HT neurotransmission in OFC (Clarke et al. 2004, 2005, 2007 ; LapizBluhm et al. 2009). In our previous study, the CIC stressinduced impairment in reversal learning was mimicked by 5-HT depletion, and was attenuated by acute
administration of Cit, as in the present study. Together
with a decrease in extracellular 5-HT levels measured
by microdialysis in OFC during testing, these findings
indicated a possible dysregulation of serotonergic
modulatory function in the OFC of rats exposed to CIC
stress (Lapiz-Bluhm et al. 2009). The present study
provides further evidence that 5-HT facilitates reversal
learning, in that chronic Cit treatment reversed the
detrimental effect of CIC stress, even though stress
exposure continued during drug treatment.
By contrast with Cit, the selective NE reuptake
blocker, Des, had no effect on the CIC stress-induced
reversal learning impairment, acutely or chronically.
Anecdotally, as noted in the Methods section, CICstressed rats seemed more sensitive to the noncognitive, locomotor-inhibitory effects of acute Des. At
doses that did not hinder performance of controls
in completing the AST, CIC-stressed rats remained
inactive, although seemingly alert, for several hours
post-injection. Further experimentation would be
required to investigate mechanisms of any possible
increase in sensitivity to certain effects of Des, but
these observations may be consistent with previous
observations indicating that CIC stress can sensitize
the response of subcortical noradrenergic receptors to
NE (Ma & Morilak, 2005).
In the present experiments, only the first reversal
task (R1) was consistently affected by CIC stress,
Antidepressants and cognitive effects of stress
whereas the second (R2) and third (R3) reversals were
generally unaffected. Similarly inconsistent effects on
the later reversal tasks have been reported in other
contexts (e.g. Bondi et al. 2008 ; Boulougouris et al.
2008 ; Hatcher et al. 2005 ; Tait & Brown, 2008). We
have speculated that R1 may be more vulnerable to
manipulation because of differences in the difficulty
involved in reversing the contingencies established
in the stages immediately preceding each of the reversals. R2 is similar to R1, as it follows a new acquisition within the same stimulus dimension, but it
differs in that the animal has just had prior experience
performing a reversal in R1. The ease with which rats
typically master R2 may reflect a ‘learning-to-learn ’
phenomenon, that may make R2 more resistant to
disruption. By contrast, R3 is preceded by the ED setshift, in which rats must abandon a cognitive set
that had been reinforced repeatedly in all stages up to
that point. Thus, the new contingency acquired in the
ED task may not be as ‘strong ’ as those established
in earlier acquisitions, and less ‘flexibility ’ may be
required to achieve subsequent reversal in R3, making
R3 less prone to disruption.
The selective reversal deficit after the chronic
metabolic stressor in the present study was also different from cognitive effects reported previously after
a more psychogenic stress treatment, CUS, which
consistently induced ED set-shifting deficits, with less
consistent effects on reversal learning (Bondi et al.
2008). CIC stress affected ED set-shifting in only one
experiment in the present study. Moreover, chronic
Des prevented the deficit in ED set-shifting induced
by CUS (Bondi et al. 2008), but did not alter the effect
of CIC stress on ED set-shifting in the present study,
consistent with its lack of effect on the reversal deficit
induced by CIC stress. Thus, it would seem that the
prefrontal substrates affected by CIC stress are subtly
different than those affected by CUS, and different
types of stress may impact different forms of cognitive
flexibility, dependent on different subregions of prefrontal cortex. CUS might produce effects that reflect
functional changes in mPFC (see Bondi et al. 2008),
while reversal learning, affected selectively by CIC
stress, may reflect changes more specifically in functioning of OFC.
