Mirror neurons: the DNA of psychology?
The chances are good that you have heard of mirror neurons. They were once thought to be the harbingers of a scientific revolution, and promised to “do for psychology what DNA did for biology” (according to the prominent neuroscientist V. S. Ramachandran in an Edge essay in 2000). But the revolution didn’t happen. In fact, recent research suggests that mirror neurons have — at best — a very modest role to play in psychology.
What makes mirror neurons special is that they fire not only when an action is performed, but also when the same action is perceived. In other words: they fire not only when a monkey grasps a banana, but also when a monkey sees another monkey grasp a banana.
Giaccamo Rizzollati and his colleagues at the University of Parma immediately saw the potential of their discovery in the early 1990s. They suggested that mirror neurons could explain how humans and animals understand the actions of others in a direct way, or “from the inside”, by simulating them in the motor area of their own brain.
If true, this would have been extremely powerful. Imagine the audience’s excitement: humanity’s inherited predisposition toward mirroring, implemented at the level of the brain, conferred such a huge advantage to our ancestors that mirror neurons become universal in the human population. This would then imply that we are “hard-wired” by evolution to empathize, imitate, and even understand others. As a corollary, the discovery could also shed light on the development of mental disorders.
Most notably, Ramachandran and other neuroscientists (e.g., Lindsay Oberman and Marco Iacoboni) introduced the “broken mirrors” theory of autism. This suggested that it is dysfunction in the mirror neuron system that explains the impairments in autism. Other phenomena for which mirror neuron explanations have been proposed include self-awareness, emotion recognition, music processing, schizophrenia, multiple sclerosis, smoking addiction, and obesity. The list continues to grow (and now includes, e.g., the bedroom intruder hallucination in sleep paralysis).
Much of this is now widely known, because mirror neuron research has been widely publicized. What is less known is that subsequent research has undermined almost all of the claims made in the popular literature.
Granted, some of these claims were easy targets. The “broken mirrors” theory of autism is rather implausible to start with, for example, because autistic individuals are not particularly bad at imitation or in understanding the actions of others (which were supposed to be the core functions of mirror neurons). However, even basic ideas on the nature and origin of mirror neurons have been challenged.
As the Oxford psychologist Cecilia Heyes has extensively argued, virtually all of the evidence for mirror neurons is compatible with the hypothesis that mirroring simply results from associative learning. This alternative explanation goes roughly as follows: a baby claps its hands, and the mother responds by clapping her hands as well. As this is repeated, the baby creates an association between the visual representation of clapping and the motor representation of clapping. When it then later sees the mother clap her hands, the corresponding visual neurons are activated, but–due to the learned association–the motor neurons for clapping are also activated.
Thus, an activation network has been formed, in the usual Hebbian way, which results in motor neurons that have learned to behave as “mirror neurons”.
The difference is crucial. If the mirroring functions of neurons are not something we are born with, but rather just the neural correlates of learned associations between visual and motor representations, then it is hard to see how they could be the foundation for our ability to imitate and empathize (or explain language learning and the rise of human civilization). Instead, they may play just a modest role in explaining cognition by showing how the brain creates connections between visual and motor information.
Of course, this alternative view of mirror neurons is an empirical hypothesis that needs to be confirmed. However, so far it does indeed seem to be well-supported by empirical evidence. For example: some mirror neurons respond maximally to unnatural stimuli, such as perceiving robotic pliers or the sound of paper tearing. This would be strange if mirror neurons were hard-wired by evolution, but is perfectly logical if they are a result of learning: Associations between visual and motor representations can also be learned for artificial stimuli. Indeed, it seems to be possible to rewire connections between motor and visual areas to create new mirror responses, such as by asking participants to perform a certain movement whenever a specific visual stimulus is presented. This again is in conflict with the view that the properties of mirror neurons are just due to a genetic predisposition, and speaks in favor of the alternative learning-based theory.
One possible objection to this is neonate imitation, which is one of the cornerstones in the mirror neuron literature. The idea is that newborn babies already have the capability to imitate. Recall Andrew Meltzoff’s famous demonstration that if you stick out your tongue, even a 10-minute-old infant will stick out its tongue as well. This would seem to support the view that mirror neurons and corresponding imitation abilities are something we are born with, and not just a result of learning. However, the evidence for neonate imitation has itself always been inconclusive.
Recently, a large study (N=106) was carried out test this in the most comprehensive way so far, and no evidence for neonate imitation was found. Sure enough, when you stick out your tongue, the baby will at some point stick out its tongue as well. But it turns out that the baby also does this after perceiving other facial expressions, such as smiles and grimaces. The researchers found no significant difference whatsoever between the conditions. In other words, it does not seem that newborn babies are hard-wired to imitate; they just like to stick out their tongues (and to yawn, and blink, especially when they get some positive response).
All in all, the revolutionary claims about the importance of mirror neurons seem to be unfounded. This does not mean that mirror neurons are scientifically irrelevant. They may still play a role in explaining the neural basis for action perception and understanding. Claims like “cells that read minds” of course make better headlines than “cells that may be one part of the explanation for action understanding”. However, it should be clear by now that although mirror neurons were certainly an important discovery, they will not be the DNA of psychology.
Cook, R., Bird, G., Catmur, C., Press, C., & Heyes, C. (2014). Mirror neurons: from origin to function. Behavioral and Brain Sciences, 37(2), 177-192.
Heyes, C. (2010). Where do mirror neurons come from? Neuroscience & Biobehavioral Reviews, 34(4), 575-583.
Oostenbroek, J., Suddendorf, T., Nielsen, M., Redshaw, J., Kennedy-Costantini, S., Davis, J., … & Slaughter, V. (2016). Comprehensive longitudinal study challenges the existence of neonatal imitation in humans. Current Biology, 26(10), 1334-1338.
Ramachandran, V. S., & Oberman, L. M. (2006). Broken mirrors. Scientific American, 295(5), 62-69.