Functional magnetic resonance imaging of Parkinson*s disease patients to investigate the effects of dopamine on visual attention.

Parkinson *s disease (PD) is a disease of the central nervous system. Patients
typically present with motor symptoms, and these remain the primary target of
clinical interventions. Nevertheless, it has long been recognized that PD is
also characterized by non-motor symptoms, such as autonomic dysregulations,
cognitive dysfunctions and neuropsychiatric disorders, and these have
increasingly become the focus of study (e.g. Bosboom et al., 2004). Such
symptoms are often related to degeneration of the dopaminergic system, and to
medication that works through this system, such as L-DOPA or DA-agonists. In
particular, it has been suggested that decision-making is altered both in the
state of relative dopamine deprivation that characterizes PD patients during
withdrawal from medication, and also in the medicated state itself (e.g., Frank
et al., 2004). Such alterations of decision-making have in turn been linked to
changes in behaviour of PD patients on medication, in particular to a proneness
to gamble, hypersexuality, and compulsive eating and buying (Seedat et al.,
2000; Wolters et al., 2008; Van den Heuvel et al., 2010 ). These are
collectively referred to as impulse control disorders (ICDs).
We recently showed that previously rewarded stimuli automatically
capture attention, even when that stimulus is no longer rewarding (Hickey et
al., 2010). Disruption of pathways from the dorsal part of the substantia nigra
(SN) and the ventral tegmental area (VTA) to the nucleus accumbens (NAcc) and
caudate nucleus (CN) in PD patients has been associated with impaired
motivational functions (Middleton & Strick, 2000a,b). Some studies have
suggested a role for dopaminergic reward processing in guiding attention by
altering the priority given to certain features over others (Berridge &
Robinson, 1998; Hickey et al., 2010; Anderson et al. ,2011a, b). These and
other results have led to the suggestion that dopamine may be important to the
assignment of priority to environmental stimuli. Abnormalities in dopaminergic
neurotransmission in PD may lead to aberrant prioritization of stimuli which
could potentially explain the reward-driven behaviours in PD patients with ICD
(i.e., gambling, hypersexuality, binge eating). In other words, these
reward-driven behaviours may be explained by processes that automatically steer
attention to valued objects.
The present study is premised on the idea that prioritization of
sensory input will be negatively affected in PD patients. Patients will show
decreased sensitivity of attention to reward contingencies in the environment
(i.e. a decreased tendency to pay attention to rewarding stimuli). Dopaminergic
drug treatments should restore normal processing, as compared to controls, or
perhaps even enhanced processing of sensory input in the case of ICD patients,
as a result of normal or enhanced sensitivity to reward contingencies
respectively.
The primary goal of this study is thus to investigate the role of
dopamine in reward-modulated attention and visual processing in PD patients
with varying levels of ICD symptoms. For example, it may be that extra
attention given to rewarded features leading to positive reinforcement is an
important element in learning (Roelfsema et al., 2010; Frank et al., 2004;
Jocham et al., 2011 ) which may contribute to the cognitive and
neuropsychiatric symptomatology of PD. Greater understanding of learning
mechanisms in PD patients may thus have practical implications, since
dopaminergic medication likely underlies behavioural alterations such as ICD.
If alterations in attention due to dopaminergic medication indeed underlie
deficits in learning and decision making, the research may also guide the
development of strategies to mitigate these behavioural abnormalities.

