Available online at www.sciencedirect.

com

Behavioural Brain Research 189 (2008) 170–179

Research report

A sensitizing regimen of amphetamine that disrupts attentional

set-shifting does not disrupt working or long-term memory
Robert E. Featherstone a,∗ , Zoe Rizos a , Shitij Kapur b,c , Paul J. Fletcher a,b,d
a Section of Biopsychology, Centre for Addiction and Mental Health, 250 College Street, Toronto, Ontario, Canada M5T 1R8
b Department of Psychiatry, University of Toronto, Canada
c Schizophrenia/PET Centre, Centre for Addiction and Mental Health, Canada
d Department of Psychology, University of Toronto, Canada

Received 18 September 2007; received in revised form 11 December 2007; accepted 27 December 2007
Available online 20 January 2008

Abstract
Exposure to an intermittent, escalating dose of amphetamine induces a sensitized state that, both behaviourally and neurochemically, mirrors
several features linked to the positive symptoms of schizophrenia. Increasingly it is being realized that cognitive deficits are a core component
of schizophrenia; therefore we sought to assess the effects of inducing an amphetamine-sensitized state on memory (working and long-term) and
cognitive flexibility, two cognitive domains impaired in schizophrenia. Rats were exposed to a sensitizing regimen of amphetamine (1–5 mg/kg;
three times per week for 5 weeks; escalating at 1 mg/kg per week) or saline. In experiment 1, animals were tested on an operant delayed non-match
to position task (working memory). Experiment 2 used a standard fixed-platform location water maze task (long-term memory), while experiment 3
used a variable-platform location water maze task (long-term memory and working memory). Amphetamine-sensitized animals were not impaired
on any of these tasks. In experiment 4, animals were assessed on a strategy selection task in which they were first required to learn to locate
a food reward using a particular learning strategy (place or response) then to learn to shift to an alternate learning strategy (response or place).
Amphetamine-sensitized animals were not impaired on this task. In the final experiment animals were found to be impaired in performance of the
extra-dimensional shift component of an attentional set-shifting task. These results suggest that while amphetamine sensitization does not produce
memory impairments similar to those seen in schizophrenia, it does produce strong impairments in set-shifting, suggesting changes in prefrontal
function similar to those seen in schizophrenia.
© 2008 Elsevier B.V. All rights reserved.

Keywords: Animal model; Schizophrenia; Dopamine; Cognition; Amphetamine sensitization; Morris water maze; Delayed non-matching to position; Attentional
set-shifting

1. Introduction rupted in schizophrenia, such as pre-pulse inhibition [44,51–53]

Amphetamine sensitization has been suggested to model
some aspects of schizophrenia [30,31,40,42,51–53]. Briefly,
amphetamine sensitization refers to a process whereby responses
to amphetamine increase through repeated exposure to the drug
[40]. This increased responsiveness is maintained during with-
drawal from the drug, and has been shown to last up to 1-year post
drug exposure in rats [35]. Following the induction of sensitiza-
tion, animals show impaired performance on a number of tasks
thought to involve psychological processes similar to those dis-
∗ Corresponding author. Tel.: +1 416 535 8501x6242; fax: +1 416 979 6942.
E-mail address: Rob Featherstone@camh.net (R.E. Featherstone).
although see [31,43], latent inhibition [30,31,43,53], attentional
vigilance [8], sustained attention [18] and attentional set-shifting
[13,16]. In the latter task, amphetamine-sensitized animals
displayed an impairment in making an extra-dimensional atten-
tional shift, as well as difficulties in reversal learning, similar to
what has been found in animals with prefrontal lesions [1,28],
schizophrenic patients [34] and human amphetamine abusers
[32].
These findings suggest that amphetamine sensitization repro-
duces many of the attentional and pre-attentional deficits asso-
ciated with schizophrenia. What is less well known, however, is
the extent to which amphetamine sensitization can model other
areas of cognitive impairment in schizophrenia. Schizophrenia is
associated with impairments in both working [20] and long-term

