The normal pattern of performance on list-learning tasks is to recall more words from the beginning (primacy) and end (recency) of the list. This pattern is also seen in patients with closed head injury, but malingerers tend to recall less words from the beginning of word lists, leading to a suppressed primacy effect. The present study examined this pattern on both learning trials and delayed recall of the Auditory Verbal Learning Test (AVLT) in 34 persons performing with normal effort, 38 naive malingerers, 33 warned malingerers, and 29 head-injured patients. Both malingering groups had lower scores on the primacy portion of the list during learning trials, while normals and head-injured patients had normal serial position curves. During delayed recall, normals and head-injured patients did better than the two malingering groups on middle and recency portions of the list. Findings suggest that the serial position effect during learning trials may be a useful pattern of performance to watch for when suspicious of malingering.