Sacred Psychiatry in Ancient Greece - Georgios Tzeferakos & Douzenis Athanasios

This open access article from the Annals of General Psychiatry offers an interesting glimpse into the role of the psychological healer in ancient Greece, and to surprise, the role that shamanism played in their healing model. Cool stuff.

Full Citation:
Tzeferakos, G, and Athanasios, D. (2014, Apr 12). 'Sacred psychiatry in ancient Greece. Annals of General Psychiatry; 13:11. doi:10.1186/1744-859X-13-11

'Sacred psychiatry in ancient Greece'

Georgios Tzeferakos and Douzenis Athanasios
Author Affiliations
Published: 12 April 2014

Abstract (provisional)

From the ancient times, there are three basic approaches for the interpretation of the different psychic phenomena: the organic, the psychological, and the sacred approach. The sacred approach forms the primordial foundation for any psychopathological development, innate to the prelogical human mind. Until the second millennium B.C., the Great Mother ruled the Universe and shamans cured the different mental disorders. But, around 1500 B.C., the predominance of the Hellenic civilization over the Pelasgic brought great changes in the theological and psychopathological fields. The Hellenes eliminated the cult of the Great Mother and worshiped Dias, a male deity, the father of gods and humans. With the Father's help and divinatory powers, the warrior-hero made diagnoses and found the right therapies for mental illness; in this way, sacerdotal psychiatry was born.

The complete article is available as a provisional PDF. The fully formatted PDF and HTML versions are in production. 

Introduction

Three basic trends in psychiatric thought can be traced back to earliest times: (a) organic approach, the attempt to explain diseases of the mind in physical terms; (b) psychological approach, the attempt to find a psychological explanation for mental disturbances; and (c) sacred or magical approach, which can be further divided into the animistic, mythological and demonological models [1]. The origin of the word ‘magic’ leads us back to the Persian religion. The prophet Zoroaster (sixth century B.C.) helped man in his struggle against evil. Aiding Zoroaster in his proselytization of the right road were the priests known as Mah (pronounced Mag), which meant ‘the greatest ones’. In subsequent years, the great Magi lost their high reputation and became known as charlatans and tricksters, hence, the connotation to the word ‘magic’ [2].
 
The magical sacred approach forms the primordial foundation for any psychopathological development because it reflects a modality of interpretation of reality that is innate to the prelogical human mind. The animistic model is based on prelogical, emotional reasoning, originating from certain historical conditions. Primitive man lived in deep communion with nature and perceived all phenomena to be connected by mysterious forces. Chance does not exist, and everything that happens has a precise meaning because the world is inhabited by animated entities that support every single event. Different feelings and emotions, psychosensorial disturbances, and delusions are the work of obscure and ineffable forces that people the world of nature and can act on a man's mind and soul [3].
 
Greek thought in the middle of the second millennium B.C. transformed the animistic conception into a naturalistic, anthropomorphic theology, in which indistinct and fluid forces were materialized in myths. Every symptom was thought to be caused by a certain deity, which could, if implored, benevolently cure it. The human passions, the emotional suffering from endopsychic conflicts, and the different psychiatric symptoms were projected and concentrated in a divine symbol. The myth was a form of knowledge that took place by symbolizing in concrete divine shapes the phenomena of nature and the complex life of the soul and mind. The ‘anthropomorphism’ of the Greek mythology, where even gods have feelings and emotions, is a historical breakthrough [4].
 
The genesis and the evolvement of the more elaborate mythological comprehension of the man and the universe are in a direct relationship with the historical dynamic process of an increasing complexity of the social structure. During the pre-Homeric era, in the Minoan and the Mycenaean societies, citizens were divided according to their financial status, to their position in the administrative hierarchy, and to their relationship with the kings. The king in these primitive societies was the sole judicial, legislative, and administrative authority. Another important criterion for the social position and integration in the different groups and ‘fratrias’, in the pre-Homeric and Homeric societies, was the genealogical lineage. Noble families claiming to have descended from mythological heroes gained social power and prestige.
 
The extensive colonization of the Mediterranean coast by the Greeks led to the emergence of new social groups. The merchants gained wealth and power and also became the bearers of new cultural, scientific, and political ideas taken from the neighboring nations. Gradually and through turbulent strife, the mainstay of the social structure in the ancient Greek world, the city-state (‘polis’) became the cradle of democracy. In the classical era, although the old noble families still held much of their power, the new wealthy aristocracy and the middle social clashes gained access to legislative, administrative, and judicial institutions. This social ‘democratization’ allowed and supported the great scientific and cultural changes that took place in this historical period.
 

Shamanism

Primitive man cured his minor troubles through various intuitive, empirical techniques. The first attempts to explain illness were equally intuitive: sometimes simple phenomena having a cause-and-effect relationship were easily understood (overeating and drinking may cause discomfort and thus purgatives will cure it); but that was not always the case. When the causes of an ailment were not obvious, primitive man ascribed them to the malignant influences of either other humans or divine beings and dealt with the former by magic or sorcery and the latter through magico-religious practices.
 
In these primitive societies, the typical witch, doctor, or shaman was a person capable of transcending into an ecstatic state, with the help of aromatic herbs, alcohol, seeds, and music. While being in ecstasy, he was able to communicate with the pathogenic spirits, drive them away, and thus cure the patient. In hunting societies, shamans acquire their healing powers from animal spirits [5]. A shaman was able to communicate with the beasts, travel through time and space, sink deep into the world of the spirits, and change his psychic and physical form.
 