0166-4328/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbr.2007.12.032
 R.E. Featherstone et al. / Behavioural Brain Research 189 (2008) 170–179
memory [7,24]. In non-human primates, exposure to a sensi-
tizing regimen of amphetamine produced short-term deficits in
performance of a delayed response task, a measure of working
memory. In contrast, a brief exposure regimen of amphetamine
(2.5 mg/kg for 5 days) did not disrupt performance of a radial
maze based delayed alternation task in rats [50]. Regarding long-
term memory, one study found that amphetamine sensitization
(1–5 mg/kg, three injections per day for 6 days) failed to disrupt
acquisition or retention of a fixed platform position water maze
task in rats [41]. In contrast, animals exposed to a repeated sin-
gle dose of amphetamine (3 mg/kg, five injections over 10 days)
showed deficits in an object recognition memory task which
required the retention of information about objects encountered
2 or 4 h earlier [2], suggesting that some forms of long-term
memory may be impaired in sensitized animals. Thus, there
are conflicting results from animal studies on the impact of the
amphetamine-sensitized state on memory and it is possible that
this could relate to differences between the various procedures
used to induce the sensitized state.
In our work an intermittent schedule of amphetamine, where
the drug is given three times per week over 5 weeks, with the
dose escalating at a rate of 1 mg/kg per week, induces a variety of
behavioural deficits when animals are tested several weeks after
drug treatment [18,51–53]. These deficits include poor atten-
tional set-shifting and sustained visual attention. The present
studies extended these findings to examine the impact of this
regimen of amphetamine on working and long-term memory.
Working memory was assessed on an operant delayed non-
match to position (DNMTP) task [12]. Three experiments were
conducted that assessed long-term memory. The first of these
examined the effect of amphetamine sensitization on long-term
memory in a standard fixed platform position water maze task.
The second experiment used a more demanding variable plat-
form position water maze task, in which animals were required
to locate a hidden platform in a novel position each day. In the
final experiment, animals were tested on a strategy shifting task
which required learning to locate a food reward on a plus maze
using a particular type of memory (place or response) followed
by a shift to a different type of memory (response or place). Inter-
ference with the prefrontal cortex produces a selective deficit in
the ability to switch memory strategies [37] and performance
of this and similar tasks is dependent upon prefrontal dopamine
function [19,37–39]. This task was included in order to assess the
impact of amphetamine sensitization on prefrontal control over
learning and memory processes. Finally, experiment 5 assessed
the effects of amphetamine sensitization on an attentional set-
shifting task [1,28] since we have previously found impaired
attentional set-shifting in the amphetamine-induced sensitized
state [16].
2. Methods
2.1. Subjects
Male, Spraque–Dawley (Charles Rivers, Saint-Constant, Quebec) rats
weighing between 275 and 300 g at the beginning of each experiment were
used. Animals were housed in a room maintained at 22 (±2) ◦ C and were kept
on a 12:12 light/dark cycle (lights on at 08:00). Eighty-three animals were used
171
(experiment 1: saline n = 12 and amphetamine n = 12; experiment 2: saline n = 10
and amphetamine n = 10; experiment 3: saline n = 10 and amphetamine n = 10.
A single group of animals (saline n = 9 and amphetamine n = 10) were run on the
strategy-shifting (experiment 4) and attentional set-shifting tasks (experiment 5).
2.2. Amphetamine sensitization
Amphetamine injections (IP) took place three times a week (Monday,
Wednesday and Friday) for 5 weeks. During week 1, amphetamine treated ani-
mals received a dose of 1 mg/kg (from salt), with the dose being escalated by
1 mg/kg each subsequent week (dose during the final week was 5 mg/kg). Control
animals received saline. All injections were administered in a 1 ml/kg volume.
2.3. Experiment 1: delayed non-match to position
DNMTP training took place in standard operant conditioning chambers (Med
Associates, St. Albans, Vermont). Each chamber was equipped with a pellet dis-
penser, two retractable response levers, a pellet magazine, two stimulus lights
and an infrared photobeam located in the pellet magazine on the front wall. The
pellet magazine was located in the lower centre of the chamber and the response
levers were located to the left and right of the magazine. One stimulus light
was located above each response lever. Additionally, two opaque barriers (5 cm
deep × 0.5 cm wide), running the height of the chamber were positioned per-
pendicular to the front wall, separating each response lever from the magazine.
Animals were first trained to respond on a FR1 schedule of reinforcement,
with both response levers available during these sessions. Each session lasted for
30 min or until 100 responses on either lever had been made. This training lasted
for four sessions. Animals were then trained to respond to signalled presentations
of the response lever. At the start of each trial, a stimulus light was illuminated and
the response lever immediately below that stimulus light was extended into the
chamber. Animals were required to press the lever within 20 s in order to receive
reinforcement. Once a response had occurred, the stimulus light was turned off
and the response lever was retracted. If no response occurred, the stimulus light
was turned off and the lever was retracted, followed by a 20 s time-out period.
Each session consisted of 50 presentations of each lever, for 100 trials, and train-
ing on this phase lasted for three to five sessions. Animals were then trained on the
DNMTP task. Each DNMTP trial consisted of a sample phase and a choice phase.
The sample phase was initiated by the extension of a response lever into the
chamber, along with illumination of the corresponding stimulus light. A response
on this lever caused the stimulus light to extinguish, the lever to retract, and the
delay period to begin. During the delay period animals were required to nose-
poke in the food magazine. The first nose-poke response after the delay period
had timed out triggered the start of the choice phase. Here, both response levers
were extended into the chamber, and both of the stimulus lights were illuminated.
To receive reinforcement, animals were required to respond on the lever that was
not present during the sample phase. All DNMTP sessions consisted of 100 trials.
Accuracy, defined as the percentage of trials with a correct response, was the pri-
mary measure of performance and was calculated separately for each delay value.
Both the barriers and the nose-poke requirement during the delay periods were
used to minimize the use mediating strategies that rats could use to solve the task.
Animals were initially trained only at the shortest delay (100 trials at 1 s
delay). Once performance had reached a level of 85% accuracy, the next delay
was introduced (i.e. 1 s and 4 s delay, 50 trials each). Further delays were gradu-
ally phased in until rats were being tested with delays of 1, 4, 8, 16 and 24 s, with
20 trials at each delay period per session. Once behaviour had stabilized, animals
were given a further 10 sessions of training prior to the start of drug treatment
(Baseline phase). Based on average performance on this baseline phase rats were
matched to two groups to receive amphetamine or saline injections. During the 5
weeks amphetamine sensitization period, animals were given DNMTP sessions
on non-injection days (Tuesday and Thursday) (sensitization phase). After the
sensitization phase had finished testing was conducted three times per week
(Monday, Wednesday and Friday) for 5 weeks (withdrawal phase).
2.4. Experiment 2: fixed platform location water maze
The water maze was constructed from a circular, white plastic tub, 1.5 m in
diameter and 79 cm deep. The platform was made from clear Plexiglas (15 cm
172 R.E. Featherstone et al. / Behavioural Brain Research 189 (2008) 170–179
in diameter and 50 cm tall). Water was maintained at 20–22 ◦ C and a small
amount of tempura paint (approximately 100 ml) was added to render the water
opaque. The platform was submerged 2 cm below the surface of the water. A
video camera was mounted directly above the water maze, which allowed each
session to be recorded for later analysis.
Training on the fixed platform water maze task took place 30 days following
the last amphetamine injection and consisted of three phases, acquisition, storage
and retention. Acquisition: animals were given six daily sessions of training,
with each session consisting of four trials; each trial began from one of four
equidistant start positions and lasted for 60 s, or until the rat successfully located
the platform. Animals were allowed to remain on the platform for 10 s. In the
