Effects of the norepinephrine reuptake inhibitor reboxetine on the exploitation-exploration trade-off

In daily life, people often have to choose between exploiting known sources of
reward and exploring new, and potentially better, contexts. This trade-off
between exploitation and exploration is crucial for adaptive behavior,
especially in changing environments. Recently, cognitive neuroscientists have
addressed the question how the exploitation-exploration trade-off is regulated
in the human brain (for an overview see Cohen, McClure, & Yu, 2007). Cell
recording studies and model simulations have suggested that the locus
coeruleus-norepinephrine (LC-NE) system plays an important role in the
regulation of this trade-off (e.g. Aston-Jones et al., 2000; Usher et al.,
1999; Yu & Dayan, 2005).
The nucleus locus coeruleus (LC) is situated in the brainstem and projects
widely to all levels of the brain where it releases the neuromodulating
substance norepinephrine (NE). There is widespread evidence that LC-mediated
noradrenergic innervation leads to a temporary increase in the responsivity of
efferent cortical neurons, which is thought to facilitate stimulus processing
(reviewed in Berridge & Waterhouse, 2003). Although to date it has not been
possible to directly measure the activation dynamics of the LC-NE system in
humans, cell recordings in non-human primates have yielded a wealth of
information regarding these dynamics. These studies have identified two
components of LC activity: the spontaneous (baseline) activity, which is
referred to as tonic activity, and the transient increase in activity in
response to motivationally salient stimuli, which is referred to as phasic
activity. The tonic and phasic LC activities interact, in such a way that
intermediate tonic activity is associated with large phasic responses, whereas
low or high tonic activity are associated with weak phasic responses
(Aston-Jones et al., 2000). It has been found that when a monkey is performing
well on a selective attention task, the monkey*s LC neurons exhibit
intermediate tonic activity and strong phasic responses to target stimuli. This
LC state has been referred to as the *phasic mode*. During periods of impaired
attentional performance, on the other hand, the monkey*s LC neurons exhibit a
high level of tonic activity but weak or absent phasic responses to target
stimuli. This LC state has been referred to as the *tonic mode*. It has been
suggested that the tonic and phasic LC activities play complementary roles in
regulating the balance between exploitation and exploration (e.g., Aston-Jones
et al.). In the phasic mode, norepinephrine is released specifically in
response to task-relevant events, thereby facilitating the processing of those
events which promotes exploitation. In the tonic mode, on the other hand, the
sustained release of norepinephrine facilitates the processing of all events,
regardless of their relevance for the current task, which promotes exploration.
Computational modeling studies have provided support for a role of the LC-NE
system in the regulation of the exploitation-exploration trade-off (Usher et
al., 1999; Yu & Dayan, 2005). However, empirical studies testing these
hypotheses in humans have not been conducted yet. One way to address this issue
is to manipulate the LC-NE system pharmacologically. The proposed study will do
this by using the selective norepinephrine reuptake inhibitor reboxetine.
Reboxetine inhibits the reuptake of NE, which increases the availability of NE
and thus shifts the LC towards the tonic mode. Based on the hypotheses of the
role of the LC-NE system in the exploration-exploitation trade-off, the
expected effect of reboxetine is an increase in explorative behavior.
Besides its effect on the noradrenergic system, reboxetine also has a general
effect on alertness. To control for this effect we will test a control group
that receives the selective serotonin reuptake inhibitor citalopram. Citalopam
is used as a positive control because it has a similar alerting effect as
reboxetine, but does not affect the noradrenergic system. Citalopram does
affect the serotonin system, but there is no evidence that this system is
involved in the exploration-exploitation trade-off. Therefore, the differences
in exploitative/explorative behavior between the reboxetine and citalopram
group is likely to reflect the noradrenergic manipulation. Finally, we will
also test a group of participants who receive a passive placebo.

The objective of this study is to assess the effects of an increased NE level
on the trade-off between exploitation and exploration. More specifically, this
study will compare the exploitative/explorative behavior of participants who
received either reboxetine, citalopram or a placebo in various cognitive tasks.
Secondary objective is to evaluate the pharmacokinetics and pharmacodynamics of
reboxetine.

The study is a double-blind placebo-controlled intervention study, using a
between-subjects design. We will test three experimental groups, each comprised
of 16 participants. Participants in group 1 receive a single oral dose of 4 mg
reboxetine. Participants in group 2 receive a single oral dose of 30 mg
citalopram. Participants in group 3 receive a placebo. All experimental
treatments will be administered in identical capsules.
Participants will take part in one experimental session taking approximately 6
hours. The protocol will start with a medical screening. On the study day,
participants will be administered 2 mg of granisetron, to prevent potential
nausea as a side effect of the citalopram. Sixty minutes after administration
of the granisetron, the experimental treatment (reboxetine, citalopram or
placebo) will be administered. Participants then wait for 90 minutes before
they are tested on the three experimental tasks. The total duration of the
tasks is approximately 60 minutes.
After completion of the last cognitive task, participants will perform a simple
reaction time task which lasts for approximately 1 minute. The mean reaction
times on this task will provide an objective measurement of the participants*
alertness.
Before administration of the experimental treatment, in between the
experimental treatment and the first cognitive task, and after each cognitive
task, participants fill in 16 Bond-Lader visual analogue scales measuring
alertness, calmness and contentment (Bond & Lader, 1974).
Upon completing the experimental tasks, participants are kept under supervision
of a physician until the end of the six-hour session. Participants* blood
pressure, pulse, ECG will be measured at regular tine points as well as
pupilsize, saccadic eye movements, body sway, adaptive tracking, cortisol,
ACTH, prolactine, en reboxetine plasma levels.

One experimental group will receive a single oral dose of 4 mg reboxetine.
Reboxetine is a selective norepinephrine reuptake inhibitor. The second
experimental group will receive a single oral dose of 30 mg citalopram.
Citalopram is a selective serotonin reuptake inhibitor. Because citalopram*s
alerting effects are comparable to those of atomoxetine citalopram is used as a
positive control. The third experimental group will receive a placebo
(lactose).

Risks associated with reboxetine intake: Reboxetine is used as a drug for the
treatment of depression. It is well tolerated in man for doses up to 10 mg
daily.

Risks associated with citalopram intake: Citalopram is used as an
antidepressant drug. It is well tolerated in doses of 20 to 60 mg/day. Previous
studies have found that a single dose of citalopram does not have disruptive
effects on psychomotor performance (Lader et al., 1986), does not affect heart
rate and blood pressure (Penttillä et al., 2001), and does not have serious
side effects, apart from a mild feeling of sedation and a dry mouth (Lader et
al., 1986).

Risks associated with the cognitive tasks: there are no risks associated with
the cognitive tasks, except the possibility of some frustration with poor
performance or fatigue.

Benefit of the study: the proposed study is expected to make an important
contribution to our understanding of the role of the noradrenergic system in
the exploitation-exploration trade-off. The importance of the scientific
contribution outweighs the minimal risks involved.