When the Greeks colonized the Black Sea, during the seventh century B.C., they came in contact with shamanic rituals and beliefs. A key figure of shamanism was Pythagoras (sixth century B.C.). He was, in today's terms, a mathematician, astronomer, psychologist, psychiatrist, physician, musicologist, mystic, and philosopher. He gathered excessive wisdom by living through 10 or 20 human generations [6], and he believed in reincarnation (‘metempsychosis’). Through an indefinite cycle of psychic reincarnations, one could achieve immortality, a privilege seized only by the gods. Pythagoras could be considered the ‘father’ of Psychology since, as Porphyrios says, ‘He was the first one to define with precision the anthropocentric science, which teaches us the nature of an individual’ [7]. He was the founder of the encephalocentric doctrine which considered the brain as the seat of human consciousness, sensation, and knowledge and claimed that the psychic organ has a tripartite division, closely resembling the structural theory of Freud: (1) reason, which was the innate category of truth, (2) intelligence, which carried out the synthesis of sensory sensations, and (3) impulse, which derived from the soma. The rational part had its seat in the brain and the irrational one in the heart. Pythagoras considered the mental life to be a harmony supported by the relationship between antithetical forms: love-hate, good-bad, etc. Life itself was regulated according to this theory by opposite rhythmic movements, e.g., sleep-wakefulness, and mental symptoms originated from a disequilibrium of this basic harmony. The work of Pythagoras and Empedocles, originator of the cosmogenic theory of the four classical elements (fire, earth, air, and water), formed the basis for the humoral theory of Hippocrates [8]. Pythagoras stressed the value of group psychotherapy, medical herbs (opium for anxiety, cauliflower and scilla against depression, anis against epilepsy), and music for the treatment of emotionally ill patients [9]. On the other hand, the Pythagoreans avoided cauterizations and incisions [10]. According to Edebstein [11], the ‘Hippocratic oath’ is of Pythagorean origin because some of its main principles are the rejection of assisted suicide and abortion, the prohibition of surgical procedures, and public disclosure of medical cases. Hippocrates and his followers performed surgeries, administered drugs for abortion, and publicly discussed case reports, with direct reference to the patients' names [12].
 
The psychic immortality was a common belief between the Pythagoreans and the Orphic religious cult, which, according to Herodotus, originated from the Egyptian religion [13]. Fundamental feature of Orphism was the psychic ‘catharsis’ or cleansing, from somatic impulses and passions, through shamanic-like rituals, music, strict dietary practices, and exorcisms [14]. Psychically depressed patients were stimulated with Phrygian music, while the excited ones were sedated with the Doric tonalities. Orphic mysteries mainly practiced in Thrace descended to the Hellenic world from Northern Europe and Siberia [15].
 

Between legend and history: Melampus and Asclepius

Ancient Greek medicine was a complex practice perceived as something between myth and reality, as an expression of a magical divinatory, hieratic, and empirical technical practice. Examples of such interrelationship are the myths of Melampus, a priest-psychiatrist who allegedly lived in Argos 200 years before the Trojan War [16], and Asclepius who was considered to be the ‘god of medicine’ by the ancient Greeks [17].

Until the second millennium B.C., a female deity governed the Universe, according to the archaic Pelasgic religion. She was the Great Mother Earth, named Cybele or Selene or Hecate, who dominated man and predated other deities. She gave birth to all things, fertilized not by any male opposite but by symbolic seeds in the form of the wind, beans, or insects [18]. The Great Mother regulated the sexual and affective life and if angered could unleash malevolent influences that brought about zoopathic psychoses. The place of origin of her following is uncertain, but it is thought that she had popular followings in Thrace. Her most important sanctuary was Lagina, a theocratic city-state in which the goddess was served by eunuchs, called Dactyls [19].
 
The predominance of the Hellenic civilization, in the second millennium B.C., over the Pelasgic, modified the psychopathological interpretation. This fundamental change almost led to a civil war, with lots of killings, especially in the Delphi temple. The Hellenes eliminated the cult of Hecate and worshiped Dias, a male deity, the father of gods and humans. The warrior cult of the hero replaced that of the Mother earth. It was the hero who set himself up as a physician and priest against the forces of evil. With the Father's help and divinatory powers, he made diagnoses and found the right therapies for mental illness; in this way, sacerdotal psychiatry was born around 1500 B.C. [1].

Read the whole article by downloading the PDF from the link above.

As my regular readers well know, I don't think we will ever have human-like robots who can interact with us as though they are not machines. This article from Nautilus presents recent advances in what is known as subsymbolic approaches to AI, "Trying to get computers to behave intelligently without worrying about whether the code actually “represents” thinking at all."

A.I. Has Grown Up and Left Home

It matters only that we think, not how we think.

By David Auerbach Illustration by Olimpia Zagnoli December 19, 2013

"The history of Artificial Intelligence,” said my computer science professor on the first day of class, “is a history of failure.” This harsh judgment summed up 50 years of trying to get computers to think. Sure, they could crunch numbers a billion times faster in 2000 than they could in 1950, but computer science pioneer and genius Alan Turing had predicted in 1950 that machines would be thinking by 2000: Capable of human levels of creativity, problem solving, personality, and adaptive behavior. Maybe they wouldn’t be conscious (that question is for the philosophers), but they would have personalities and motivations, like Robbie the Robot or HAL 9000. Not only did we miss the deadline, but we don’t even seem to be close. And this is a double failure, because it also means that we don’t understand what thinking really is.