event that an animal did not find the platform during the 60-s trial, it was gently
picked up and placed onto the platform for a 10-s period. The latency to locate
the submerged platform served as the primary measure of behaviour. Storage:
after the sixth session of training, animals were given two probe trials. The
first probe trial occurred immediately following the end of training. The second
probe trial was given on the next day, 24 h after the first probe trial. During each
probe trial, the platform was removed and the animal was started from a novel
start position. Each trial lasted for 60 s. The number of times that the animal
crossed the position where the platform had been during training was recorded.
An identically sized area located in the opposite quadrant was used as a control
location, and the number of crossings of this location was also recorded. Long-
term retention: in order to assess long-term retention of the platform position,
subjects were given a final probe trial 10 days following the end of the second
phase of training, followed by two sessions of re-acquisition training.
2.5. Experiment 3: variable platform location water maze
The procedure for the variable platform location water maze was similar to
that of the acquisition phase of the fixed location version, with the exceptions that
(1) the platform was located in a different position during each session, and (2)
animals received eight trials per session (two from each of the four equidistant
start positions), instead of four. Animals were run sequentially, such that an
intertrial interval of approximately 20 min separated each trial for each animal.
At the end of acquisition training, the task was modified to allow for assessment
of short-term memory [49]. Animals were given a single initial trial on each of 3
days using a novel platform location. After a fixed delay period (16 s, 1 h or 24 h),
animals were given a second trial, with the platform located in the same position
as on the first trial. The experiment was counterbalanced such that, for any given
day, roughly equal numbers of animals were tested at each delay period.
2.6. Experiment 4: strategy-shifting
Training took place on an eight arm radial maze constructed from dark grey
Plexiglas. Each individual arm (60 cm long and 9 cm wide) was attached to a
central octagonal platform, measuring 40 cm across, elevated 60 cm above the
floor. An insert was placed into the centre of the maze that constricted the shape
of the central octagonal platform into a plus shape. Additionally, an insert was
used to block individual arms when needed. Four metal cups, in which the food
reward (a half piece of Froot-Loop©) would be located, were placed at the far
end of each arm.
The procedures for this task followed those of [37]. Training began with a
series of habituation sessions, designed to accustom the animals to eating on the
maze and to being repeatedly handled by the experimenter. During the first day
of training, five half pieces of Froot-Loop©were placed within each arm of the
maze (two half-pieces in the food cup and three half-pieces on the arm), and the
animals were given 15 min of free access to the maze. Arms were rebaited in
the event that an animal ate all of the Froot-Loops©. From days 2 to 6, animals
underwent a similar procedure, with the exception that only one half-piece of
Froot-Loop©was placed in each food cup, and that animals were picked up and
placed back into the centre of the maze after consuming each Froot-Loop©piece.
Sessions lasted for 15 min or until the animal had depleted the maze four times.
On the final day of habituation, animals were tested for turn bias. One arm of
the maze was blocked and the animal was started from the arm opposite to the
blocked arm. Thus, an animal was forced to make either a left or right hand turn
when it reached the end of the arm. Both arms were baited with food, and animals
were given enough trials so that they were able to successfully find food in both
of the arms (although only the response made during the first trial was considered
indicative of a turn bias). Seven series of trials were conducted with each animal,
with each subsequent series starting from the start point immediately clockwise
from the last. An animal was considered to have a turn bias for whichever turn
it made most frequently during this phase of training.
Once this phase had been completed, training began on the strategy-shift
task. Training required 2 days, with acquisition training taking place on the first
day and training on the shift occurring on the following day. During acquisition
training, approximately one-half of the animals were trained to use a response
strategy (e.g., turn left) to locate food reward, while the other half were trained
to use a place strategy (e.g., go to the north position). On the second day, animals
had to learn to locate food reward using the alternate strategy to that used on
the first day. Animals were trained until they performed the correct response 10
times consecutively, at which point they were given a probe trial from a novel
start position. A probe trial was considered correct if the animal chose the arm
consistent with the strategy it had been trained to use. In the event of a correct
probe trial, training was considered to be complete. If an incorrect choice was
made, animals were retrained on the task until they had reached an additional
criterion of five consecutive correct responses, at which time an additional probe
trial was given. This process continued until a correct choice was made during
a probe trial. The procedure for training on the strategy-shift was identical to
that used during acquisition. Animals were considered to have mastered the shift
when they were able to complete 10 consecutive correct trials, followed by a
correct probe trial. Probe trials were conducted in a manner identical to that used
during acquisition.
2.7. Experiment 5: attentional set-shifting task
Animals were given 4 weeks rest following radial maze training before begin-
ning training on the attentional set-shifting task. Training occurred in a chamber
constructed from black Plexiglas (60 cm long × 42 cm wide × 30 cm tall). The
chamber was divided into a front and rear compartment, with the front compart-
ment divided into two separate compartments (21.5 cm long × 20.5 cm wide). An
opaque barrier separated the front and rear chambers, and this barrier was low-
ered at the start of each trial such that the animal, placed in the rear compartment,
was blocked from accessing the front chamber. Each of the two compartments in
the front chamber contained a small ceramic bowl (7 cm in diameter and 4 cm in
depth) that was covered with a texture (fine sandpaper, course sandpaper, velvet,
the reversed side of the velvet material, paper or plastic wrap), and was filled
with scented bedding material (Bed-O’cobs, The Andersons Inc., Maumee, OH,
scented with paprika, thyme, cinnamon, cumin, cloves or sage).
The procedure for set-shifting has been previously described [14,16].
Animals were first shaped to dig for the food reward (half piece of Froot-
Loop©cereal). The cereal was initially placed directly on top of the bedding and,
over time, as animals learned to retrieve this food, was placed deeper within the
bedding, eventually requiring the animal to learn to dig to obtain the food. This
training took place for as many trials as were needed for animals to reliably (six
trials in a row) dig and retrieve food from the bowl. Animals were then trained on
a simple discrimination, in which food was paired with one stimulus dimension
(either odor or texture) but not the other. The stimuli during this discrimination
were banana and vanilla or masking tape and paper towel. Training continued
until the animals chose correctly over six consecutive trials.
One the next day, training began on the full set of discriminations. Two
bowls were placed in the front chamber of the apparatus, with one bowl
baited with a half piece of Froot-Loop©and the other empty. The location
of the baited bowl was randomly determined on each trial to prevent the
development of a positional bias. The bowls differed according to odor and
texture, with food being consistently associated with a particular odor or texture
depending on the current discrimination. Animals were first trained on a simple
discrimination involving an exemplar from a single stimulus dimension (either
odor or texture). During the next stage of training (complex discrimination),
an exemplar from the same stimulus dimension was added, although food
remained paired with the original stimulus (e.g., odor 1+/odor 2−, or texture
1+/texture 2−). The third discrimination (reversal 1) was arranged such that the
previously irrelevant exemplar (e.g., odor 2 or texture 2) was now paired with
food while the previously relevant exemplar was not. The fourth discrimination
(intra-dimensional shift, IDS) required animals to choose between a new series
of exemplars, with the previously relevant stimulus dimension still being paired
 R.E. Featherstone et al. / Behavioural Brain Research 189 (2008) 170–179 173

