Studies on the cognitive function have highlighted promising results concerning attention. In contrast, in the field of memory, the single cohort study that has been carried out, has not been randomised, with non comparable groups, the significant outcomes are therefore to be balanced.

In this study, patients with focal and structural temporal lobe epilepsy will be randomised in 2 groups : the COMETE group where patients will attend a specific rehabilitation programme of memory and the control group where patients will benefit from a standard treatment.

Epilepsy is a chronic neurological disease among the most frequent ones. It causes cognitive disorders in 20 to 50 percent of patients with structural epilepsy, that is to say in connexion with an identified or presumed cerebral lesion. Compared to seizures, these cognitive disorders are a major additional factor of professional, social or family disability. They are particularly prevalent (50%) in temporal lobe epilepsy and preferentially affect language memory and abilities. These disorders are well described, they are multifactorial but no (drug and/or non drug) therapy has been validated yet.

Recently, techniques of cognitive rehabilitation have shown benefits in some neurological diseases, like multiple sclerosis.

Concerning adulthood epilepsy, the few cognitive rehabilitation studies that have been conducted present huge methodological gaps limiting the scope of theirs results.

Studies on the cognitive function have highlighted promising results concerning attention. In contrast, in the field of memory, the single cohort study that has been carried out, has not been randomised, with non comparable groups, the significant outcomes are therefore to be balanced.

Thus, in this study, patients with focal and structural temporal lobe epilepsy will be randomised in 2 groups : the COMETE group where patients will attend a specific rehabilitation programme of memory and the control group where patients will benefit from a standard treatment.

Work's hypothesis would be that patients undergoing the rehabilitation programme would present improved learning abilities (with impact on their memory performances) compared to patients with standard treatment.

This hypothesis is based on underlying theories of cognitive rehabilitation. This method predicts that by reinforcing some cognitive abilities that transfer to others, the improved learning abilities will also result in improved abilities in immediate memory, work memory, but will have no effect on verbal inhibition abilities, which will constitute a specific marker/indicator ? for the intervention. Moreover, patient care may have an impact on their quality of life and mood which are predictive factors for a good social and professional integration: that's why these indexes/criteria have to be measured.

Inclusion Criteria:

• Adults with structural temporal lobe epilepsy, according to Fischer et al (2005) criteria.
• Patients aged 18-70
• Patients with epilepsy for more than 2 year
• Patients with mild to severe memory impairment (minimum score : -1,6σ / average at Buschke Reminding Test, learning index
• Patients with equal or higher Intellectual Quotient than 70 (WAIS-IV)
• stabilized anti epileptic treatment (same treatment for the last 6 months)
• Patient who has given informed consent in writing
• Patients with social insurance
• Patients with internet connexion for weekly training at home

Exclusion Criteria:

• other progressive associated neurological (history of stroke for example) or psychiatric disease
• temporal lobe surgery within following 12 months
• presence of a progressive brain lesion
• legally protected adults or adults incapable of giving informed consent
• persons who are deprived of their liberty by judicial or administrative decision