Smooth pursuit eye movements have been investigated as a diagnostic tool for mild traumatic brain injury (mTBI). However, the degree to which smooth pursuit differentiates mTBI patients from healthy controls (i.e. its diagnostic performance) is only moderate. Our goal was to establish if simultaneous performance of smooth pursuit and a working memory task increased the diagnostic performance of pursuit metrics following mTBI. We integrated an n-back task with two levels of working memory load into a pursuit target, and tested single- and dual-task pursuit in mTBI patients and healthy controls. We assessed pursuit using measures of velocity accuracy, positional accuracy and positional variability. The mTBI group had higher pursuit variability than the control group in all conditions. Performing a concurrent 1-back task decreased pursuit variability for both the mTBI and control groups. Performing a concurrent 2-back task produced differential effects between the groups: Pursuit variability was significantly decreased in the control group, but not in the mTBI group. Diagnostic indices were improved when pursuit was combined with the 2-back task, and increased by 20% for the most sensitive variable. Smooth pursuit with simultaneous working memory load may be a superior diagnostic tool for mTBI than measuring smooth pursuit alone.