with food. This discrimination was followed by a reversal (reversal 2) in which

the alternate exemplar of the relevant stimulus dimension was now paired with
food. The sixth discrimination (extra-dimensional shift, EDS) required animals
to shift attention to the previously irrelevant stimulus dimension while the final
discrimination involved a reversal of the extra-dimensional shift discrimination
in which the animal had to choose the previously irrelevant exemplar of
the new stimulus dimension. Animals were determined to have mastered a
discrimination once they had made six correct responses in a row.

2.8. Locomotor activity testing

For each experiment, locomotor responsiveness to amphetamine was

assessed following completion of behavioural training. The monitoring system
consisted of 16 Plexiglas cages, each of which was equipped with 6 photo-
beam cells capable of detecting horizontal movement. The purpose of this stage
of testing was to confirm that amphetamine sensitization had, indeed, occurred.
Animals were given three, 2 h sessions of habituation training, followed by a test
day. On the test day, animals were given 30 min in which to habituate to the boxes,
at which point they received an injection of amphetamine (0.5 mg/kg). Activity
was then assessed for 60 min. For experiments 1–3, locomotor activity testing
took place between days 60 and 80 of amphetamine withdrawal. Animals from
experiments 4 and 5 were given locomotor testing around day 100 of withdrawal.

3. Results

3.1. Experiment 1: DNMTP

Fig. 1 depicts data from the DNMTP task. Rats were assigned
to either the amphetamine or saline groups based on baseline
testing, such that the two groups had equivalent levels of per-
formance. As such, no significant differences were found for
Fig. 1. Graph depicts DNMTP performance in saline treated (white circle) and
group or group × delay during baseline testing. A significant
amphetamine-sensitized (black circle) animals. Graph A depicts average choice
main effect was found for delay, with performance decreasing accuracy during both of the drug free days (Tuesday and Thursday) for each
as the delay increased [F(4, 88) = 139.04, p < 0.05]. For testing week of sensitization (S1–S5). Graph B depicts average choice performance for
that took place during amphetamine sensitization, data from the each week during the 5-week withdrawal period (W1–W5). Although animals
two sessions per week were averaged for each of the 5 weeks. displayed a delay-dependent decrease in accuracy throughout each session, no
group differences were observed during sensitization (A: S1–S5) or withdrawal
Separate repeated measures ANOVAs were conducted on the
(B: W1–W5). Error bars depict S.E.M.
data from each week, using group as the between subject fac-
tor and delay as the repeated within subject factor. A significant
main effect of delay was found for each of the 5 weeks [small-
est F(4, 88) = 56.18, p < 0.05]. Thus, accuracy declined with
increases in the delay period. Neither the main effect of group nor
the group × delay interaction terms were significant, indicating
that there were no differences between saline and amphetamine-
sensitized animals in the performance of this task. Data from the
5 weeks of testing during withdrawal were averaged across each
week (three sessions per week) and separate repeated measures
ANOVAs were conducted for each week. Again a significant
main effect was found for delay during each week (smallest
F(4, 88) = 68.61, p < 0.05), indicating that accuracy deteriorated
as the length of the delay period increased. Neither group nor
the group × delay interaction term was significant, showing that
amphetamine sensitization had no detrimental impact on the
performance of this task.
Fig. 2. Graph depicts acquisition of the fixed platform water maze task in saline
3.2. Experiment 2: fixed platform location water maze treated (saline) and amphetamine-sensitized (AMPH) animals. The figure shows
the average latency (s) over each of the four daily trials that comprised each ses-
sion. Amphetamine-sensitized animals showed significantly longer latencies to
A repeated measure ANOVA was carried out on latency
find the platform on sessions 1, 4 and 6 (p < 0.05). No other significant differ-
to find the platform, using group as a between-subject vari- ences were found. Error bars depict S.E.M. Asterisks denote significant group
able and session as a within-subject variable. Fig. 2 depicts differences.
174 R.E. Featherstone et al. / Behavioural Brain Research 189 (2008) 170–179
Fig. 3. Graph depicts the number of crossings of the previously trained platform
position vs. a control location (opposite quadrant) in saline treated (saline) and
amphetamine-sensitized (AMPH) animals during probe trials conducted imme-
diately, 24 h or 10 days after the end of training. Both groups crossed the platform
position significantly more often than the control condition during each of the
three test periods (p < 0.05). There were no differences between the two groups.
Error bars depict S.E.M.
the effects of prior amphetamine treatment on the acquisition
phase of the water maze task. A significant main effect was
found for group [F(1, 18) = 10.17, p < 0.05], and session [F(5,
90) = 22.87, p < 0.05], while the group by session interaction
term was not significant. Planned comparison analyses were
conducted which looked at group differences in latency scores
during each daily session. Amphetamine-sensitized animals per-
formed significantly poorer than saline treated animals on days
one [F(1, 18) = 6.36, p < 0.05], four [F(1, 18) = 6.98, p < 0.05]
and six [F(1, 18) = 4.65, p < 0.05]. At the end of acquisition train-
ing on the water maze, animals were given a series of probe tests
(immediate and 24 h). Two separate repeated measure ANOVAs
were used to analyze the number of platform location cross-
ings from these probe tests, with group as a between subjects
factor and platform location (trained platform versus opposite
quadrant location) as a within subjects variable. Fig. 3 depicts
performance during these probe trials. A main effect was found
during the probe tests for platform location [immediate: F(1,
18) = 106.1, p < 0.05 and 24 h: F(1, 18) = 7.86, p < 0.05], with
Fig. 4. Graph depicts performance on the variable platform position water maze task in saline treated (white) and amphetamine-sensitized (black) animals. Saline
treated animals took a greater amount of time to learn the platform location on day 1 and during the first trial on day 3. No other differences were significant. Error
bars depict S.E.M.
animals in the two groups making significantly more crossings
of the location where the platform had been during training in
comparison to a similar location in the opposite quadrant. Nei-
ther group nor the group by platform interaction was significant
during either probe test.
To assess long-term retention of the platform position, ani-
mals were given a single probe test 10 days following the end
of acquisition training. A repeated measure ANOVA was con-
ducted on the number of platform location crossings, with group
as the between subjects factor and location (number of cross-
ings of the trained platform location versus a neutral location
in the opposite quadrant) as the within subjects factor. A sig-
nificant main effect was observed for platform location [F(1,
18) = 24.5, p < 0.05], showing that animals in both groups made
a greater number of crossings of the trained platform location
than they did the neutral location (see Fig. 3). No significance
was observed for the main effect of group or the group × location
interaction.
3.3. Experiment 3: variable platform location water maze
Individual repeated measure ANOVAs were conducted on the
latency scores from each day of testing. Fig. 4 depicts perfor-
mance on the variable platform position water maze task. Each
ANOVA used group as a between subjects factor and trial as a
repeated within subjects variable. Significant main effects were
found for trial on each of the 8 days of training [smallest F(7,
126) = 2.84, p < 0.05]. Additionally, a marginal group effect was
seen on day 1 [F(1, 18) = 4.69, p < 0.05], with saline treated ani-
mals showing higher latencies, and a significant group by trial
interaction was observed on day 3 [F(7, 126) = 2.58, p < 0.05].
Planned comparison analyses suggested that this latter effect
was due to saline treated animals taking a significantly longer
time than amphetamine-sensitized animals in finding the plat-
form during the first trial [F(1, 18) = 4.55, p < 0.05]. Later trials
showed no differences between the two groups.
After acquisition, animals were given additional sessions
which assessed retention of a novel platform position across
three delay periods (16 s, 1 h and 24 h). Fig. 5 depicts these data.
Repeated measure ANOVAs were conducted separately for each
 R.E. Featherstone et al. / Behavioural Brain Research 189 (2008) 170–179 175

Fig. 5. Graph depicts latency to locate a submerged platform during an initial

trial, when the platform location was novel (trial 1), and during a second trial (trial
2) with the same platform position, in saline treated (saline) and amphetamine-
sensitized (AMPH) animals. The two trials were separated by a delay of 16 s, 1 h
or 24 h. While the two groups showed significantly shorter latencies during the
second trial at both the 16 s and 1 h delays (but not after a 24 h delay), no signifi-
cant differences were observed between the two groups. Error bars depict S.E.M.

delay period, with group serving as the between subject factor

and time (trial 1 versus trial 2) as the repeated within subject
factor. No significant differences were found during any of the
three delay periods for group or for the group by time interac-
tion. The main effect of time was significant during the 16 s [F(1,
18) = 8.95, p < 0.05] and 1 h [F(1, 18) = 23.33, p < 0.05] delay
periods, showing that animals in both groups were able to retain
information about the platform position across the delay period.