Our approach to thinking, from the early days of the computer era, focused on the question of how to represent the knowledge about which thoughts are thought, and the rules that operate on that knowledge. So when advances in technology made artificial intelligence a viable field in the 1940s and 1950s, researchers turned to formal symbolic processes. After all, it seemed easy to represent “There’s a cat on the mat” in terms of symbols and logic:

Literally translated, this reads as “there exists variable x and variable y such that x is a cat, y is a mat, and x is sitting on y.” Which is no doubt part of the puzzle. But does this get us close to understanding what it is to think that there is a cat sitting on the mat? The answer has turned out be “no,” in part because of those constants in the equation. “Cat,” “mat,” and “sitting” aren’t as simple as they seem. Stripping them of their relationship to real-world objects, and all of the complexity that entails, dooms the project of making anything resembling a human thought.

This lack of context was also the Achilles heel of the final attempted moonshot of symbolic artificial intelligence. The Cyc Project was a decades-long effort, begun in 1984, that attempted to create a general-purpose “expert system” that understood everything about the world. A team of researchers under the direction of Douglas Lenat set about manually coding a comprehensive store of general knowledge. What it boiled down to was the formal representation of millions of rules, such as “Cats have four legs” and “Richard Nixon was the 37th President of the United States.” Using formal logic, the Cyc (from “encyclopedia”) knowledge base could then draw inferences. For example, it could conclude that the author of Ulysses was less than 8 feet tall:

(implies

(writtenBy Ulysses-Book ? SPEAKER)

(equals ?SPEAKER JamesJoyce))

(isa JamesJoyce IrishCitizen)

(isa JamesJoyce Human)

(implies

(isa ?SOMEONE Human)

(maximumHeightInFeet ?SOMEONE 8)

Unfortunately, not all facts are so clear-cut. Take the statement “Cats have four legs.” Some cats have three legs, and perhaps there is some mutant cat with five legs out there. (And Cat Stevens only has two legs.) So Cyc needed a more complicated rule, like “Most cats have four legs, but some cats can have fewer due to injuries, and it’s not out of the realm of possibility that a cat could have more than four legs.” Specifying both rules and their exceptions led to a snowballing programming burden.

After more than 25 years, Cyc now contains 5 million assertions. Lenat has said that 100 million would be required before Cyc would be able to reason like a human does. No significant applications of its knowledge base currently exist, but in a sign of the times, the project in recent years has begun developing a “Terrorist Knowledge Base.” Lenat announced in 2003 that Cyc had “predicted” the anthrax mail attacks six months before they had occurred. This feat is less impressive when you consider the other predictions Cyc had made, including the possibility that Al Qaeda might bomb the Hoover Dam using trained dolphins.

Cyc, and the formal symbolic logic on which it rested, implicitly make a crucial and troublesome assumption about thinking. By gathering together in a single virtual “space” all of the information and relationships relevant to a particular thought, the symbolic approach pursues what Daniel Dennett has called a “Cartesian theater”—a kind of home for consciousness and thinking. It is in this theater that the various strands necessary for a thought are gathered, combined, and transformed in the right kinds of ways, whatever those may be. In Dennett’s words, the theater is necessary to the “view that there is a crucial finish line or boundary somewhere in the brain, marking a place where the order of arrival equals the order of ‘presentation’ in experience because what happens there is what you are conscious of.” The theater, he goes on to say, is a remnant of a mind-body dualism which most modern philosophers have sworn off, but which subtly persists in our thinking about consciousness.

The impetus to believe in something like the Cartesian theater is clear. We humans, more or less, behave like unified rational agents, with a linear style of thinking. And so, since we think of ourselves as unified, we tend to reduce ourselves not to a single body but to a single thinker, some “ghost in the machine” that animates and controls our biological body. It doesn’t have to be in the head—the Greeks put the spirit (thymos) in the chest and the breath—but it remains a single, indivisible entity, our soul living in the house of the senses and memory. Therefore, if we can be boiled to an indivisible entity, surely that entity must be contained or located somewhere.

Philosophy of mind: René Descartes’ illustration of dualism.Wikimedia Commons

This has prompted much research looking for “the area” where thought happens. Descartes hypothesized that our immortal soul interacted with our animal brain through the pineal gland. Today, studies of brain-damaged patients (as Oliver Sacks has chronicled in his books) have shown how functioning is corrupted by damage to different parts of the brain. We know facts like, language processing occurs in Broca’s area in the frontal lobe of the left hemisphere. But some patients with their Broca’s area destroyed can still understand language, due to the immense neuroplasticity of the brain. And language, in turn, is just a part of what we call “thinking.” If we can’t even pin down where the brain processes language, we are a far way from locating that mysterious entity, “consciousness.” That may be because it doesn’t exist in a spot you can point at.

Symbolic artificial intelligence, the Cartesian theater, and the shadows of mind-body dualism plagued the early decades of research into consciousness and thinking. But eventually researchers began to throw the yoke off. Around 1960, linguistics pioneer Noam Chomsky made a bold argument: Forget about meaning, forget about thinking, just focus on syntax. He claimed that linguistic syntax could be represented formally, was a computational problem, and was universal to all humans and hard-coded into every baby’s head. The process of exposure to language caused certain switches to be flipped on or off to determine what particular form the grammar would take (English, Chinese, Inuit, and so on). But the process was one of selection, not acquisition. The rules of grammar, however they were implemented, became the target of research programs around the world, supplanting a search for “the home of thought.”

Chomsky made progress by abandoning the attempt to directly explain meaning and thought. But he remained firmly inside the Cartesian camp. His theories were symbolic in nature, postulating relationships among a variety of potential vocabularies rooted in native rational faculties, and never making any predictions that proved true without exception. Modern artificial intelligence programs have gone one step further, by giving up on the idea of any form of knowledge representation. These so-called subsymbolic approaches, which also go under such names as connectionism, neural networks, and parallel distributed processing take a unique approach. Rather than going from the inside out—injecting symbolic “thoughts” into computer code and praying that the program will exhibit sufficiently human-like thinking—subsymbolic approaches proceed from the outside in: Trying to get computers to behave intelligently without worrying about whether the code actually “represents” thinking at all.