Further, CIC stress may be even more broadly useful for modelling other dimensions of depression as
well. We have shown that CIC stress induced a dysregulation of the hypothalamic–pituitary–adrenal
axis, as well as changes in NE release and postsynaptic adrenergic receptor sensitivity in stressresponsive subcortical brain regions, including the
paraventricular nucleus of the hypothalamus (Ma &
1007
Morilak, 2005 ; Pardon et al. 2003). However, in the
present experiment there was no anxiety-like reduction in open-arm exploration on the EPM after
CIC stress, as was seen previously after CUS (Bondi
et al. 2008). Moreover, some of the effects of CIC stress
in previous studies were more pronounced in WKY
rats, a genetically stress-vulnerable strain, compared
to the more resilient Sprague–Dawley rats used in
the present experiments. Thus, CIC as a model of
metabolic stress might be useful specifically in addressing mechanisms underlying vulnerability to
stress-induced behavioural and physiological pathology. Further research is required to understand how
the behavioural, cognitive, physiological and affective
dimensions affected by CIC stress might be related to
each other, to the mechanisms underlying long-term
consequences of chronic stress, or to differences in the
efficacy of therapeutic interventions.
Acknowledgements
We thank Mr Gus Rodriguez, Ms. Julianne Jett and
Ms. Ashley Furr for their expert technical assistance.
Support was provided by a Young Investigator Award
from NARSAD : The Mental Health Association (to
M.D.S.L.-B.), and by research grants from the National
Institute of Mental Health (MH083669 and MH072672)
to D.A.M.
Statement of Interest
None.
References
Anisman H, Zacharko RM (1986). Behavioral and
neurochemical consequences associated with stressors.
Annals of the New York Academy of Sciences 467, 205–225.
Beats BC, Sahakian BJ, Levy R (1996). Cognitive
performance in tests sensitive to frontal lobe dysfunction
in the elderly depressed. Psychological Medicine 26,
591–604.
Beck AT (1976). Cognitive Therapy and the Emotional Disorders.
New York : International Universities Press.
Beck AT (2005). The current state of cognitive therapy :
a 40-year retrospective. Archives of General Psychiatry 62,
953–959.
Beck AT, Brown G, Steer RA, Eidelson JI, et al. (1987).
Differentiating anxiety and depression : a test of the
cognitive content-specificity hypothesis. Journal of
Abnormal Psychology 96, 179–183.
Birrell JM, Brown VJ (2000). Medial frontal cortex mediates
perceptual attentional set shifting in the rat. Journal of
Neuroscience 20, 4320–4324.
1008
M. D. S. Lapiz-Bluhm and D. A. Morilak
Bondi CO, Rodriguez G, Gould GG, Frazer A, et al. (2008).
Chronic unpredictable stress induces a cognitive deficit
and anxiety-like behavior in rats that is prevented by
chronic antidepressant drug treatment.
Neuropsychopharmacology 33, 320–331.
Boulougouris V, Dalley JW, Robbins TW (2007). Effects
of orbitofrontal, infralimbic and prelimbic cortical lesions
on serial spatial reversal learning in the rat. Behavioural
Brain Research 179, 219–228.
Boulougouris V, Glennon JC, Robbins TW (2008).
Dissociable effects of selective 5-HT2A and 5-HT2C receptor
antagonists on serial spatial reversal learning in rats.
Neuropsychopharmacology 33, 2007–2019.
Channon S (1996). Executive dysfunction in depression : the
Wisconsin Card Sorting test. Journal of Affective Disorders
39, 107–114.
Clarke HF, Dalley JW, Crofts HS, Robbins TW, et al. (2004).
Cognitive inflexibility after prefrontal serotonin depletion.
Science 304, 878–880.
Clarke HF, Walker SC, Crofts HS, Dalley JW, et al. (2005).
Prefrontal serotonin depletion affects reversal learning but
not attentional set shifting. Journal of Neuroscience 25,
532–538.
Clarke HF, Walker SC, Dalley JW, Robbins TW, et al.
(2007). Cognitive inflexibility after prefrontal serotonin
depletion is behaviorally and neurochemically specific.
Cerebral Cortex 17, 18–27.
Coleshaw SRK, Van Someren RNM, Wolff AH, Davis HM,
et al. (1983). Impaired memory registration and speed
of reasoning caused by low body temperature. Journal
of Applied Physiology 55, 27–31.