3.4. Experiment 4: strategy-shifting task

Fig. 6 depicts the data from the strategy-shift task. Two ani-
mals (one saline and one amphetamine) were removed from the
study due to an inability to learn the initial phase of the task. Data
were analyzed with a repeated measure ANOVA, using group
(amphetamine-sensitized versus saline) and shift type (response
to place versus place to response) as between subject variables
and phase (acquisition versus shift) as a repeated within subject
variable. No significant main effects were found for group, shift
type or phase. Likewise, neither the group by phase nor the three-
way interaction was significant. A significant effect was found
for the interaction between shift type and phase [F(1, 14) = 24.8, Fig. 6. Graph depicts performance on the strategy-shifting task in saline treated
(saline) and amphetamine-sensitized (AMPH) animals. Rats were required to
p < 0.05], likely due to the fact that, overall, animals learned the use either a place or response strategy to locate food on the maze. Once animals
place discrimination more easily than the response discrimina- reached a criterion of 10 consecutive correct trials, they were required to switch
tion. Nonetheless, amphetamine-sensitized animals performed strategies (i.e. from place to response or response to place). No differences were
as well as saline treated animals on both the place to response observed between the two groups in the number of trials needed for animals
and the response to place versions of the task. to reach criteria either during acquisition or reversal. The top graph depicts
performance collapsed across both shift types (place to response and response to
place). The middle graph depicts performance on the place and response versions
3.5. Experiment 5: attentional set-shifting task of the task during acquisition while the bottom graph depicts performance on
the place and response versions of the task after being required to shift strategy.
Data from the attentional set-shifting task were analyzed
using a repeated measure ANOVA, with group as a between 17) = 23.9, p < 0.05], for stage [F(6, 102) = 126.5, p < 0.05] and
subject variable and stage (simple discrimination, complex the group × time interaction term [F(6, 102) = 2.94, p < 0.05].
discrimination, etc.) as the repeated within subject vari- Two planned comparison analyses were conducted to assess
able. A significant main effect was found for group [F(1, group differences during each stage of the task. Significant dif-
176 R.E. Featherstone et al. / Behavioural Brain Research 189 (2008) 170–179
Fig. 7. Graph depicts performance on an attentional set-shifting task. Compared
to saline treated animals (saline) amphetamine-sensitized (AMPH) animals
required a significantly greater (p < 0.05) number of trials to learn the extra-
dimensional shift component of the task. A similar difference was observed for
the third reversal (Rev 3). No differences were observed during any of the other
stages of the task. Error bars depict S.E.M. Asterisks denote significant group
differences.
ferences were observed between the amphetamine-sensitized
and saline treated groups only during the extra-dimensional shift
discrimination [F(1, 17) = 19.3, p < 0.05] and one of the reversals
(reversal 3) [F(1, 17) = 7.1, p < 0.05]. Fig. 7 depicts performance
on the set-shifting task.
3.6. Locomotor activity test
Five weeks following completion of each experiment, the
locomotor response to a challenge dose of amphetamine was
Fig. 8. Following behavioural training, animals were assessed for locomo-
tor changes in response to the administration of an amphetamine challenge
(0.05 mg/kg). The graph depicts the mean number of beam breaks in saline
treated (saline) and amphetamine-sensitized (AMPH) animals for each of
the four experiments (exp 1 = experiment 1, exp 2 = experiment 2, etc.).
Amphetamine-sensitized animals had a significantly greater (p < 0.05) number of
beam breaks than did saline treated animals during the post-injection period for
all four experiments. For simplicity, performance prior to amphetamine treat-
ment (habituation) is not depicted. Error bars depict S.E.M. Asterisks denote
significant group differences.
assessed. Fig. 8 depicts locomotor data from each of the four
groups of animals. Subjects were given an initial 30 min habitua-
tion session, followed by an injection of 0.5 mg/kg amphetamine
and an additional 60 min locomotor test. For each experiment,
two t-test were conducted; one that compared group differences
in locomotion following saline injection and the other that com-
pared group differences following amphetamine treatment. In
all four experiments, amphetamine-sensitized animals had sig-
nificantly higher rates of locomotion than saline treated animals
during the 60 min post injection period [experiment 1, t22 = 2.59;
experiment 2, t14 = 4.58; experiment 3, t14 = 2.29; experiment 4
and 5, t17 = 3.62, all p < 0.05], but not during the 30-min habit-
uation session.
4. Discussion
Prior exposure to a sensitizing regimen of amphetamine had
little impact on the performance of a delayed non-match to posi-
tion task, suggesting that working memory was not altered in
sensitized animals. Operant based working memory tasks such
as the DNMTP have been criticized as being open to the influ-
ence of mediating behaviours that can occur during delay periods
and which aid in the retention of positional information [6,22].
While we attempted to minimize such behaviours by installing
barriers between the lever and the magazine, and by requiring
rats to perform a nose poke task during the delay period, it is
difficult to rule out the possibility that such behaviours may
have affected task performance. The fact that both groups of
animals showed increasing rates of error as the delay interval
increased suggests that these animals were relying to a large
extent on memory processes to perform the task. In additional
support of this position, sensitized animals did not show any
impairment on the variable platform position task in the water
maze when assessed on the short-term retention (16 s) of infor-
mation about the platform position. Taken together, the two sets
of results suggest that working memory ability was unaffected
by the induction of amphetamine sensitization.
Two prior studies have examined the impact of amphetamine
sensitization on working memory in rodents, using delayed alter-
nation paradigms [47,50]. As with the present experiments, both
studies showed a lack of impairment of working memory fol-
lowing chronic amphetamine exposure. In contrast, impairments
in the performance of a delayed response task were observed in
non-human primates following the induction of amphetamine
sensitization [5]. While this discrepancy could be suggested to
reflect differences in amphetamine regimen or differences in the
nature of the tasks, the present study would seem to rule out such
explanations. First, the amphetamine regimen used here was
clearly effective in inducing robust sensitization, as indicated
by increased locomotor activity to an amphetamine challenge.
Secondly, both the rodent and primate studies have used delayed
response tasks that are quite similar to one another. It is possible,
given that many of the sensitized animals in the primate study
showed difficulties at the 0 s delay when trained following the
establishment of sensitization [5], that the deficits observed in
