Background: Cognitive dysfunction in multiple sclerosis (MS) can occur even in the earliest disease stages and its prevalence increases with disease duration. Whether and how the neural correlates change in parallel is incompletely understood. Objective: We therefore created a functional magnetic resonance imaging (fMRI) paradigm with applicability across a large clinical spectrum of MS aiming to elicit activation in a widespread cognitive network. Methods: We investigated 40 MS patients n=10 clinically isolated syndrome (Expanded Disability Status Scale(EDSS) score 0-2, age 23-50), n=10 relapsing-remitting MS (0-3.5, 24-49), n=10; secondary progressive MS (3.5-7.5, 25-57), n=10; long-term low disability MS (0-3.5, 33-58) and 21 healthy age-matched controls using a 3T scanner and a `Go/No Go' paradigm. Participants also underwent extensive behavioral and neuropsychological testing (including the Brief Repeatable Battery for Neuropsychological Tests, the Wisconsin Card Sorting Test, the `Zahlen-Verbindungstest', the Nine-Hole Peg Test and the Maximal Finger Tapping Rate). Results: Behaviorally, reaction times and the number of correct responses were lower in patients than controls. This was mostly attributable to a worse performance of secondary progressive MS patients. On fMRI, both patients and controls demonstrated robust activation of a frontoparietal cerebral network. Depending on task difficulty, differential recruitment of network components became evident. Also, brain activity increased with increasing cognitive demand in the pre- and supplementary motor area (SMA) in both groups. However, in contrast with healthy controls, MS patients showed additional activity in posterior parietal areas with the most difficult task. Differences in activation were more pronounced in patients with longstanding disease. Conclusions: Our results confirm previous studies on the role of pre-SMA/SMA in response inhibition. They also support the concept of increasing cortical recruitment with increasing cognitive demands which is supported by cortical reorganization in longstanding MS.