Dias R, Robbins TW, Roberts AC (1996). Primate analogue
of the Wisconsin Card Sorting Test : effects of excitotoxic
lesions of the prefrontal cortex of the marmoset.
Behavioral Neuroscience 110, 872–886.
Fellows LK, Farah MJ (2003). Ventromedial frontal cortex
mediates affective shifting in humans : evidence
from reversal learning paradigm. Brain 126, 1830–1837.
Fossati P, Amar G, Raoux N, Ergis AM, et al. (1999).
Executive functioning and verbal memory in young
patients with unipolar depression and schizophrenia.
Psychiatry Research 89, 171–187.
Goldapple K, Segal Z, Garson C, Lau M, et al. (2004).
Modulation of cortico-limbic pathways in major
depression : treatment-specific effects of cognitive
behavior therapy. Archives of General Psychiatry 61, 34–41.
Hatcher PD, Brown VJ, Tait DS, Bate S, et al. (2005). 5-HT6
receptor antagonists improve performance in an
attentional set shifting task in rats. Psychopharmacology
181, 253–259.
Hornak J, O’Doherty J, Bramham J, Rolls ET, et al. (2004).
Reward-related reversal learning after surgical
excisions in orbito-frontal or dorsolateral prefrontal
cortex in humans. Journal of Cognitive Neuroscience 16,
463–478.
Idris NF, Repeto P, Neill JC, Large CH (2005). Investigation
of the effects of lamotrigine and clozapine in improving
reversal-learning impairments induced by acute
phencyclidine and D-amphetamine in the rat.
Psychopharmacology (Berlin) 179, 336–348.
Kennedy SH, Evans KR, Kruger S, Mayberg HS, et al.
(2001). Changes in regional brain glucose metabolism
measured with positron emission tomography after
paroxetine treatment of major depression. American Journal
of Psychiatry 158, 899–905.
Lapiz MDS, Bondi CO, Morilak DA (2007). Chronic
treatment with desipramine improves cognitive
performance of rats in an attentional set shifting test.
Neuropsychopharmacology 32, 1000–1010.
Lapiz MDS, Morilak DA (2006). Noradrenergic modulation
of cognitive function in rat medial prefrontal cortex as
measured by attentional set shifting capability.
Neuroscience 137, 1039–1049.
Lapiz-Bluhm MDS, Jett JD, Rodriguez G, Morilak DA
(2008). Temporal characterization of the cognitive deficit
induced by chronic intermittent cold stress on an
attentional set shifting test in rats. Society for Neuroscience
Abstracts 34, Online, Program number 196.6.
Lapiz-Bluhm MDS, Soto-Piña AE, Hensler JG, Morilak DA
(2009). Chronic intermittent cold stress and serotonin
depletion induce deficits of reversal learning in an
attentional set-shifting test in rats. Psychopharmacology 202,
329–341.
Lee B, Groman S, London ED, Jentsch JD (2007). Dopamine
D2/D3 receptors play a specific role in the reversal of
a learned visual discrimination in monkeys.
Neuropsychopharmacology 32, 2125–2134.
Ma S, Morilak DA (2005). Chronic intermittent cold stress
sensitizes the HPA response to a novel acute stress by
enhancing noradrenergic influence in the rat
paraventricular nucleus. Journal of Neuroendocrinology 17,
761–769.
Mahoney CR, Castellani J, Kramer FM, Young A, et al.
(2007). Tyrosine supplementation mitigates working
memory decrements during cold exposure. Physiology and
Behavior 92, 575–582.
Makinen TM (2007). Human cold exposure, adaptation,
and performance in high latitude environments. American
Journal of Human Biology 19, 155–164.
Martin B, Pearson M, Kebejian L, Golden E, et al. (2007).
Sex-dependent metabolic, neuroendocrine, and cognitive
responses to dietary energy restriction and excess.
Endocrinology 148, 4318–4333.