these animals could have been partly due to the disruption of
some non-mnemonic process, such as attention.
 R.E. Featherstone et al. / Behavioural Brain Research 189 (2008) 170–179
Working memory performance has been linked to altered
dopaminergic neurotransmission in the prefrontal cortex. Prior
exposure to a sensitizing regimen of amphetamine or cocaine
disrupts multiple aspects of prefrontal cortex function, includ-
ing blunting dopaminergic function [5,25,48,54]. In primates,
reduced prefrontal dopaminergic function has been shown to be
associated with poor working memory [45,46,55]. In rats, how-
ever, the picture is less clear-cut. For example, direct infusion
of the dopamine D1 receptor antagonist, SCH23390 into the
medial prefrontal cortex in rats produced performance impair-
ments rather than memory impairments in a DNMTP task similar
to the one used here [3]. SCH23390 also failed to affect working
memory in a delayed alteration task in rats [56]. Interestingly,
that study showed that a high dose of a D1 receptor agonist
impaired working memory, but that this disruption could be
blocked by administration of a D1 antagonist. [56]. These data
suggest that in the rat it is increased, rather than reduced, PFC DA
function that impairs working memory. Given that amphetamine
sensitization appears to decrease prefrontal dopamine activity
[5,25,48,54] it is not surprising that working memory remains
intact in sensitized rodents. Alternatively, tasks typically used to
assess working memory in rodents often heavily depend upon
the use of hippocampal memory processes [4], a factor which
may make the performance of such tasks less dependent upon
prefrontal involvement and, thus, less susceptible to disruption
by the amphetamine-induced sensitized state.
Amphetamine-sensitized animals also failed to show an
impairment in long-term, hippocampal dependent memory, as
assessed using either a fixed platform or variable platform posi-
tion water maze task. Sensitized animals showed slightly higher
escape latencies on the fixed platform position water maze task
on days 1, 4 and 6 of training. That this impairment was due
to a disruption in spatial learning and memory is unlikely given
that no differences were seen between the two groups during the
post acquisition probe tests carried out immediately and 24 h
following the last trial of training. Such probe tests represent a
purer measure of learning, since animals are started from a novel
position on the maze and, as a result, are unable to rely upon non-
spatial navigation strategies, such as route learning or response
learning, which may be used when started from a previously used
start position. Animals use both spatial and non-spatial infor-
mation simultaneously on navigation tasks [9,10,29,33] and it
is possible that the deficit observed during acquisition reflected
a change in ability in non-spatial memory. No differences were
observed during retesting on the water maze 10 days follow-
ing the end of acquisition training, demonstrating that the two
groups were able to retain information about the platform posi-
tion equally well over the retention interval. Sensitized animals
also showed normal performance of the variable platform posi-
tion task. In this task performance depends upon the use of
a number of different cognitive processes and, hence, may be
dependent upon a wide network of neural structures. For exam-
ple, in addition to remembering a particular platform position
(spatial memory), animals need to develop strategies to effi-
ciently search the water maze (strategy learning), an ability that
appears to be dependent, in part, upon the prefrontal cortex [26],
and animals are required to remember the novel platform loca-
177
tion over a short delay from the first to second trial (working
memory) [15]. Thus, it was expected that the variable platform
task would provide a more demanding test of memory than that
provided by the fixed platform position task. The fact that per-
formance was not disrupted on the variable platform task further
strengthens the conclusion that long-term memory is not altered
by the induction of amphetamine sensitization.
The lack of effect of amphetamine sensitization on water
maze performance reported in the present experiment is consis-
tent with the results of a previous study that showed no effect of
amphetamine sensitization on either the acquisition or retention
of a fixed platform position water maze task [41]. Sensitized
animals did show a decrease in latency to find a novel platform
position following training (interpreted as reversal learning), as
well as an increase in number of crossings of the former platform
position during probe trials relative to saline treated animals.
While this latter effect may suggest improved spatial learning
in amphetamine treated animals, the effect was not large and
was not always observed. Thus, while deficits in long-term,
hippocampal dependent memory have been widely reported
in schizophrenia [7,23,24], such deficits are not reproduced in
rodents following the induction of amphetamine sensitization.
It is likely that amphetamine sensitization is not a useful model
of this aspect of schizophrenia.
The final two experiments assessed behavioural flexibil-
ity in amphetamine-sensitized rats using both a radial maze
based strategy selection task and an attention set-shifting task.
Amphetamine-sensitized animals did not show an impairment
on the radial maze based strategy-shifting task. In contrast,
amphetamine-sensitized animals were impaired in performance
of the extra-dimensional shift component of the set-shifting task,
despite showing good performance during the earlier stages of
the task, as has been previously reported [16]. Superficially, the
two tasks would appear to be measuring attentional set-shifting
and both have been shown to be impaired by disruptions of pre-
frontal cortex function [1,19,28,37–39]. Likewise, deficits have
been reported on a radial maze based strategy selection task fol-
lowing the establishment of cocaine sensitization [21], although
this was limited to an increase in perseverative responding in
sensitized animals with no change in number of trials needed to
reach criterion. Further, interfering with prefrontal cortex activ-
ity by direct infusion of the dopamine antagonist SCH23390, a
manipulation which may mirror the changes in dopamine func-
tion induced following amphetamine sensitization, produces a
deficit in switching memory strategy [36]. Taken together, these
results suggest that the two tasks are both mediated by the
prefrontal cortex. As such, it is surprising that amphetamine
sensitization did not induce similar impairments in performance
of the two tasks.
A number of explanations could be put forth to explain the dif-
ferential effects of amphetamine sensitization on the set-shifting
and strategy-shifting tasks. First, major procedural differences
between the two tasks exist that could place different demands
on cognitive processing. For example, on the set-shifting task,
animals learn a series of discriminations in which a particular
stimulus dimension is always used to signal reinforcement. The
relevant stimulus dimension is repeatedly reinforced across dif-
178 R.E. Featherstone et al. / Behavioural Brain Research 189 (2008) 170–179
ferent stimulus pairings/combinations, most likely resulting in
the development of a strong attentional set. In contrast, animals
