From Scientific American:
Eighteen years ago, in a laboratory at the University of Parma in Italy, a neuroscientist named Giacomo Rizzolatti and his graduate students were recording electrical activity from neurons in the brain of a macaque monkey. It was a typical study in neurophysiology: needle thin electrodes ran into the monkey’s head through a small window cut out of its skull; the tips of the electrodes were placed within individual neurons in a brain region called the premotor cortex. At the time, the premotor cortex was known to be involved in the planning and initiation of movements, and, just as Rizzolatti expected, when the monkey moved its arm to grab an object the electrodes signaled that premotor neurons were firing. And then, neglecting to turn off their equipment, Rizzolatti and his team got lunch.
What followed lunch that day was a serendipitous discovery. One of Rizzolatti’s graduate students decided to have an ice cream cone for dessert, which he ate in full view of the wired-up monkey. To his surprise, the electrodes suddenly began to signal a spike in cellular activity in the premotor cortex, even though the monkey was motionless.
More here.