McAlonan K, Brown VJ (2003). Orbital prefrontal cortex
mediates reversal learning and not attentional set shifting
in the rat. Behavioural Brain Research 146, 97–103.
Murphy FC, Michael A, Robbins TW, Sahakian BJ (2003).
Neuropsychological impairments in patients with major
depressive disorder : the effects of feedback on task
performance. Psychological Medicine 33, 455–467.
Murphy FC, Sahakian BJ, Rubinsztein JS, Michael A, et al.
(1999). Emotional bias and inhibitory control processes in
mania and depression. Psychological Medicine 29,
1307–1321.
Murphy FC, Smith KA, Cowen PJ, Robbins TW, et al.
(2002). The effects of tryptophan depletion on cognitive
Antidepressants and cognitive effects of stress
and affective processing in healthy volunteers.
Psychopharmacology (Berlin) 163, 42–53.
Pardon M-C, Ma S, Morilak DA (2003). Chronic cold stress
sensitizes brain noradrenergic reactivity and
noradrenergic facilitation of the HPA stress response in
Wistar Kyoto rats. Brain Research 971, 55–65.
Patil PG, Apfelbaum JL, Zacny JP (1995). Effects of
a cold-water stressor on psychomotor and cognitive
functioning in humans. Physiology and Behavior 58,
1281–1286.
Prasko J, Hornacek J, Zalesky R, Kopecek M, et al. (2004).
The change of regional brain metabolism (18FDG PET) in
panic disorder during the treatment with cognitive
behavioral therapy or antidepressants. Neuroendocrinology
Letters 25, 340–348.
Roberge MC, Hotte-Bernard J, Messier C, Plamondon H
(2008a). Food restriction attenuates ischemia-induced
spatial learning and memory deficits despite extensive
CA1 ischemic injury. Behavioural Brain Research 187,
123–132.
Roberge MC, Messier C, Staines WA, Plamondon H (2008b).
Food restriction induces long-lasting recovery of spatial
memory deficits following global ischemia in delayed
matching and non-matching-to-sample radial arm maze
tasks. Neuroscience 156, 11–29.
Roberts AC, Robbins TW, Everitt BJ, Muir JL (1992). A
specific form of cognitive rigidity following excitotoxic
lesions of the basal forebrain in marmosets. Neuroscience 47,
251–264.
Rogers MA, Kasai K, Koji M, Fukuda R, et al. (2004).
Executive and prefrontal dysfunction in unipolar
1009
depression : a review of neuropsychological and imaging
evidence. Neuroscience Research 50, 1–11.
Rogers RD, Andrews TC, Grasby PM, Brooks DJ, et al.
(2000). Contrasting cortical and subcortical activations
produced by attentional set shifting and reversal
learning in humans. Journal of Cognitive Neuroscience 12,
142–162.
Rolls ET, Hornak J, Wade D, McGrath J (1994).
Emotion-related learning in patients with social and
emotional changes associated with frontal lobe damage.
Journal of Neurology, Neurosurgery and Psychiatry 57,
1518–1524.
Sheline YI (2003). Neuroimaging studies of mood
disorder effects on the brain. Biological Psychiatry 54,
338–352.
Shurtleff D, Thomas JR, Ahlers ST, Schrot J (1993). Tyrosine
ameliorates a cold-induced delayed matching-to-sample
performance decrement in rats. Psychopharmacology (Berlin)
112, 228–232.
Tait DS, Brown VJ (2008). Lesions of the basal
forebrain impair reversal learning but not shifting of
attentional set in rats. Behavioural Brain Research 187,
100–108.
van der Meulen JAJ, Joosten RNJMA, de Bruin JPC,
Feenstra MGP (2007). Dopamine and noradrenaline efflux
in the medial prefrontal cortex during serial reversals and
extinction of instrumental goal-directed behaviour.
Cerebral Cortex 17, 1444–1453.
Yanai S, Okaichi Y, Okaichi H (2004). Long-term dietary
restriction causes negative effects on cognitive functions in
rats. Neurobiology of Aging 25, 325–332.