Subsymbolic approaches were pioneered in the late 1950s and 1960s, but lay fallow for years because they initially seemed to generate worse results than symbolic approaches. In 1957, Frank Rosenblatt pioneered what he called the “perceptron,” which used a re-entrant feedback algorithm in order to “train” itself to compute various logical functions correctly, and thereby “learn” in the loosest sense of the term. This approach was also called “connectionism” and gave rise to the term “neural networks,” though a perceptron is vastly simpler than an actual neuron. Rosenblatt was drawing on oddball cybernetic pioneers like Norbert Wiener, Warren McCulloch, Ross Ashby, and Grey Walter, who theorized and even experimented with homeostatic machines that sought equilibrium with their environment, such as Grey Walter’s light-seeking robotic “turtles” and Claude Shannon’s maze-running “rats.”

In 1969, Rosenblatt was met with a scathing attack by symbolic artificial intelligence advocate Marvin Minsky. The attack was so successful that subsymbolic approaches were more or less abandoned during the 1970s, a time which has been called the AI Winter. As symbolic approaches continued to flail in the 1970s and 1980s, people like Terrence Sejnowski and David Rumelhart returned to subsymbolic artificial intelligence, modeling it after learning in biological systems. They studied how simple organisms relate to their environment, and how the evolution of these organisms gradually built up increasingly complex behavior. Biology, genetics, and neuropsychology are what figured here, rather than logic and ontology.

This approach more or less abandons knowledge as a starting point. In contrast to Chomsky, a subsymbolic approach to grammar would say that grammar is determined and conditioned by environmental and organismic constraints (what psychologist Joshua Hartshorne calls “design constraints”), not by a set of hardcoded computational rules in the brain. These constraints aren’t expressed in strictly formal terms. Rather, they are looser contextual demands such as, “There must be a way for an organism to refer to itself” and “There must be a way to express a change in the environment.”

By abandoning the search for a Cartesian theater, containing a library of symbols and rules, researchers made the leap from instilling machines with data, to instilling them with knowledge. The essential truth behind subsymbolism is that language and behavior exist in relation to an environment, not in a vacuum, and they gain meaning from their usage in that environment. To use language is to use it for some purpose. To behave is to behave for some end. In this view, any attempt to generate a universal set of rules will always be riddled with exceptions, because contexts are constantly shifting. Without the drive toward concrete environmental goals, representation of knowledge in a computer is meaningless, and fruitless. It remains locked in the realm of data.

For certain classes of problems, modern subsymbolic approaches have proved far more generalizable and ubiquitous than any previous symbolic approach to the same problems. This success speaks to the advantage of not worrying about whether a computer “knows” or “understands” the problem it is working on. For example, genetic approaches represent algorithms with varying parameters as chromosomal “strings,” and “breed” successful algorithms with one another. These approaches do not improve through better understanding of the problem. All that matters is the fitness of the algorithm with respect to its environment—in other words, how the algorithm behaves. This black-box approach has yielded successful applications in everything from bioinformatics to economics, yet one can never give a concise explanation of just why the fittest algorithm is the most fit.

Neural networks are another successful subsymbolic technology, and are used for image, facial, and voice recognition applications. No representation of concepts is hardcoded into them, and the factors that they use to identify a particular subclass of images emerge from the operation of the algorithm itself. They can also be surprising: Pornographic images, for instance, are frequently identified not by the presence of particular body parts or structural features, but by the dominance of certain colors in the images.

These networks are usually “primed” with test data, so that they can refine their recognition skills on carefully selected samples. Humans are often involved in assembling this test data, in which case the learning environment is called “supervised learning.” But even the requirement for training is being left behind. Influenced by theories arguing that parts of the brain are specifically devoted to identifying particular types of visual imagery, such as faces or hands, a 2012 paper by Stanford and Google computer scientists showed some progress in getting a neural network to identify faces without priming data, among images that both did and did not contain faces. Nowhere in the programming was any explicit designation made of what constituted a “face.” The network evolved this category on its own. It did the same for “cat faces” and “human bodies” with similar success rates (about 80 percent).

While the successes behind subsymbolic artificial intelligence are impressive, there is a catch that is very nearly Faustian: The terms of success may prohibit any insight into how thinking “works,” but instead will confirm that there is no secret to be had—at least not in the way that we’ve historically conceived of it. It is increasingly clear that the Cartesian model is nothing more than a convenient abstraction, a shorthand for irreducibly complex operations that somehow (we don’t know how) give the appearance, both to ourselves and to others, of thinking. New models for artificial intelligence ask us to, in the words of philosopher Thomas Metzinger, rid ourselves of an “Ego Tunnel,” and understand that, while our sense of self dominates our thoughts, it does not dominate our brains.

Instead of locating where in our brains we have the concept of “face,” we have made a computer whose code also seems to lack the concept of “face.” Surprisingly, this approach succeeds where others have failed, giving the computer an inkling of the very idea whose explicit definition we gave up on trying to communicate. In moving out of our preconceived notion of the home of thought, we have gained in proportion not just a new level of artificial intelligence, but perhaps also a kind of self-knowledge.

David Auerbach is a writer and software engineer who lives in New York. He writes the Bitwise column for Slate.

Intranasal Ketamine for Depression - First Controlled Evidence for the Rapid Antidepressant Effects

 

This is a huge step in terms of how to make ketamine an effective treatment for depression with the risks associated with the oral method of ingestion (urinary tract issues and liver toxicity). There is already of history of intranasal ketamine for anesthesia.

