Princeton University
Department of Molecular Biology

Upcoming Tiger Talk

No events

Tiger Talks


How does the brain make predictions?

Tue, Apr 20, 2010
Location - TBA


Professor of Molecular Biology


Professor Michael Berry introduced the audience of 200 students and teachers to neuroscience by asking the question "Why have a brain?" He gave the example of the sea squirt that begins life as a free swimming larvae. When the sea squirt finds a location it desires, it attaches itself and becomes a sessile adult. During the metamorphosis its brain dissolves and the energy is used to develop its body. The brains of vertebrates are needed for movement, allows them to get to a desirable location, evade predators or catch prey. Prof. Berry made the case that the fundamental purpose of the brain is to make predictions. The brain is made of 1011 neurons. There is a tremendous diversity of sizes and shapes of neurons. Neurons communicate by propagating electrical signals. An electrical pulse is called an action potential. Prof. Berry went on to ask "How does the brain perform complex tasks?" Current research points to networks of neurons becoming larger and larger in bigger brains. The brain seems to operate by dividing complex tasks into steps using many local circuits. There is also evidence for functional segregation in the brain initially discovered in lesion studies. One approach, which is the focus of Prof. Berry's research, is to study local circuits in detail using the retina. The retina is equivalent to a camera. The sheet forms a pixel map of the visual field. When a ganglion fires there is an electrical output. Prof. Berry assembled an array of retinal cells on a multi-electrode array. Patterns of light are the input and the output signals are analyzed. For a single flash, there is a 70millisecond delay in response. Further experiments show that a moving bar of light produces waves of activity that indicate that the retinal array anticipates the light movement. The peak of activity is at the leading edge. The neural image travels ahead of the moving bar. Ganglion cells have large dendrites and the movement activates the leading edge. To demonstrate how the brain makes predictions, Prof. Berry called for volunteers and threw Nerf footballs to them. If the pass was smooth, then the student could anticipate the trajectory and catch the ball. If the ball bounced then it was impossible to predict where the ball would go.

more information


Lewis Thomas Lab
Washington Road
Princeton, NJ 08544

P 609.258.2075
E This email address is being protected from spambots. You need JavaScript enabled to view it.

Howard Hughes Medical Institute

Princeton Center for Quantitative Biology