The efficacy of a cognitive intervention consisting of training in face-name associations, spaced retrieval, and cognitive stimulation was tested in a sample of 37 patients (16 men, 21 women) with probable Alzheimer disease (AD). Patients with AD were randomly assigned to receive either the cognitive intervention or a mock (placebo) intervention for 5 weeks. The placebo group then crossed over to receive the intervention. During the intervention, AD patients showed significant improvement in recall of personal information, face-name recall, and performance on the Verbal Series Attention Test. Improvement did not generalize to additional neuropsychologic measures of dementia severity, verbal memory, visual memory, word generation, or motor speed, or to caregiver-assessed patient quality of life. Results suggest that although face-name training, spaced retrieval, and cognitive stimulation may produce small gains in learning personal information and on a measure of attention, improvement does not generalize to overall neuropsychologic functioning or patient quality of life.