on the strategy-shifting task only experience one particular dis-
crimination, wherein a single stimulus or collection of stimuli
signals reinforcement over a number of trials. As such, it is pos-
sible that animals develop a much stronger attentional set on the
set-shifting task than on the strategy-shifting task and, as a result,
find the shift required by the former task more difficult. It should
be noted that, while robust, the deficit observed on the attentional
set-shifting task following sensitization is relatively mild in
comparison to that observed following lesions of the prefrontal
cortex (as reported by [1]). It is possible that the strategy-shifting
task is not sensitive enough to detect changes in the relatively
mildly impaired sensitized animal, despite being sufficient to
detect changes in the more impaired prefrontal lesioned animal.
The attentional set-shifting task differs from the other tasks
used in the present study in that it uses odor cues and, thus,
places a demand on olfactory processing. Some evidence sug-
gests that acute treatment with psychostimulants may impair
olfactory-based learning. For example, one study showed that
exposure to cocaine could disrupt performance of an odor-based
working memory task in animals chronically exposed to cocaine
through self-administration training, despite the fact that such
training failed to alter the ability of cocaine to disrupt perfor-
mance on a similar visual based memory task [27]. Likewise,
animals exposed to amphetamine showed reduced detection
rates for an odor cue on a signal detection task [11]. It could
thus be argued that amphetamine-sensitized rats were deficient
on the attentional set-shifting task because of altered olfactory
processing induced by amphetamine treatment. However, this
is unlikely for a number of reasons. First, the present animals
were tested in a drug-free state, and neither of these previous
studies allows us to determine whether a sensitizing regimen of
amphetamine is sufficient to disrupt olfactory processing in the
absence of the acute effects of the drug. Second, amphetamine-
sensitized animals were equally impaired when required to make
an extradimensional shift from odor to texture or from texture to
odor, suggesting that the deficits observed in sensitized animals
on the attentional set-shifting task also occur in the somatosen-
sory modality. Finally, if the amphetamine-sensitized state alters
olfactory based learning then it might be predicted that sensi-
tized rats would be impaired on other types of discriminations
involving odor, most notably the compound discrimination when
a second irrelevant stimulus dimension was introduced. This did
not occur; amphetamine-sensitized rats were impaired only on
the EDS.
The current study assessed working and long-term memory in
rats that had undergone amphetamine sensitization. While prior
work has demonstrated that this model replicates some of the
cognitive deficits associated with schizophrenia, amphetamine-
sensitized animals did not show evidence of memory deficits in
any of the tasks used in the current experiments. These results
suggest that amphetamine sensitization does not produce simi-
lar memory impairments as are seen in schizophrenia, and, thus,
might not be an adequate model of this aspect of schizophrenic
cognition. The deficit in attentional set-shifting is consistent
with a deficit in cognitive flexibility, a common feature of
schizophrenia, although the lack of effect on the similar strategy-
shifting task suggests that the cognitive processes affected by
sensitization may be specific to those required to complete an
extradimensional shift. As such, amphetamine sensitization can-
not be regarded as a complete model of schizophrenia, as it
does not capture the entirety of cognitive deficit observed in
this disease. The reliability with which extradimensional shift
performance is disrupted provides further confirmation that
amphetamine sensitization is a good model of the attentional
disruptions associated with schizophrenia.
Acknowledgements
This work was supported by a special initiative grant from
the Ontario Mental Health Foundation. REF was supported by a
Research Fellowship from the Ontario Mental Health Founda-
tion and SK was supported by a Canadian Research Chair.
References
[1] Birrell JM, Brown VJ. Medial frontal cortex mediates perceptual attentional
set shifting in the rat. J Neurosci 2000;20:4320–4.
[2] Bisagno V, Ferguson D, Luine VN. Chronic D-amphetamine induces sex-
ually dimorphic effects on locomotion, recognition memory, and brain
monoamines. Pharmacol Biochem Behav 2003;74:859–67.
[3] Broersen LM, Heinsbroek RP, de Bruin JP, Uylings HB, Olivier B. The role
of the medial prefrontal cortex of rats in short-term memory functioning:
further support for involvement of cholinergic, rather than dopaminergic
mechanisms. Brain Res 1995;674:221–9.
[4] Castner SA, Goldman-Rakic PS, Williams GV. Animal models of work-
ing memory: insights for targeting cognitive dysfunction in schizophrenia.
Psychopharmacology (Berl) 2004;174:111–25.
[5] Castner SA, Vosler PS, Goldman-Rakic PS. Amphetamine sensitization
impairs cognition and reduces dopamine turnover in primate prefrontal
cortex. Biol Psychiatry 2005;57:743–51.
[6] Chudasama Y, Muir JL. A behavioural analysis of the delayed non-
matching to position task: the effects of scopolamine, lesions of the fornix
and of the prelimbic region on mediating behaviours by rats. Psychophar-
macology (Berl) 1997;134:73–82.
[7] Clare L, McKenna PJ, Mortimer AM, Baddeley AD. Memory in
schizophrenia: what is impaired and what is preserved? Neuropsychologia
1993;31:1225–41.
[8] Deller T, Sarter M. Effects of repeated administration of amphetamine on
behavioral vigilance: evidence for “sensitized” attentional impairments.
Psychopharmacology (Berl) 1998;137:410–4.
[9] Devan BD, McDonald RJ, White NM. Effects of medial and lateral caudate-
putamen lesions on place- and cue-guided behaviors in the water maze:
relation to thigmotaxis. Behav Brain Res 1999;100:5–14.
[10] Devan BD, White NM. Parallel information processing in the dorsal stria-
tum: relation to hippocampal function. J Neurosci 1999;19:2789–98.
[11] Doty RL, Ferguson-Segall M. Odor detection performance of rats following
d-amphetamine treatment: a signal detection analysis. Psychopharmacol-
ogy (Berl) 1987;93:87–93.
[12] Dunnett SB, Evenden JL, Iversen SD. Delay-dependent short-term memory
deficits in aged rats. Psychopharmacology (Berl) 1988;96:174–80.
[13] Featherstone RE, Kapur S, Fletcher PJ. The amphetamine-induced sensi-
tized state as a model of schizophrenia. Prog Neuropsychopharmacol Biol
Psychiatry 2007;31:1556–71.
[14] Featherstone RE, Rizos Z, Nobrega JN, Kapur S, Fletcher PJ. Gestational
methylazoxymethanol acetate treatment impairs select cognitive functions:
parallels to schizophrenia. Neuropsychopharmacology 2007;32:483–92.
[15] Ferbinteanu J, Ray C, McDonald RJ. Both dorsal and ventral hippocam-
pus contribute to spatial learning in Long–Evans rats. Neurosci Lett
2003;345:131–5.
 R.E. Featherstone et al. / Behavioural Brain Research 189 (2008) 170–179
[16] Fletcher PJ, Tenn CC, Rizos Z, Lovic V, Kapur S. Sensitization to
amphetamine, but not PCP, impairs attentional set shifting: reversal by
a D(1) receptor agonist injected into the medial prefrontal cortex. Psy-
chopharmacology (Berl) 2005;183:190–200.
[18] Fletcher PJ, Tenn CC, Sinyard J, Rizos Z, Kapur S. A sensitizing reg-