We aim to extend our recent work on fundamental mechanisms that underlie
automatic influences of reward on attention to clinical populations that suffer
from reduced and excessive reward-driven behaviours. That is, we will
investigate whether impaired reward processing in PD patients is caused by
alterations in selective attention. We will examine this mechanism in PD
patients with varying levels of ICD symptoms, as we expect individuals with
more ICD symptoms to show greater sensitivity to reward value and a greater
automaticity in the allocation of attention toward rewarding objects than
patients with lower levels or with no ICD symptoms.
Firstly, we plan to establish that all PD patients in a state of
overnight withdrawal from medication are less sensitive to reward in their
attentive response than normal controls. We will investigate this by examining
known fMRI correlates of reward processing. We are particularly interested in
regions where activation is elicited by reward feedback; ventral tegmental area
(VTA), substantia nigra pars compacta (SNpc), and the basal ganglia (BG),
including the ventral striatum (VSr), and their relationship to other regions
that indicate the deployment of selective attention (visual cortex; areas
V1-V4, and lateral occipital cortex (LOC)). Our expectation is that patients
who are off medication should be less sensitive to reward than controls,
reflected in reduced visual attention for rewarded stimuli. When the same
patients are on medication, they should be equally or more sensitive than
controls. In the case of patients with greater ICD symptoms, they should be
more sensitive to rewarded stimuli when on medication (Frank et al., 2004) and
should show enhanced patterns of visual attention toward stimuli, as compared
to controls. Selective visual attention will be quantified using an fMRI data
analysis technique called multivariate pattern analysis (MVPA).
Secondly, in addition to analysing task-related fMRI BOLD signal to
investigate our hypotheses, we will also look at resting-state fMRI activity,
in particular the default mode network (DMN) and visual attention networks. We
intend to assess how reward-related regions are connected to occipital visual
areas. In line with our hypotheses, we expect to see differences in the
functional connectivity of the basal ganglia to visual areas between patients
and controls, and also within patients when they are on and off their
medication.
As a final objective, we aim to relate the amount of pain participants
experience as a result of their disease to connectivity measures of
resting-state fMRI activity. Alterations in the DMN have been associated with
pain in individuals experiencing various forms of chronic pain, in particular
decreased connectivity of the medial prefrontal cortex (MPFC) to posterior
parts of the DMN, and increased connectivity to the insular cortex (Baliki et
al., 2014; Otti et al., 2013). In this study, pain-related questionnaires will
be included during the neuropsychological assessment phase, and patients will
give a brief update before and at a particular point during fMRI scanning.
Correlations between these measures and resting-state fMRI activity will then
be investigated.

A schematic of the study design is provided in Fig. 1 (see Protocol). Patients
will first undergo neuropsychological assessment. For the fMRI sessions, we
will perform a counterbalanced repeated measures study in which patients are
scanned once during a state of overnight withdrawal from L-DOPA or 24 hour
withdrawal from DA agonists, and once in a medicated state. This means that
every patient undergoes the task session twice. On one day, they will undergo
our task experiment before they have taken their morning medication - this is
the state of withdrawal. On an evening later that same weekend (for one half of
patients), or on an evening earlier that weekend (for other half of patients),
they are tested after ingestion of their standard doses of dopamine medication
- this is the medicated state. Both fMRI sessions occur in the same weekend,
but are not on the same day, so that effects of an un-medicated morning don*t
spill into the second fMRI session. The main neurophysiological difference
between the withdrawal and the medicated state is thus the level of dopamine
available within the brain of the patient. The counterbalancing of the on/off
session being first or second is used to control for the effects of repeated
testing. Healthy controls will undergo one fMRI session, with one half of
controls participating in the morning and the other half in the evening.

The experimental task is the same for the two sessions, however the trial
sequence is randomized in each. Participants visit the Transitorium building of
the Vrije Universiteit (VU) on one occasion and are subsequently scanned at the
out-patients building of the VU Medical Centre, VUmc (Polikliniekgebouw, De
Boelelaan, Amsterdam) over two separate fMRI sessions.

Cognitive performance measurements of PD patients when on L-DOPA medication and
during a practically-defined off-medication state (i.e. following overnight
withdrawal of dopamine medication).

There will be no direct benefit for the subjects to participate in the study.
However, participation is expected to result in increased insight into the
neurobiological background of the disorder, which will contribute subsequently
to the development of treatment alternatives, and particularly the mitigation
of the cognitive side effects associated with pharmacological treatment.
Cognitive impairments and especially impairments in visual attention in PD have
been relatively under-investigated, as compared to the motor deficits
associated with the disease. This project aims to dissociate effects on visual
selective attention as a result of the disease itself, the prescribed
medication, and the level of ICD symptoms in these patients. Results of this
project will also have important implications in other groups of people with
disturbances in their reward system, such as patients with addiction or
obsessive compulsive disorder.

The risks can be considered minimally above negligible. In fMRI a strong
magnetic field and radio waves are used to generate radio signals in the body.
These signals are picked up by an antenna (frame), which is placed over the
participant*s head. Patients will hear various noises at varying volumes during
scanning. Thus, ear protection is always provided to reduce noise levels. In
addition, an emergency button is always available in case the patient
experiences too much discomfort and wishes to be taken out of the scanner.
Patients may also experience some discomfort during withdrawal from medication.
To minimize the duration of this, patients will be provided either with
accommodation close to the imaging centre the evening before their
off-medication session, or a taxi to chauffer them directly to the VUmc so that
they can partake in the session in the early morning. A medical doctor will be
on-site at the VUmc in case any assistance is required. Moreover, patients will
be encouraged to take their medication if they feel excessive discomfort.