Max Planck Institute for Biological Cybernetics

Researchers test cognitive abilities of the language model GPT-3

New insights about the thalamus may improve understanding of brain disorders and intervention

What we perceive might sometimes reflect the outcome of a value-based decision-making process, a new analysis of the literature suggests

PhD student Franziska Bröker from the Max Planck Institute for Biological Cybernetics initiated the CaCTüS Internship programme for less advantaged students

The cooperation strengthens application-related research on artificial intelligence in Germany

The brain of a housefly weighs around one-thousandth of a gram. Nonetheless, thanks to this miniscule control center, the insect can evaluate images in fractions of a second and steer its way through lightning-fast flight maneuvers. It was Werner Reichardt, Founding Director of the Max Planck Institute for Biological Cybernetics in Tübingen, who, more than 50 years ago, described how the motion detectors in the fly brain work.

For Valentin Braitenberg, the brain was the most interesting research subject in the world, apart from the world itself. A former Director at the Max Planck Institute for Biological Cybernetics in Tübingen, he spent thousands of hours poring over a microscope to get to the bottom of this most complex of organs. His purpose was to examine the fiber pathways in various areas of the brain and to search for their functions.

Machines are naturally dumb. They lack flexibility and the ability to react appropriately and at the right time. Scientists are trying to teach robots something akin to intelligence.

Computer games are originally designed for entertainment. We use them to research human behavior. This opens up new approaches in psychological research on the one hand, and our results contribute to the improvement of artificially intelligent systems.

Although there is a great deal of research on vision and the processing of vision, still little is known about the processing of visual stimuli in higher brain areas. This is because an important research question has not been properly asked: the attentional selection of information. Certain brain areas coordinate where we direct our gaze; others decide what further information to retrieve and report back suggestions for interpreting the visual input. Our hypotheses set a new framework for future understanding of how vision works in our brains.

The brain is he most poorly understood organ within the human body. It is permanently in action, for example, it processes visual inputs and then rapidly decides how to interact with our environment. To achieve this, the brain consumes about 20 percent of the total energy demand of the body, and about 50 percent more than the heart. Our research team has developed a miniaturized magnetic resonance sensor (NMR) that can measure nerve activity and blood regulation with much higher spatial and temporal resolution than conventional systems.

Neuropsychiatric conditions are disorders originating in the nervous system. The past 50 years show a steady increase in neurochemical studies on rodent brain. Thereby, neuroscientists have joined the big-data club that was traditionally reserved for astronomers and physicists. By systematically analyzing data from thousands of studies, we were able to develop a free database for optimizing further research. We also demonstrated a mismatch between the current classification of neuropsychiatric drugs and spatiotemporal neurostransmitter response patterns at the systems level.

We cannot process any individual feature in a face without the other parts of the face influencing our perception. So far, this so-called holistic perception had been demonstrated mostly with static faces (images) or with objects that we know very well. However, the Max-Planck scientists have shown that dynamic faces and unknown objects (displaying specific properties) are also perceived holistically. Their results pose a challenge to current dominant theories about holistic processing.