Intranasal ketamine confers rapid antidepressant effect in depression

Posted By News On April 8, 2014

A research team from the Icahn School of Medicine at Mount Sinai published the first controlled evidence showing that an intranasal ketamine spray conferred an unusually rapid antidepressant effect –within 24 hours—and was well tolerated in patients with treatment-resistant major depressive disorder. This is the first study to show benefits with an intranasal formulation of ketamine. Results from the study were published online in the peer-reviewed journal Biological Psychiatry on April 2, 2014.

Of 18 patients completing two treatment days with ketamine or saline, eight met response criteria to ketamine within 24 hours versus one on saline. Ketamine proved safe with minimal dissociative effects or changes in hemodynamic dimensions.

The study randomized 20 patients with major depressive disorder to ketamine (a single 50 mg dose) or saline in a double-blind, crossover study. Change in depression severity was measured using the Montgomery-Asberg Depression Rating Scale. Secondary outcomes included the durability of response, changes in self-reports of depression, anxiety, and the proportion of responders.

"One of the primary effects of ketamine in the brain is to block the NMDA [N-methyl-d-aspartate] glutamate receptor," said James W. Murrough, MD, principal investigator of the study, and Assistant Professor of Psychiatry and Neuroscience, and Associate Director of the Mood and Anxiety Disorders Program at the Icahn School of Medicine at Mount Sinai. "There is an urgent clinical need for new treatments for depression with novel mechanisms of action. With further research and development, this could lay the groundwork for using NMDA targeted treatments for major depressive disorder."

"We found intranasal ketamine to be well tolerated with few side effects," said Kyle Lapidus, MD, PhD, Assistant Professor of Psychiatry, at the Icahn School of Medicine at Mount Sinai.

One of the most common NMDA receptor antagonists, ketamine is an FDA-approved anesthetic. It has been used in animals and humans for years. Ketamine has also been a drug of abuse and can lead to untoward psychiatric or cognitive problems when misused. In low doses, ketamine shows promise in providing rapid relief of depression, with tolerable side effects.

Study co-author Dennis S. Charney, MD, Anne and Joel Ehrenkranz Dean of the Icahn School of Medicine at Mount Sinai and President for Academic Affairs for the Mount Sinai Health System, and a world expert on the neurobiology and treatment of mood disorders, said: "What we have here is a proof of concept study and we consider the results very promising. We hope to see this line of research further developed so that we have more treatments to offer patients with severe, difficult-to-treat major depressive disorder."

Going forward, the Mount Sinai research team hopes to examine the mechanism of action, dose ranging, and use functional brain imaging to further elucidate how ketamine works.

* * * * *

The article is still in-press - this is a pre-publication version.

A Randomized Controlled Trial of Intranasal Ketamine in Major Depressive Disorder

Kyle A.B. Lapidus, Cara F. Levitch, Andrew M. Perez, Jess W. Brallier, Michael K. Parides, Laili Soleimani, Adriana Feder, Dan V. Iosifescu, Dennis S. Charney, James W. Murrough

Published Online: April 02, 2014
DOI: http://dx.doi.org/10.1016/j.biopsych.2014.03.026

Abstract

Background

The N-methyl-d-aspartate glutamate receptor antagonist ketamine, delivered via an intravenous route, has shown rapid antidepressant effects in patients with treatment-resistant depression. The current study was designed to test the safety, tolerability and efficacy of intranasal ketamine in patients with depression who had failed at least one prior antidepressant trial.

Methods

Twenty patients with major depression were randomized and 18 completed two treatment days with intranasal ketamine hydrochloride (50 mg) or saline solution in a randomized, double-blind, crossover study. The primary efficacy outcome measure was change in depression severity 24 hours following ketamine or placebo, measured using the Montgomery-Asberg Depression Rating Scale. Secondary outcomes included persistence of benefit, changes in self-reports of depression, changes in anxiety, and proportion of responders. Potential psychotomimetic, dissociative, hemodynamic, and general adverse effects associated with ketamine were also measured.

Results

Patients showed significant improvement in depressive symptoms at 24 hours following ketamine compared to placebo [t=4.39, p<0.001; estimated mean MADRS score difference of 7.6 ± 3.7 (95% CI: 3.9 – 11.3)]. Eight of 18 patients (44%) met response criteria 24 hours following ketamine administration, compared to 1 of 18 (6%) following placebo (p=0.033). Intranasal ketamine was well tolerated with minimal psychotomimetic or dissociative effects and was not associated with clinically significant changes in hemodynamic parameters.

Conclusions

This study provides the first controlled evidence for the rapid antidepressant effects of intranasal ketamine. Treatment was associated with minimal adverse effects. If replicated, these findings may lead to novel approaches to the pharmacologic treatment of patients with major depression.

What We Know Currently about Mirror Neurons - The Most [Over] Hyped Topic in Neuroscience

If you ask Christian Jarrett, Ph.D, editor of the British Psychological Society's Research Digest blog and staff writer on their magazine The Psychologist, mirror neurons are the most (over) hyped topic in neuroscience.

Back in December of 2012 he posted an entry at his Psychology Today blog entitled, "Mirror Neurons: The Most Hyped Concept in Neuroscience?" A year later, he authored an article for WIRED on the study presented below. Be sure top check out his overview of the article if the actual article proves too long or geeky for you - he offers an excellent assessment.

For your entertainment, here is Dr. Jarrett's original piece, followed by the new article published (open access) in Current Biology.

Mirror Neurons: The Most Hyped Concept in Neuroscience?

