Deconstructing the symbol digit modalities test in multiple sclerosis: The role of memory

Background: The Symbol Digit Modalities Test (SDMT) is a sensitive measure of impaired cognition in people with MS. While the SDMT is primarily considered a test of information processing speed, other components such as visual scanning and oral-motor ability have also been linked to performance. The objective of this study was to determine the role of memory in the performance of the SDMT. Methods: Two version of a modified computerized SDMT (c-SDMT) were employed, a fixed and a variable. For each group 50 MS and 33 healthy control (HC) participants were recruited. In the fixed c-SDMT, the symbol-digit code is kept constant for the entire test whereas in the variable version, it changes eight times. Unlike the traditional SDMT which records the correct number of responses, the c-SDMT presented here measures the mean response time (in seconds) for the eight trials. Results: MS participants were slower than HC on the fixed (p < 0.001) and variable (p = 0.005) c-SDMT. Trend analysis showed performance improvement on the fixed, but not on the variable c-SDMT in both MS and HC groups. Furthermore, immediate visual memory recall was associated with the fixed (13 = 0.299, p = 0.017), but not variable (B = 0.057, p = 0.260) c-SDMT. Immediate verbal memory was not associated with either versions of the c-SDMT. Conclusions: Given that the fixed and variable c-SDMTs are identical in every way apart from the fixity of the code, the ability of participants to speed up responses over the course of the fixed version only points to the contribution of incidental visual memory in test performance.