The purpose of this study is to determine whether cognitive function evaluation (sustained attention, executive functions, working memory) during acute post-stroke is predictive of a 3 months post-stroke depression.

1. Baseline visit: after neurologic stability, 14 days or less after stroke and before discharge.

2. First assessment :

   1. Neurologic data: morphologic data (MRI of the brain), stroke severity (NIHSScore), aphasia assessment (Boston Diagnostic Aphasia Examination) if NIHS verbal score ≥1, hemineglect assessment (Bells test).
   2. Personal data assessment :

   i. Socio demographic : age, gender, marital status, employment, level of education ii. Medical history: medical treatments, psychiatric and non psychiatric history.

   c. Psychiatric assessment i. Mini International Neuropsychiatric Interview - depression ii. Beck Depression Inventory iii. Clinical Global Impression iv. Standardized Assessment of Personality - Abbreviated Scale (SAPAS) v. Fageström-C vi. Alcohol Use Disorders Identification Test (AUDIT-C) d. Cognitive assessment: i. Verbal memory : Dubois's five words ii. Clock drawing test iii. D2test (computerized) iv. Verbal fluency (Cardebat) v. Working memory (digit span WAIS 3) vi. Brixton (computerized)

3. Assessment 3 months after the first assessment. Phone call and psychiatric assessment: MINI depression, Beck Depression Inventory. Barthel index and Quality of Life assessment.

Inclusion Criteria:

• Age 18 or more
• Ischemic or hemorragic stroke, diagnosed with radiology : cerebral CT or MRI, 14 days or less after stroke
• Signed Consent
• Social insurance regime affiliation

Exclusion Criteria:

• Death
• No possibility of follow-up
• No French-speaker : insufficient command of French
• Deprivation of liberty : judicial or administrative decision
• D2 test assessment impossibility
• Aphasia with major disorder of the understanding (Boston <8)
• Antidepressant treatment during stroke
• Depression during stroke
• protected people