A computer model of the brain containing 2.5 million virtual neurons exhibits complex behaviours such as recognizing numerical patterns and doing simple arithmetic.

Chris Eliasmith and his team at the University of Waterloo, Canada, have developed a brain model that can 'see' images and perform tasks, writing out answers with a robotic arm. The model simulates the physiology of each of its 'neurons', such as the spikes of electricity that flow through them. The cells are divided into groups corresponding to specific parts of the brain that process images, control movements and store short-term memories. These regions are wired together, allowing the 'brain' to perform at least eight different tasks.

However, the model is limited: it is slow, simulates only a small fraction of the human brain and cannot learn completely new tasks or deal with inputs beyond the ten numerals.

Science 338, 1202–1205 (2012)