Mirror neurons are fascinating but they aren’t the answer to what makes us human

Published on December 10, 2012 by Christian Jarrett, Ph.D in Brain Myths


Back in the 1990s neuroscientists at the University of Parma identified cells in the premotor cortex of monkeys that had an unusual response pattern. They were activated when the monkeys performed a given action and, mirror-like, when they saw another individual perform that same movement. Since then, the precise function and influence of these neurons has become perhaps the most hyped topic in neuroscience.

The hype

In 2000, Vilayanur Ramachandran, the charismatic neuroscientist, made a bold prediction: “mirror neurons will do for psychology what DNA did for biology.” He's at the forefront of a frenzy of excitement that has followed these cells ever since their discovery. For many, they have come to represent all that makes us human.

Perhaps, in those early heady years, Ramachandran was just getting a little carried away? Not at all. For his 2011 book, The Tell-Tale Brain, Ramachandran took his claims further. In the chapter “The neurons that shaped civilisation”, he argues that mirror neurons underlie empathy, allow us to imitate other people, that they accelerated the evolution of the brain, that they help explain the origin of language, and most impressively of all, that they prompted the great leap forward in human culture that happened about 60,000 years ago.

“We could say mirror neurons served the same role in early hominin evolution as the Internet, Wikipedia, and blogging do today,” he concludes. “Once the cascade was set in motion, there was no turning back from the path to humanity.”

Ramachandran is not alone. Writing for The Times (London) in 2009 about our interest in the lives of celebrities, the eminent philosopher AC Grayling traced it all back to those mirror neurons. “We have a great gift for empathy,” he wrote. “This is a biologically evolved capacity, as shown by the function of ‘mirror neurons’.” In the same newspaper this year, Eva Simpson wrote on why people were so moved when Tennis champ Andy Murray broke down in tears. “Crying is like yawning,” she said, “blame mirror neurons, brain cells that make us react in the same way as someone we’re watching (emphasis added)”. In a New York Times article in 2007, about one man’s heroic actions to save another, those cells featured again: “people have ‘mirror neurons,’” Cara Buckley wrote, “which make them feel what someone else is experiencing (emphasis added)”.

If mirror neurons grant us the ability to empathise with others, it follows that attention should be drawn to these cells in attempts to explain why certain people struggle to take the perspective of others – such as can happen in autism. Lo and behold the “broken mirror hypothesis” of autism.

The reality

The ubiquitous idea that mirror neurons “cause” us to feel other people’s emotions can be traced back to the original context in which they were discovered – the motor cells in the monkey brain that responded to the sight of another person performing an action. This led to the suggestion that mirror neurons play a causal role in allowing us to understand the goals behind other people’s actions. By representing other people’s actions in the movement-pathways of our own brain, so the reasoning goes, these cells provide us with an instant simulation of their intentions – a highly effective foundation for empathy.

It’s a simple and seductive idea. What the newspaper reporters (and over-enthusiastic neuroscientists) don’t tell you is just how controversial it is. The biggest and most obvious problem for anyone advocating the idea that mirror neurons play a central role in our ability to understand other people’s actions, is that we are quite clearly capable of understanding actions that we are unable to perform.

A non-player tennis fan who’s never held a racket doesn’t sit baffled as Roger Federer swings his way to another victory. They understand fully what his aims are, even though they can’t simulate his actions with their own racket-swinging motor cells. Similarly, we understand flying, slithering, coiling and any number of other creaturely movements, even if we don’t have the necessary motor cells to simulate them. From the medical literature there are also numerous examples of comprehension surviving after damage to motor networks – people who can understand speech, though they can’t produce it; others who recognise facial expressions, though their own facial movements are compromised. Perhaps most awkward of all, there’s evidence that mirror neuron activity is greater when we view actions that are less familiar – such as a meaningless gesture – as compared with gestures that are imbued with cultural meaning, such as the victory sign.

Mirror neuron fans generally accept that action understanding is possible without corresponding mirror neuron activity, but they say mirror neurons bring an extra depth to understanding. In a journal debate published this year in Perspectives in Psychological Science, Marco Iacoboni insists mirror neurons are important for action understanding, and he quotes others saying how they allow “an understanding from within”. Critics in the field believe otherwise. Gregory Hickok at the University of California Irvine thinks the function of mirror neurons is not about understanding others’ actions per se, but about using others actions’ in the process of making our own choice of how to act. Seen this way, mirror neuron activity is just as likely a consequence of action understanding, as a cause.

What about the grand claims that mirror neurons played a central role in accelerating human social and cultural evolution by making us empathise with each other? Troublesome findings here include the fact that mirror neurons appear to acquire their properties through experience. Research by Cecelia Heyes and others has shown that learning experiences can reverse, nullify or exaggerate mirror-like properties in motor cells. It can’t reasonably be claimed that mirror neurons made us imitate and empathise with each other, if the way we choose to behave instead dictates the way our mirror neurons work. On their role in cultural evolution, Heyes says mirror neurons are affected by cultural practices, such as dancing and music, just as much they influenced them.

Finally, what about the suggestion that mirror neurons play a role in autism? It’s here that the hype is probably the least justified. There are numerous findings showing that people with autism have no problem understanding other people’s actions (contrary to the broken mirror hypothesis) and that they show normal imitation abilities and reflexes. For a new review paper, Antonia Hamilton assessed the results from 25 relevant studies, concluding: “there is little evidence for a global dysfunction of the mirror system in autism.”

