Background and Objective: We previously reported a randomized, sham-controlled trial of 5 Hz dorsolateral prefrontal left- and right-side repetitive transcranial magnetic stimulation (rTMS) in 48 participants with a medically refractory major depressive disorder. Depression improved most with right-side cranial stimulation, both rTMS and sham, and to a lesser degree with left rTMS. Because depression is often associated with cognitive impairment, in this study we sought to determine whether our earlier participants had treatment-induced changes in cognition, which cognitive domains (language, executive, visuospatial, verbal episodic memory, attention) were affected, and whether treatment-induced cognitive changes were related either to improvement in depression or to other treatment variables, such as right versus left treatment and rTMS versus sham. Methods: We used hierarchical regression analyses to determine how variables measured at baseline or associated with treatment affected changes in neuropsychological functions. The variables were neuropsychological function in the 5 domains, severity of depression, change in depression with treatment, rTMS versus sham, laterality of stimulation, and rTMS-laterality interaction. Results: Compared to sham, right rTMS was associated with 1.24 standard deviations greater gain in language function, 1.09 standard deviations greater gain in visuospatial function, and 2.38 standard deviations greater gain in verbal episodic memory than left rTMS. These improvements did not appear to be directly related to the relief from depression. Conclusions: Our results suggest that disorders of cognition and mood in depression may have different mechanisms, but right rTMS may treat both. We propose potential mechanisms underlying the right-side rTMS effect. CLINICAL TRIAL REGISTRATION: Clinicaltrials.gov NCT00711568. © 2014 by Lippincott Williams & Wilkins.