The glutamate N-methyl-D-aspartate (NMDA) receptor antagonist ketamine displays rapid antidepressant effects in patients with treatment-resistant depression (TRD); however, the potential for adverse neurocognitive effects in this population has not received adequate study. The current study was designed to investigate the delayed neurocognitive impact of ketamine in TRD and examine baseline antidepressant response predictors in the context of a randomized controlled trial. In the current study, 62 patients (mean age = 46.2 ± 12.2) with TRD free of concomitant antidepressant medication underwent neurocognitive assessments using components of the MATRICS Consensus Cognitive Battery (MCCB) before and after a single intravenous infusion of ketamine (0.5 mg/kg) or midazolam (0.045 mg/kg). Participants were randomized to ketamine or midazolam in a 2:1 fashion under double-blind conditions and underwent depression symptom assessments at 24, 48, 72 h, and 7 days post treatment using the Montgomery-Asberg Depression Rating Scale (MADRS). Post-treatment neurocognitive assessment was conducted once at 7 days. Neurocognitive performance improved following the treatment regardless of treatment condition. There was no differential effect of treatment on neurocognitive performance and no association with antidepressant response. Slower processing speed at baseline uniquely predicted greater improvement in depression at 24 h following ketamine (t = 2.3, p = 0.027), while controlling for age, depression severity, and performance on other neurocognitive domains. In the current study, we found that ketamine was devoid of adverse neurocognitive effects at 7 days post treatment and that slower baseline processing speed was associated with greater antidepressant response. Future studies are required to further define the neurocognitive profile of ketamine in clinical samples and to identify clinically useful response moderators.