The human lesion method for the study of dreaming has been criticized on the grounds that patients' reports of dream cessation can be conceived as a secondary effect of memory failure, rather than as a direct consequence of neurological insults. There are two arguments within this criticism. First, dream recall is difficult in nature, even in normal people. Second, neurological patients' memories, and therefore their dream recall, are poor. This paper aims to determine whether dream recall is difficult in nature by summarizing the findings generated by the major methods utilized by researchers to study dream recall. Moreover, the controversy of dream cessation as a secondary effect of memory failure is examined, taking into consideration the clinical attributes of the major categories of neurological memory dysfunction. The analyses indicate an essential differentiation between memory failure and dream cessation.