--

Motor cells that respond to the sight of other people moving are intriguing, there’s no doubt. It’s likely they play a role in important social cognitions. But to claim that they make us empathic, and to raise them up as neuroscience’s holy grail, as the ultimate brain-based root of humanity, is ridiculous. The evidence I’ve mentioned is admittedly somewhat biased, designed to counteract the hype and show just how much debate and doubt persists. In fact, the very existence of mirror neurons in the human brain is still disputed by some. That’s where we’re at with the study of these cells. We’re still trying to find out whether they exist in humans, where they are, and what exactly it is they do. Mirror neurons are fascinating but they aren’t the answer to what makes us human.

Update December 2013: A new review paper has provided a sober assessment of what we currently know about mirror neurons. [This is the overview I referred to above.]

And now, your featured article.

Full Citation:

Kilner, JM, Lemon, RN. (2013, Dec 2). What We Know Currently about Mirror Neurons. Current Biology; 23(23), pR1057–R1062. DOI: http://dx.doi.org/10.1016/j.cub.2013.10.051

What We Know Currently about Mirror Neurons

J.M. Kilner, R.N. Lemon
Sobell Department of Motor Neuroscience and Movement Disorders, UCL Institute of Neurology, London, UK, WC1N 3BG

Summary

Mirror neurons were discovered over twenty years ago in the ventral premotor region F5 of the macaque monkey. Since their discovery much has been written about these neurons, both in the scientific literature and in the popular press. They have been proposed to be the neuronal substrate underlying a vast array of different functions. Indeed so much has been written about mirror neurons that last year they were referred to, rightly or wrongly, as “The most hyped concept in neuroscience”. Here we try to cut through some of this hyperbole and review what is currently known (and not known) about mirror neurons.

Introduction

Mirror neurons are a class of neuron that modulate their activity both when an individual executes a specific motor act and when they observe the same or similar act performed by another individual. They were first reported in the macaque monkey ventral premotor area F5 [1] and were named mirror neurons in a subsequent publication from the same group [2]. Ever since their discovery, there has been great interest in mirror neurons and much speculation about their possible functional role with a particular focus on their proposed role in social cognition. As Heyes [3] wrote “[mirror neurons] intrigue both specialists and non-specialists, celebrated as a ‘revolution’ in understanding social behaviour … and ‘the driving force’ behind ‘the great leap forward’ in human evolution…”. Indeed so much has been written in both peer-review literature and elsewhere about mirror neurons and their proposed functional role(s) that they have recently been given the moniker “The most hyped concept in neuroscience” [4].

For us, the discovery of mirror neurons was exciting because it has led to a new way of thinking about how we generate our own actions and how we monitor and interpret the actions of others. This discovery prompted the notion that, from a functional viewpoint, action execution and observation are closely-related processes, and indeed that our ability to interpret the actions of others requires the involvement of our own motor system.

The aim of this article is not to add to this literature on the putative functional role(s) of mirror neurons, but instead to provide a review of the studies that have directly recorded mirror neuron activity. To date, there have been over 800 published papers on mirror neurons (from a PubMed search using: “mirror neuron” OR “mirror neurons”). Here, we restrict our attention to only the primary literature on mirror neurons. Mirror neurons were originally defined as neurons which “discharged both during monkey’s active movements and when the monkey observed meaningful hand movements made by the experimenter” [2]. Thus, the key characteristics of mirror neurons are that their activity is modulated both by action execution and action observation, and that this activity shows a degree of action specificity. This distinguishes mirror neurons from other ‘motor’ or ‘sensory’ neurons whose discharge is associated with either execution or observation, but not both. It also distinguishes mirror neuron responses from other types of response to vision of objects or other non-action stimuli. As the activity of mirror neurons cannot yet be unambiguously detected using neuroimaging techniques, we have excluded human and non-human primate imaging studies from this review. We therefore focus on the 25 papers [1, 2, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27] that have reported quantitative results of recording mirror neurons or mirror-like neurons in macaque monkeys since 1992 (Table 1).

Table 1 

Proportion of neurons recorded in macaque premotor cortex (area F5) and posterior parietal cortex that showed mirror neuron properties.

Reference	Recording area	No. neurons	No. mirror	% mirror	1Action specificity	Observed effector
Bonini et al. [5]	F5	154	36	23.4%	y	Hand
Caggiano et al. [6]	F5	299	149	49.8%	n	Hand
Caggiano et al. [8]	F5	219	105	48%	n	Hand
Caggiano et al. [7]	F5	224	123	54.9%	n	Hand (video)
Caggiano et al. [9]	F5	785	247	31.5%	n	Hand (video)
Ferrari et al. [11]	F5	485	130	26.8%	y	Mouth
Ferrari et al. [12]	F5	209	52	24.9%	y	Hand
Gallese et al. [2]	F5	532	92	17.3%	y	Hand
Kohler et al. [16]2	F5	497	63	12.7%	y	Auditory
Kraskov et al. [17]	F5	64	31	48.4%	y	Hand (PTNs)
di Pellegrino et al. [1]	F5	184	18	9.8%	y	Hand
Rizzolatti et al. [18]	F5	300	60	20%	y	Hand
Rochat et al. [19]	F5	282	92	32.6%	y	Hand
Umilta et al. [23]	F5	220	103	46.8%	y	Hand
Bonini et al. [5]	IPL	120	28	23.3%	y	Hand
Fogassi et al. [13]	IPL	165	41	24.8%	y	Hand
Rozzi et al. [20]	IPL	423	51	12%	y	Hand
Shepherd et al. [21]	LIP	153	30	19.6%	n	Eye-gaze
Dushanova and Donoghue [10]	M1	303	105	34.6%	y	Reaching
Tkach et al. [22]	M1	829	581	70.1%	y	Tracking arm
Vigneswaran et al. [24]	M1	132	77	58.3%	n	Hand (PTNs)
Tkach et al. [22]	PMd	128	77	60.1%	y	Tracking arm
Ishida et al. [14]	VIP	541	48	8.9%	y	Bimodal tactile/visual
Fujii et al. [27]	PM3	148	_	3–14%4	n	Hand
	IPS5	148	-	10–42%4	n

1This column indicates if mirror neurons were tested for any form of action specificity.

