Issue link: https://www.e-digitaleditions.com/i/485941
New Research on Cognitive Fatigue in Multiple Sclerosis Multiple sclerosis (MS) is a disabling neurological disorder with pathology that involves lesions, plaques and axonal demylenation across the central nervous system. The clinical expression of the disease is extremely heterogeneous across patients, with impaired sensory, motor, emotional and cognitive function. Fatigue may be one of the most common symptoms that MS patients experience and prevalence estimates of fatigue range between 70% and 90% 1 . Operationally defining cognitive fatigue (CF) has proven to be challenging; however, from a patients' perspective, CF might be best characterized as the feeling of being "mentally drained". The experience of CF can result in subjective sensations or objective performance changes, recently referred to as fatigue and fatigability, respectively 2 . Over the past few years, research in our lab has focused on understanding the behavioral parameters that bring about feelings of CF, as well as the neural correlates of CF in MS. Identifying the underlying causes of CF and isolating the neural networks involved in CF will lead to more efficient clinical techniques that can be used to manage CF in MS. In this brief article we highlight some of our recent work in this area and provide a few take home practice points for clinicians. Cognitive Fatigue and Behavior CF is multifaceted and the measurement of the construct is complex 3 . Questionnaires exist that are designed to capture general fatigue; however, many scales lack items that specifically capture CF. The scales that do include CF items often ask patients to retrospectively estimate their fatigue over the past week or few weeks. A large proportion of prior research has focused on using these "trait" measures to understand CF in MS. We have taken a different approach and that is to look at "state" CF and try and identify what task demands elicit subjective feelings of CF. Taking online measurements of CF (i.e., state CF), asking participants how mentally fatigued they are at this very moment, has been valuable in trying to understand how different task demands lead to subjective feelings of CF in MS. In our earlier work, we recorded subjective trait CF ratings at four separate intervals over a 3-hour neuropsychological testing session and found that CF ratings were unrelated to behavioral performance. In fact, MS participants' neuropsychological performance improved similar to healthy controls (a practice effect), suggesting that higher fatigue does not impair performance 4 . Cognitively impaired and unimpaired MS patients also show similar CF profiles 5 and high and low fatigue periods in the same MS patients do not differ in their cognitive performance 6 . These findings suggest that subjective CF does not change as a function of cognitive impairment and that individual differences in CF do not impact cognitive performance. More importantly, this work shows that subjective CF does not correlate with objective performance, a consistent finding with over 100 years of research. In order to further identify the task parameters associated with CF, we recently asked MS participants to complete two cognitive tasks Bulletin vol. 29 no. 1 | 25 Joshua Sandry, Ph.D., Ekaterina Dobryakova, Ph.D., and John DeLuca, Ph.D. that varied in task difficulty 7 . One task relied on working memory resources (n-back) while the other task relied on processing speed resources (modified symbol digit modalities test) and each task included either a low or high cognitive demand. Participants' "online" or state CF was measured at multiple specific intervals allowing us to identify how CF manifests as a function of: [1] cognitive domain (working memory vs processing speed) [2] cognitive load (high vs low) or [3] time on task, as well as the interaction among these factors. The findings revealed that subjective CF was more pronounced as the length of the task increased and this was more extreme for MS patients compared to healthy controls. There was no effect for cognitive domain or cognitive load. Additionally, we found that trait and state measures of fatigue were uncorrelated in MS patients, suggesting that the two constructs are independent. Finally, consistent with prior research 3 , higher levels of CF did not result in worse behavioral performance, supporting the fatigue-behavior dissociation. At this point, future research needs to test whether minimizing time on task (or taking more breaks) reduces CF and whether this will have important implications for clinical practice. Neural Correlates of Cognitive Fatigue Recent advances in functional neuroimaging have helped to understand the neurocognitive and neurobiological basis of CF. Neuroimaging evidence is particularly valuable since self-report instruments (e.g. Fatigue Severity Scale or modified Fatigue Impact Scale) that are often used to assess CF show little correlation with objective measures of performance as described above. According to one hypothesis, CF might arise due to the "failure of the non-motor functions of the basal ganglia"( 8 p.40). The basal ganglia is an aggregation of subcortical nuclei, previously thought to be responsible primarily for motor function and control 9 . However, today there is ample evidence that shows that the basal ganglia plays an important role in learning, motivation, addiction and reward-guided behavior. Largely, the involvement of the basal ganglia in higher-order behaviors can be explained by widespread topographical projections from the prefrontal cortex (PFC) 10 . The hypothesis proposed by Chaudhuri and Behan 8 was developed based on evidence from animal and clinical studies that showed the effects of the basal ganglia damage to be similar to the symptoms of patients who experience CF. Functional neuroimaging studies can provide important converging evidence regarding the activity of brain areas underlying fatigue. To date, there are very few functional neuroimaging studies that investigate the neural correlates of CF in either healthy participants or in individuals with neurological damage. For example, functional magnetic resonance imaging (fMRI) studies from our lab showed that the pattern of activation in the striatum, the primary input nucleus of the basal ganglia, is different in individuals with MS and traumatic brain injury (TBI; prevalence estimates of fatigue up to 80%) compared to healthy participants 11,12 . Specifically, in both studies, we observed