When "long-term memory" no longer means "forever": analysis of accelerated long-term forgetting in a patient with temporal lobe epilepsy.

Classical amnesia involves a difficulty in transferring information to long-term memory and can be detected with standard clinical tests. However, there are some patients who pass these tests but nonetheless show longer-term memory impairments. A case study is presented of a patient, RY, with temporal lobe epilepsy, who exhibited such a profile of "accelerated long-term forgetting". To investigate the effect of recalling information on later retention, recall and recognition for pairs of novel stories were tested at five intervals ranging from 30 min to 4 weeks; we also manipulated whether or not recall and recognition were repeatedly tested for stories. Two studies are reported, one before RY commenced treatment with anticonvulsant medication, and one following 6 months of treatment. Very similar memory profiles were observed in both settings. Against a background of above average cognitive function, results showed that RY's free recall, although initially average or above, was significantly impaired at extended delays (within 24h) for non-repeatedly recalled episodic information. However, this contrasted with normal performance for information that had been repeatedly recalled. An unresolved issue in the field is the impact of anticonvulsant medication on alleviating long-term forgetting, and the current study shows that anticonvulsant medication can have negligible beneficial effects in improving the rate of long-term forgetting in this type of patient. In addition, our study highlights the possible protective effect of active review of recent episodic memories.