This project aims at establishing bidirectional communication via the cortical areas of the brain. In recent years there have been extensive research efforts for establishing an efferent pathway from the brain by means of cortical recordings to allow patients suffering from amyotrophic lateral sclerosis (ALS), stroke or high spinal cord lesions to interact with their environment (Birbaumer and Cohen, 2007; Wolpaw et al., 2002). As an extension this project will investigate the possibility of an afferent pathway to the brain by means of cortical stimulation, since it is ex-pected that stimulation might help to increase the information transfer rate for the efferent path-way. To achieve this there are two possible stimulation paradigms to be investigated. The first is based on the identification of optimal brain states for communication and the active maintenance of these states by stimulation. Inspired by classical conditioning, the second stimulation paradigm seeks to support and accelerate the rehabilitation process in stroke patients, as well as the learning process needed for the efferent communication pathway in ALS patients. By development of visual cortical prostheses (Schmidt et al., 1996) it became apparent that there are several fundamental problems related to cortical stimulation, which need to be solved before it is possible to evoke well-defined neural responses by stimulation - a prerequisite of the stimulation paradigms mentioned above. To overcome these problems it is envisaged to adapt stimulus parameters based on the current background brain activity by a feedback system in real time. Leveraging prior knowledge from microstimulation studies the feasibility of this approach will be evaluated by simultaneous stimulation and recording from ECoG grids and accompanied by the development of suitable algorithms.