Objective: Posttraumatic stress disorder (PTSD) has been associated with neuropsychological impairments, particularly in verbal memory. It is unclear if memory impairments persist once PTSD symptoms improve, and if the memory impairments represent a risk factor for the development of PTSD or a consequence of the disorder. We hypothesized that veterans with current PTSD would show poorer memory performance compared to veterans with past PTSD and controls, and that veterans with past PTSD would perform similarly to controls. Method: We evaluated verbal memory via the California Verbal Learning Test (CVLT) and the Logical Memory subtest of the Wechsler Memory Scales III in a sample of Gulf War veterans with current PTSD (n =45), past PTSD (n = 40), and controls with no history of PTSD (n = 156). Results: There were no significant differences on verbal immediate memory between groups on either CVLT [F(2, 224) = 1.74, p = .177] or Logical Memory [F(2, 230) = .974, p = .379] after covarying for premorbid IQ, Body Mass Index (BMI) and Gulf War Illness (GWI). In contrast, on measures of delayed memory, veterans with current PTSD retained less information on both CVLT percent retention, [F(2, 222) = 3.30, p = .039], and Logical Memory percent retention [F(2, 230) = 3.63, p = .028], whereas veterans with past PTSD performed similarly to controls after covarying for premorbid IQ, BMI and GWI. Conclusion: This pattern of results shows delayed memory differences in current, but not past PTSD, suggest that memory impairments may represent a feature of current PTSD rather than a risk factor.