Prof. Dr. Dr. h.c. mult. Niels Birbaumer
Organization: University Hospital Tübingen
Phone number: +49 (0)7071 29 74219
Department: Institute of Medical Psychology and Behavioral Neurobiology
Area: CIN Members
Scientific topic: Brain-Machine Interfaces
Field of Research
Brain-machine-interfaces (BMI) connect the brain with an external device and allow the direct transfer of intentions, thoughts and emotions from the brain to the devices. The involvement and activation of the motor system and muscles including speech is not necessary and direct communication between brain and device, mostly computers, becomes possible.
Earlier work by this group showed that patients suffering from amyotrophic lateral sclerosis (ALS) were able to learn to control non-invasively recorded brain waves through time-contingent reinforcement of these waves. After acquisition of this brain-wave-control skill patients select letters of a computer menu by activating the computer with their brain activity. However, completely locked-in patients without eye movements never learned that skill.
Recently, using a new Pavlovian conditioning paradigm we have been able to demonstrate in such patients that using their brains, even completely paralyzed patients with ALS can learn to communicate “yes” and “no” answers consisting of event-related brain potentials that are recorded with electroencephalography.
We showed for the first time that hemiparetic chronic stroke patients can learn to move a paretic arm fixed to a neuroprosthetic device by changing particular brain oscillations of the central motor system. A placebo-controlled clinical trial was initiated and completed with severely paralyzed chronic stroke patients: BCI-training was combined with behavioral physiotherapy and a comparable control group received the same BCI-training but random feedback of their brain activity plus physiotherapy. The results confirmed the hypothesis and revealed a highly significant improvement in hand movement in the experimental group only. This study paves the way to widespread clinical application of BCI in chronic stroke.
The neural circuits representing emotional and motivational behaviour are located in deep subcortical structures in the human brain and are not accessible to non-invasive neuroimaging technology, except functional magnetic resonance imaging (fMRI). Most mental disorders involve emotional changes in these deep, mostly paralimbic brain structures. Using new technology developed in our laboratory known as real-time fRMI-BCI, patients with schizophrenia, psychopathy and addictions were trained to self-regulate emotional brain structures through feedback and reinforcement. After short training durations remarkable behavioural and emotional consequences confirmed the plasticity and significance of these subcortical structures. Learned modification of large brain circuits and their connections in diverse psychological diseases is the target of future research.