The purpose of this study is to characterize specific brain signals elicited by motor disturbances and errors in stroke patients. The patients will perform a motor task using both their affected and unaffected hands. There are two types of errors: low level errors and high level errors. While disturbances (low level errors) have been shown to elicit P300, uncorrectable actions (high level errors) have been shown to elicit ERN. These event related potentials (ERP) have been extensively studied in healthy subjects including a recent paper from our Lab.

The experiment includes two parts, cognitive and motor parts, which will be conducted over different sessions in different days. Patients will fill in a consent form, and pre questionnaire form in the first session, and a post questionnaire at the end the last session.

EEG activity and eye movements will be recorded using BrainProducts EEG system. At the beginning of each session an EEG cap will be placed on the patient's head and conductive gel will be inserted between the cap and the head.

Session I - cognitive part: The cognitive parts involves the well-known oddball and Eriksen flanker tasks, which are expected to elicit P300 and ERN, respectively. Both tasks will be performed in one session unless the patient will be tired, in which case they will be performed in two sessions (in two different days). In the oddball task patients will be shown a sequence of simple geometrical shapes, including a standard blue circle, a large blue circle and a square. The probability of these shapes will be 80%, 10% and 10%, respectively. Each shape will appear for 100mse every 1500-2000msec. Patients will be asked to press the space-bar (with the unaffected hand) when the large blue circle appears. There will be 3 blocks of 400 shapes each. This task will last one hour with breaks between the blocks. In the Eriksen flanker task patients will be shown a sequence of congruent, non-congruent and neutral stimuli. Each stimulus will include a central pair of arrows, flanked at the top and bottom by two other pairs of either arrows or squares. In congruent and non-congruent stimuli the top and bottom pairs include arrows that are either in the same (congruent) or opposite (non-congruent) direction as the central pair, while in neutral stimuli the top and bottom pairs are squares. Each stimulus will appear for 200msec every 1500-2000msec. Patients will be asked to press the left or right arrow on the keyboard (with the unaffected hand) according to the direction of the arrows in the central pair as fast as possible. There will be 8 blocks of 64 stimuli each and will last 45-60min, including breaks.

Sessions II-V: Patients will sit in front of a computer screen and will play a simple computer game by moving a cursor using a Joystick, and reaching the target on the screen. Each block of reaching movements will be conducted with the same hand, and different blocks will be conducted with the affected or unaffected hands. During reaching movements with the unaffected hand there will be random jumps in the position of the target and/or cursor. The number of reaching movements in each block will be adjusted to the comfort of each patient, but will be limited to 50. A total of 250 reaching movements will be performed with each hand, over 60-120min depending on the number and duration of the breaks. The experiment will be conducted over 1-4 sessions depending on how long the patient will be able to participate in each day.

Inclusion Criteria:

• First stroke
• Rehabilitation time don't exceed one year

Exclusion Criteria:

• Mental disorders
• Additional neuromuscular disease
• Significant cognitive dysfunction (as defined by a score <20 on Mini Mental Status Examination).
• Visual impairments

Stroke patients from Loewenstein Hospital or Bnai Zion medical center who are going through a similar rehabilitation process