The California Verbal Learning Test-Second Edition (CVLT-II) was performed by 200 participants, divided into moderate–severe traumatic brain injury (n=43), mild traumatic brain injury (n=57) and demographically matched control (n=100) groups. Participants with complicating premorbid histories or who scored below 15/16 on the CVLT-II forced choice recognition trial were excluded. There were statistically significant (p<.0001) effects of group status on the CVLT-II total recall discriminability and recognition discriminability indices. Logistic regression revealed that, in the classification of control versus moderate–severe traumatic brain injury, CVLT-II variables were accurate 66–71% overall, but false positive rates ranged from 49 to 54%. In conclusion, average scores on the CVLT-II differ meaningfully between patients with various degrees of severity of traumatic brain injury and controls, but this test should not be used in isolation to determine the presence or absence of acquired memory impairment.