Prospective memory deficits are common after brain injury and can create impediments to independent living. Most approaches to management of such deficits are compensatory, such as the use of notebooks or electronic devices. While these can be effective, a restorative approach, in theory, could lead to greater generalisation of treatment. In the current study a metacognitive technique, using visual imagery, was employed under conditions of rote repetition and spaced retrieval. Treatment was provided in an AB-BA crossover design with A as the active treatment and B as a no-treatment attention control to 20 individuals with brain injury. A group of 20 healthy participants served to control for effects of re-testing. Individuals with brain injury demonstrated improvement on the main outcome measure of prospective memory, the Memory for Intentions Screening Test, only after the active treatment condition. In addition, some generalisation of treatment was measured in daily life. Moreover, treatment gains were maintained for one year after treatment was completed.