The purpose of the present study was to investigate the efficacy of a memory and metamemory training program on memory performance and metamemory judgement accuracy in adults with a closed head injury. A multiple baseline across subjects design was used with six subjects. All subjects were seen at least two years post-injury. Training included general metamemory information about the nature of memory, use of a specific memory strategy to assist verbal recall (to Preview, Question, Read, State and Test- PQRST), specific metamemory information about the strategy, and a self instruction procedure (WTSC- What is the task, Select a strategy to use, Try out strategy, Check to evaluate strategy effectiveness). During the training period all subjects recalled greater than fifty percent of paragraph ideas while using PQRST. Follow-up tests showed that five of the six subjects maintained recall levels but a gradual decrease in slope was observed over eight weeks post-training. Tests of recall, recognition and metamemory judgements on Sentence and Action Tasks were used to evaluate generalisation of training. Two subjects showed improved recall and two subjects showed improved recognition performance. In addition, four subjects demonstrated greater metamemory judgement accuracy about recognition performance following training. Improved performance post-training was also observed for three subjects on the Rivermead Behavioral Memory Test and the Logical Memory subtest of the Wechsler Memory Scale-Revised, greater than that expected for repeated testing. Several factors were identified as having a role in subjects’ ability to benefit from training.