This paper reports a qualitative review of the literature on memory for pain. Most research has focused on the accuracy of memory for pain intensity. There is some evidence that recall is moderately accurate but this conclusion is tentative because of significant methodological problems. There is also some evidence that recall of acute pain is more accurate than recall of chronic pain and we make some suggestions as to why this difference might occur. We conclude that further research on memory for pain should be informed by reference to methodological practices developed in cognitive psychology and embedded within an appropriate theoretical framework.