imen of amphetamine impairs visual attention in the 5-choice serial
reaction time test: reversal by a D1 receptor agonist injected into
the medial prefrontal cortex. Neuropsychopharmacology 2007;32:1122–
32.
[19] Floresco SB, Magyar O, Ghods-Sharifi S, Vexelman C, Tse MT. Multi-
ple dopamine receptor subtypes in the medial prefrontal cortex of the rat
regulate set-shifting. Neuropsychopharmacology 2006;31:297–309.
[20] Goldman-Rakic PS. Working memory dysfunction in schizophrenia. J Neu-
ropsychiatry Clin Neurosci 1994;6:348–57.
[21] Goto Y, Grace AA. Dopaminergic modulation of limbic and cortical
drive of nucleus accumbens in goal-directed behavior. Nat Neurosci
2005;8:805–12.
[22] Gutnikov SA, Barnes JC, Rawlins JN. Working memory tasks in five-
choice operant chambers: use of relative and absolute spatial memories.
Behav Neurosci 1994;108:899–910.
[23] Hanlon FM, Weisend MP, Hamilton DA, Jones AP, Thoma RJ, Huang M,
et al. Impairment on the hippocampal-dependent virtual Morris water task
in schizophrenia. Schizophr Res 2006;87:67–80.
[24] Heckers S, Rauch SL, Goff D, Savage CR, Schacter DL, Fischman AJ, et
al. Impaired recruitment of the hippocampus during conscious recollection
in schizophrenia. Nat Neurosci 1998;1:318–23.
[25] Hedou G, Homberg J, Feldon J, Heidbreder CA. Expression of sensiti-
zation to amphetamine and dynamics of dopamine neurotransmission in
different laminae of the rat medial prefrontal cortex. Neuropharmacology
2001;40:366–82.
[26] Hoh TE, Kolb B, Eppel A, Vanderwolf CH, Cain DP. Role of the neocortex
in the water maze task in the rat: a detailed behavioral and Golgi-Cox
analysis. Behav Brain Res 2003;138:81–94.
[27] Kantak KM, Udo T, Ugalde F, Luzzo C, Di Pietro N, Eichenbaum HB.
Influence of cocaine self-administration on learning related to prefrontal
cortex or hippocampus functioning in rats. Psychopharmacology (Berl)
2005;181:227–36.
[28] McAlonan K, Brown VJ. Orbital prefrontal cortex mediates reversal
learning and not attentional set shifting in the rat. Behav Brain Res
2003;146:97–103.
[29] McDonald RJ, White NM. Parallel information processing in the water
maze: evidence for independent memory systems involving dorsal striatum
and hippocampus. Behav Neural Biol 1994;61:260–70.
[30] Murphy CA, Di Iorio L, Feldon J. Effects of psychostimulant with-
drawal on latent inhibition of conditioned active avoidance and prepulse
inhibition of the acoustic startle response. Psychopharmacology (Berl)
2001;156:155–64.
[31] Murphy CA, Fend M, Russig H, Feldon J. Latent inhibition, but not pre-
pulse inhibition, is reduced during withdrawal from an escalating dosage
schedule of amphetamine. Behav Neurosci 2001;115:1247–56.
[32] Ornstein TJ, Iddon JL, Baldacchino AM, Sahakian BJ, London M, Everitt
BJ, et al. Profiles of cognitive dysfunction in chronic amphetamine and
heroin abusers. Neuropsychopharmacology 2000;23:113–26.
[33] Packard MG, McGaugh JL. Inactivation of hippocampus or caudate nucleus
with lidocaine differentially affects expression of place and response learn-
ing. Neurobiol Learn Mem 1996;65:65–72.
[34] Pantelis C, Barber FZ, Barnes TR, Nelson HE, Owen AM, Robbins TW.
Comparison of set-shifting ability in patients with chronic schizophrenia
and frontal lobe damage. Schizophr Res 1999;37:251–70.
[35] Paulson PE, Camp DM, Robinson TE. Time course of transient behavioral
depression and persistent behavioral sensitization in relation to regional
brain monoamine concentrations during amphetamine withdrawal in rats.
Psychopharmacology (Berl) 1991;103:480–92.
179
[36] Ragozzino ME. The effects of dopamine D(1) receptor blockade in
the prelimbic-infralimbic areas on behavioral flexibility. Learn Mem
2002;9:18–28.
[37] Ragozzino ME, Detrick S, Kesner RP. Involvement of the prelimbic-
infralimbic areas of the rodent prefrontal cortex in behavioral flexibility
for place and response learning. J Neurosci 1999;19:4585–94.
[38] Ragozzino ME, Kim J, Hassert D, Minniti N, Kiang C. The contribution
of the rat prelimbic-infralimbic areas to different forms of task switching.
Behav Neurosci 2003;117:1054–65.
[39] Ragozzino ME, Wilcox C, Raso M, Kesner RP. Involvement of rodent
prefrontal cortex subregions in strategy switching. Behav Neurosci
1999;113:32–41.
[40] Robinson TE, Becker JB. Enduring changes in brain and behavior produced
by chronic amphetamine administration: a review and evaluation of animal
models of amphetamine psychosis. Brain Res 1986;396:157–98.
[41] Russig H, Durrer A, Yee BK, Murphy CA, Feldon J. The acquisition, reten-
tion and reversal of spatial learning in the morris water maze task following
withdrawal from an escalating dosage schedule of amphetamine in Wistar
rats. Neuroscience 2003;119:167–79.
[42] Russig H, Murphy CA, Feldon J. Clozapine and haloperidol reinstate latent
inhibition following its disruption during amphetamine withdrawal. Neu-
ropsychopharmacology 2002;26:765–77.
[43] Russig H, Murphy CA, Feldon J. Prepulse inhibition during withdrawal
from an escalating dosage schedule of amphetamine. Psychopharmacology
(Berl) 2003;169:340–53.
[44] Russig H, Murphy CA, Feldon J. Behavioural consequences of withdrawal
from three different administration schedules of amphetamine. Behav Brain
Res 2005;165:26–35.
[45] Sawaguchi T, Goldman-Rakic PS. D1 dopamine receptors in prefrontal
cortex: involvement in working memory. Science 1991;251:947–50.
[46] Sawaguchi T, Goldman-Rakic PS. The role of D1-dopamine receptor
in working memory: local injections of dopamine antagonists into the
prefrontal cortex of rhesus monkeys performing an oculomotor delayed-
response task. J Neurophysiol 1994;71:515–28.
[47] Shoblock JR, Maisonneuve IM, Glick SD. Differences between d-
methamphetamine and d-amphetamine in rats: working memory, tolerance,
and extinction. Psychopharmacology (Berl) 2003;170:150–6.
[48] Sorg BA, Davidson DL, Kalivas PW, Prasad BM. Repeated daily cocaine
alters subsequent cocaine-induced increase of extracellular dopamine in
the medial prefrontal cortex. J Pharmacol Exp Ther 1997;281:54–61.
[49] Steele RJ, Morris RG. Delay-dependent impairment of a matching-to-place
task with chronic and intrahippocampal infusion of the NMDA-antagonist
D-AP5. Hippocampus 1999;9:118–36.
[50] Stefani MR, Moghaddam B. Effects of repeated treatment with
amphetamine or phencyclidine on working memory in the rat. Behav Brain
Res 2002;134:267–74.
[51] Tenn CC, Fletcher PJ, Kapur S. Amphetamine-sensitized animals show
a sensorimotor gating and neurochemical abnormality similar to that of
schizophrenia. Schizophr Res 2003;64:103–14.
[52] Tenn CC, Fletcher PJ, Kapur S. A putative animal model of the “prodromal”
state of schizophrenia. Biol Psychiatry 2005;57:586–93.
[53] Tenn CC, Kapur S, Fletcher PJ. Sensitization to amphetamine, but not
phencyclidine, disrupts prepulse inhibition and latent inhibition. Psy-
chopharmacology (Berl) 2005;180:366–76.
[54] Vanderschuren LJ, Schmidt ED, De Vries TJ, Van Moorsel CA, Tilders FJ,
Schoffelmeer AN. A single exposure to amphetamine is sufficient to induce
long-term behavioral, neuroendocrine, and neurochemical sensitization in
rats. J Neurosci 1999;19:9579–86.
[55] Williams GV, Goldman-Rakic PS. Modulation of memory fields by
dopamine D1 receptors in prefrontal cortex. Nature 1995;376:572–5.
[56] Zahrt J, Taylor JR, Mathew RG, Arnsten AF. Supranormal stimulation
of D1 dopamine receptors in the rodent prefrontal cortex impairs spatial
working memory performance. J Neurosci 1997;17:8528–35.