Mirror neurons were first discovered in the brains of monkeys, where they are concentrated in two linked areas, called the ventral premotor cortex and the inferior parietal lobule, that are important for selecting appropriate motor behavior. Mirror neurons make up approximately 20 percent of the neurons in those regions, which lie close to the primary motor cortex, the area of the brain that sends electric signals to the muscles. In humans, however, mirror neurons may be located in many more regions of the brain, in varying amounts. (I hope to publish new findings about their location soon.)
Mirror neurons seem to have nothing in common with deliberate, effortful, and cognitive attempts to imagine being in somebody else’s shoes. So how do they actually predict the action that will follow an observed scene? How do they let us understand the intention associated with such an action?My hypothesis is this: we activate a chain of mirror neurons when we watch an action. This chain of neurons can anticipate a whole sequence—say, reaching for the cup, grasping it, bringing it to the mouth—and so can simulate the intention of the human we are watching.
Mirror neurons in such a chain may be of different types. One kind—so-called strictly congruent mirror neurons—respond to identical actions, either performed or observed. For instance, a strictly congruent mirror neuron fires both when a monkey grasps an object with two fingers, in a “precision” grip, and when that same monkey sees another primate grasping with a precision grip. A different mirror neuron, also strictly congruent, fires when the monkey grasps with its whole hand as well as when the monkey sees somebody else grasping in the same fashion.
Other mirror neurons, however, show a less strict correspondence between performed and observed actions. Those are known as broadly congruent mirror neurons. They fire at the sight of actions that may not be identical, but that achieve similar goals. For instance, a broadly congruent mirror neuron may fire when the monkey is grasping food with its hand, and also when the monkey sees somebody else bringing food to the mouth.
An important subset of the broadly congruent type of mirror neurons fire in anticipation of logically related actions. These logically related mirror neurons, as they are logically called, are probably the neuronal elements needed to understand intentions associated with observed actions. I see you grasping a cup with a certain kind of grip, and my grip mirror neurons fire, the strictly congruent ones. So far I am only simulating a grasping action. However, given that the context suggests drinking, my logically related mirror neurons, the ones that code for the action of bringing the cup to the mouth, fire even before the cup is brought to the mouth. By activating this chain of mirror neurons, my brain is able to simulate the intentions of others.
Why do some cells fire for actions that are logically related? No one knows for sure, but it’s likely that mirror neurons “learn” from experience—such as when babies watch or interact with their caregiver. Suppose a baby sees a caregiver’s hand put food on the table, and then the baby grasps the food to eat. Grasping food and seeing food placed become associated in the baby’s brain. It’s not that the mirror neurons know there is a logical connection between the two actions; rather, the baby-caregiver interaction links the two as part of a sequence.