Information processing speed, neural efficiency, and working memory performance in multiple sclerosis: Differential relationships with structural magnetic resonance imaging

Abstract

Multiple sclerosis (MS), a central nervous system (CNS) neurodegenerative disorder, involves lesions of both white and gray matter and reported cognitive impairments that include processing speed (PS), executive function, and working memory (WM). This study closely examined the specifics of these cognitive deficits and their relationship to structural brain damage. A visual n-back task with 3 WM load conditions was used to assess WM performance (task accuracy), PS (reaction time, RT), and a novel measure of processing efficiency (standard deviation of RT, RTSD) in MS patients and controls. These behavioral measures were related to quantitative magnetic resonance imaging (MRI) measures of white and gray matter integrity. Even when MS patients performed as well as controls, as seen for low WM load (0-back), they responded more slowly and were less efficient in their speed of responding. Accuracy findings indicated that the correct match trials were superior to correct nonmatch trials at differentiating MS patients from controls. Further, decreased accuracy during the highest WM load condition was associated with global damage that included both gray and white matter atrophy, while slowed PS and particularly processing inefficiency were associated primarily with white matter atrophy in MS. Importantly, relationships between PS, processing efficiency, performance accuracy, and structural MRI measures were seen only during the highest WM load condition, the condition that required the most executive control. These findings suggest that the MRI/behavioral relationships that were present exclusively during the 2-back condition may reflect connectivity involving frontal cortical systems, the site for executive control.

Keywords:

• Demyelination
• Neural noise
• Brain atrophy
• Executive function
• n-Back task
• Processing efficiency

Acknowledgments

Portions of this work were presented at the Society for Neurosciences Meetings in 2009. Financial support for the conduct of this research and preparation of the manuscript was from the National Multiple Sclerosis Society Grants (PP0871 to J.L.S. and PP1398 to D.W.S.).