Background: Fatigue is a common and disabling symptom in multiple sclerosis (MS) with a variety of direct and indirect influences, but remains poorly understood. Perceived fatigue and cognitive performance fatigability may only be weakly correlated and may have independent predictors. We adopted a multifactorial approach, utilising a measure of concurrent cognitive performance change in order to examine the clinical, psychological, and cognitive factors influencing perceived and cognitive performance fatigability in MS. Methods: Individuals with adult-onset MS were identified from a regional patient database and invited to complete an assessment battery during a home visit. Baseline perceived fatigue was measured using the Modified Fatigue Impact Scale, Fatigue Assessment Instrument, and a Visual Analogue Scale (VAS). The Conners Continuous Performance Test 3 (CCPT3) and VAS were administered before and after our intervention of roughly 2.5 hours of assessment, which represented a period of cognitive effort. The differences in scores formed measures of cognitive performance fatigability and perceived fatigue change, respectively. We examined differences across baseline fatigue, fatigue change and performance change classifications, using regression analysis to uncover predictors of perceived fatigue and performance change. Results: The sample comprised 61 participants who were recruited from an existing cohort of MS patients. Positive relationships with depression and emotion-focused coping, and a negative one with sleep, each predicted baseline perceived fatigue with the model explaining 53.5% of variance. Increased perceived fatigue change was not associated with baseline fatigue, cognitive impairment, disease variables or levels of disability, but was linked with higher anxiety, lower self-efficacy and gender. Most CCPT3 performance change variables did not show significant correlations with baseline clinical, psychological, or fatigue variables. However, two variables were predicted by positive relationships with estimated intelligence, whilst a negative relationship with self-efficacy and a positive one with post-intervention fatigue predicted one each. Conclusion: Fatigue in MS is a multifactorial construct, with perceived fatigue and cognitive performance fatigability largely influenced by indirect psychological and cognitive factors. Future studies need to take these influences into account when developing fatigue assessment tools. Further, targeting influential fatigue drivers such as psychological variables may improve the burden of fatigue and quality of life of people with MS.