Can the above proposed mechanisms of chronic fatigue hold true across several pathological conditions? The mechanisms proposed thus far identify brain-mediated functions that explain an abnormal perception of effort, a salient feature of chronic fatigue. Chronic fatigue is prevalent in a large number of pathological conditions: neurological, autonomic, immunological, hormonal and cardiovascular diseases not to mention a significant side effect of many pharmacological interventions. Here, I propose sensory attenuation as a fundamental mechanism that underpins effort perception, and aberrant sensory attenuation as a disease-independent mediator of fatigue, which in some cases may be the primary driver and in others, a knock-on effect of a more proximal problem. Fundamentally, fatigue can be viewed as the end point manifestation of a cascade of events activated by disease-specific triggers, and—following resolution of the primary trigger—the cascade of events reverse. However, failure to reverse changes in sensory attenuation results in chronicity of the ensuing symptom; namely, fatigue. The point of reversal failure determines if impoverished sensory attenuation is the driver or a mediator of fatigue. A key aspect of these putative failures rests on the similarity between selective attention and sensory attenuation. In active inference, these are both sides of the same coin corresponding to a centrally mediated increase and decrease in precision (i.e. postsynaptic gain of neurons encoding prediction errors at various levels of processing hierarchies). In other words, there is an intimate relationship between attention and attenuation; both of which have to be carefully orchestrated through descending predictions of precision—or top-down gain control. On this view, disruption to higher order cognitive functions such as endogenous attention may contribute to, or be intimately conflated with, the development of fatigue.
Fatigue and agencyThe inability to suppress anticipated neuromuscular sensory consequences of motor commands as a primary cause of pathological fatigue is proposed here for the first time. However, poor sensory attenuation, as demonstrated by a perceptual behavioural task, is an established correlate of disorders of agency (Lafargue and Franck, 2009). How then, can one reconcile the idea of fatigue and disruption of agency as both driven by similar mechanisms? For this, we must examine the common denominator between the two; namely, effort. In fatigue there is a greater sense of effort and in disorders of agency there is a lack (external attribution) of effort. Research is at very early stages and there is little evidence to draw further conclusions; however, one might plausibly speculate on how effort ties together the two seemingly unrelated phenomena. In fatigue (post-stroke), patients often report a loss of control over their body and simple movements require high effort. In disorders of agency, patients attribute control to an external agent or having no control of their actions. One might speculate that different degrees of reported loss of control could be mapped on to different levels of sensory attenuation with: (i) normal sensory attenuation relating to reports of full control over their movements; (ii) in partial sensory attenuation, patients report partial loss of control over their movements, or find it difficult to move their bodies; and (iii) in a state where sensory attenuation fails completely, a total loss of control over their movements is reported, or loss of agency. Moreover, recent evidence suggests effort and sense of agency interact (Demanet et al., 2013). Therefore, fatigue, a perceptual disorder of the sensorimotor system, the system through which the agent fulfils predictions of agency, can be placed within the spectrum of agency-related disorders.
Concluding remarks and future perspectivesThus far, I have elaborated a framework within which one can define and understand the symptom of fatigue from a physiological standpoint and place it in the broader context of multiple pathologies. Recent investigations provide indirect evidence for pathological fatigue being a disorder of sensory attenuation in neurological conditions. Low motor cortex resting-state excitability in the stroke affected hemisphere in the fatigue group, despite no difference in sensorimotor impairment and behavioural outcome (Kuppuswamy et al., 2015b) raises two questions, the first, ‘does low excitability give rise to fatigue and how?’ Suppression of motor cortex excitability using inhibitory repetitive transcranial magnetic stimulation protocols in healthy volunteers, results in poor sensory attenuation as evidenced by more veridical force matching in the force-matching task, hence low excitability of stroke affected hemisphere may reflect poor sensory attenuation. Second, if sensory attenuation is critical for motor initiation, how might we reconcile the lack of sensorimotor impairment with poor attenuation? In this study (Kuppuswamy et al., 2015b) sensorimotor impairment was measured using standard clinical tests that are not sensitive enough to capture subtle alterations in sensorimotor control. However, in a further study we showed that self-selected ballistic movement speeds were indeed compromised in the affected limb of the high fatigue group (Kuppuswamy et al., 2015a). Limb heaviness in stroke survivors in relation to fatigue but not muscle weakness, indicates abnormal sensory processing of the affected limb, possibly poor attenuation of sensory afferent information from resting state muscle tone in the affected limb? Attention deficits have been related to post-stroke fatigue (Radman et al., 2012) and attention is inextricably linked to sensory attenuation. Fatigue in multiple sclerosis has been thought of as a disorder of movement preparation as evidenced by altered pre-movement motor cortex excitability and movement preparation includes predictions of sensory consequences—a key feature of sensory attenuation. Interventions targeted at altering motor cortex excitability have thus far shown some positive effects on fatigue; however, direct evidence linking poor sensory attenuation to pathological fatigue is yet to emerge. Future investigations could investigate sensory attenuation in patient cohorts with a wide range of fatigue levels. A simple, established, robust behavioural paradigm that allows one to quantify sensory attenuation is the force matching illusion; see above (Shergill et al., 2003). Furthermore, high frequency neuronal oscillatory activity in motor areas that encode prediction errors and movement parameters both appear to be causally linked to sensory attenuation and neuro-modulatory protocols also support a functional role for motor cortex in sensory attenuation. Therefore, using a combination of behavioural, imaging and neuromodulation techniques, future studies can, in principle, confirm or negate the hypothesis: pathological fatigue is a disorder of sensory attenuation. Interestingly, an explanation based on aberrant sensory attenuation for other affective symptoms, is starting to emerge in computational psychiatry—with the combined use of neuroimaging and dynamic causal modelling to measure sensory attenuation and attention in terms of neuronal gain; i.e. the synaptic efficacy of intrinsic neuronal connections.
Fatigue is a perceptual state that is experienced by all humans, albeit transiently, but when fatigue is non-transient it starts to significantly impact all aspects of the sufferer’s lives. Such fatigue is a hallmark of many pathological conditions and despite more than a hundred years of trying to understand fatigue, there has been very little progress. Here, I highlight the features of fatigue that render it difficult for scientific investigation, propose a unifying definition from a physiological standpoint, elaborate on a disease-independent mechanism that might underlie fatigue, discuss evidence in support of the proposed mechanism and suggest further experiments to verify the hypothesis. If proven to be true, this framework may provide us with the much needed foundation on which to build our understanding of fatigue and, more broadly, a robust link between mind and body.
AcknowledgementsI would like to extend my sincere thanks to Profs Karl Friston and John Rothwell for the many useful discussions and valuable input to this manuscript. I would also like to thank Dr Sasha Ondobaka and Mr William De Doncker for their feedback on this manuscript.
FundingA.K. is funded by the Wellcome Trust 202346/Z/16/Z.
10 references used to produce this.