Neural correlates of chronic fatigue syndrome

Chronic fatigue syndrome (CFS) is characterized by profound disabling fatigue
with an unknown aetiology. CFS is currently treated with Cognitive behavioural
therapy (CBT), which has proven to be a successful therapy leading to a
reduction in fatigue and disability. Consistent with cognitive behavioural
models of CFS, recent clinical research has shown that mainly cognitive factors
mediate successful therapy outcome. Accordingly, with support from neuroimaging
studies, it has been suggested that central (cognitive) mechanisms play a role
in CFS and its treatment. This project aims at identifying the neural
correlates of central mechanisms that perpetuate CFS symptoms and underlie the
mechanisms of change of CBT. Our hypotheses are derived from a neurobiological
hierarchical Bayesian model of medically unexplained symptoms (Edwards et al.,
2012) that emphasizes the influence of dysfunctional beliefs on perception.
According t this model, somatoform symptoms arise from an inference failure
between prior beliefs and sensory evidence. Thus, it is hypothesized that
fatigue-related beliefs may bias perception towards experiencing fatigue. This
project aims at investigating neural correlates associated with inference
processes that are thought to underlie CFS symptoms. In addition, we will
assess how these mechanisms change during cognitive behavioural therapy.

Primary objectives:
- Identify neural correlates and behavioural measures that underlie cognitive
processes that perpetuate CFS symptoms
- Identify neural mechanisms of change that mediate successful CBT

We will focus our investigation on three hypotheses: We will assess the
hypothesis that, compared to healthy controls

1> CFS patients show a general tendency to base perceptual decisions more on
prior expectations than on sensory inputs by investigating the influence of
expectancy-cues in a moving-dot paradigm. Identification of such a bias in CFS
patients within the visual domain would indicate a fundamental vulnerability
that may predispose subjects to develop medically unexplained symptoms.

2> CFS patients have a perceptual bias about how errors in physical performance
are interpreted, such that an error due to too little physical effort is
attributed or processed differently than an error due to too much physical
effort. This will be tested by investigating feedback processing and associated
neural mechanisms in a force-estimation task. The finding that CFS patients
have biased feedback processing in a physical performance task provides an
explanation for the systematically found underperformance of CFS patients in
physical exercise tests.

3> CFS patients have a higher tendency to infer contingencies between cues and
outcomes that are inconsistent with sensory evidence, by investigating adaptive
and aberrant salience and associated neural mechanisms in the salience
attribution task. This learning process is proposed to contribute to the
development and retention of dysfunctional beliefs which is proposed to play an
important role in the maintenance and aggregation of symptoms.

Secondary objectives:
- Identify biomarkers (i.e. stress hormones, cytokines and sleep) as correlates
of changes in symptoms after CBT

Phase 1, Pilot: Observational behavioural study in which untreated CFS and
healthy controls will be tested once on computerized behavioural tasks.

Phase 2, RCT: Randomized controlled trial (RCT) in which CFS patients receiving
CBT will be compared with CFS patients on a waiting list and all patients will
be compared with healthy controls.

There is no risk associated with this study. The time that subjects need to
invest in this study encompasses one (pilot) or two (RCT) visits to the Donders
Centre for Cognitive neuro-imaging. These visits wall take about 3* hours
during which subjects perform two tasks in the fMRI scanner and one
computerized behavioural task in a quit testing room. Subjects will have to lie
in an MRI scanner for ~2 hours of which in total 60 minutes includes active
involvement in computerized tasks. Patients in the waiting list condition will
perform one additional baseline assessment (standard protocol for all referred
patients before diagnosis and after treatment) after the second test session
and prior to the start of their treatment. Additional measurements include
saliva sampling, blood sampling and hair sampling.

Patients will receive care as usual provided by the expert centre for chronic
fatigue. Patients in the waiting list condition will not wait longer than usual
before they start with CBT, as the average waiting period for starting
treatment is at least 6 months. The benefit of participating in the study is
that patients will have a 66 percent chance that they can start directly with
CBT instead of waiting for 6 months. Both patients and healthy controls will
receive a small financial compensation for their time investment according to
the Donders Institute regulations on participant compensation.