2These data were further analysed by Keysers et al. [15].

3Including area F5.

4See text.

5Included anterior bank of the intraparietal sulcus (IPS). PTN, pyramidal tract neuron.

Mirror neurons were first described in the rostral division of the ventral premotor cortex (area F5) of the macaque brain, and have subsequently been reported in the inferior parietal lobule, including the lateral and ventral intraparietal areas, and in the dorsal premotor and primary motor cortex. But despite the large array of areas in which mirror neurons have been reported, the majority of mirror neuron research has studied the activity of mirror neurons in area F5 (15/25 papers; Figure 1A).

Figure 1. Number of mirror neurons recorded in areas F5 and in the IPL.

(A) The percentage of mirror neurons as a function of publication year for studies reporting mirror neurons in F5 when observing hand actions. The black line shows the line of best fit. (B) The percentage of mirror neurons in premotor area F5 and in the inferior parietal lobule (IPL). The average percentage of mirror neurons for each region is shown in black and the percentage of total mirror neurons is shown in grey with the total number of mirror neurons and neurons recorded given above.

Mirror Neurons in Ventral Premotor Region F5

Of the 15 papers reporting mirror neuron activity in area F5, 11 provide details of the number of mirror neurons recorded when observing the experimenter (not a video) reaching and grasping objects with their hand. On average, 33.6% of neurons recorded in F5 have been described as mirror neurons when the monkey observed hand actions performed by a human experimenter in front of them (ranging from 9.8–49.8%; Figure 1A,B). It is of note that the percentage of mirror neurons reported appears to increase as a function of time. This most likely reflects a sampling bias during data collection.

The first three papers [1, 2, 18] described the basic properties of mirror neurons, and their percentages are low compared with later studies. The more recent papers, in general, have investigated modulations of mirror neuron activity with some form of task manipulation. The methodological approach of these later papers is to first select neurons based on their motor properties (for example, selectivity for grasping) and then investigate the responses of this neuronal population to observed actions. This subtle change in the experimental strategy might explain the apparent increase in the percentage of mirror neurons in F5 as a function of time. Some investigators have avoided the sampling bias based on mirror properties by studying identified pyramidal tract neurons in area F5, selected on the basis of their antidromic response and not for their properties during action execution or observation [18]. A large proportion of pyramidal tract neurons in F5 and in M1 appear to show mirror-like responses (Table 1).

The three early papers [1, 2, 18] provided details about the relative selectivity of mirror neuron discharge during action execution and observation. On average, 48.9% of mirror neurons were classified as broadly congruent. Some mirror neurons discharged for only one action type, such as grasping, during both execution and observation, but showed no specificity for the type of grasp, for example precision grip or whole hand prehension. Others discharged for more than one type of observed action, for example grasping and holding. One of the three papers [2] describes a further category of mirror neurons, strictly congruent mirror neurons; these are defined as mirror neurons that respond selectively to one action type, such as precision grip, during both action execution and observation, and are reported as constituting 31.5% of mirror neurons recorded. Two of the three papers [2, 18] report a further category of neuron in F5 that discharged during action observation but not during action execution; on average these neurons, which would not be included as mirror neurons, have been reported as making up 5.1% of the neurons in F5.

Further neuroanatomical studies of area F5 have revealed three interconnected sub-divisions [28]. The sub-division in which mirror neurons are located is suggested to be on the convexity of the precentral gyrus, adjacent to the inferior limb of the arcuate sulcus, and referred to as area F5c. This is distinguished from area F5p (posterior), which is reciprocally connected both with posterior parietal area AIP and primary motor cortex M1, and from area F5a (anterior) in the depth of the sulcus, which has prefrontal connections [29].

Two studies [7, 9] have been reported that have shown that F5 mirror neurons discharged both to the observation of an action performed in front of the monkey by the experimenter and to videos of the same action. On average 26.9% of F5 neurons discharged when the monkey observed a video of a grasping action. One of the two studies [7] reported the relative number of mirror neurons that discharged to real and to videoed actions: 46.4% of neurons in F5 that responded to an executed action also responded when observing a real action, whereas only 22.3% responded when observing a videoed action. Although fewer mirror neurons responded when the monkey was observing the video of an action, for those mirror neurons that did discharge, there was no significant difference in the pattern or rate of mirror neuron discharge between real and videoed actions.

Two of the early papers [2, 18] on mirror neurons reported that they could not find any neurons that discharged when monkeys observed an object being grasped with a tool. Subsequently, two studies [12, 19] showed that mirror neurons did respond to such a tool-based action. In both these latter cases, however, the monkeys had received a high exposure to tool use during the training period prior to the recordings. One study [12] reported that 20% of F5 neurons were tool-responding mirror neurons, whereas the other reported the much higher percentage of 66.6% [20]. This high percentage most likely reflects a combination of a small sample size (n = 27) and strict inclusion criteria.

Two papers [15, 16] have reported that neurons in F5 responded to the sound of an action: so-called auditory mirror neurons. On average, 17% of F5 neurons have been reported to have auditory properties (12.7% and 21.3%, respectively, in the two papers). Four papers [6, 7, 8, 23] have reported that mirror neurons not only discharged during action observation but that their firing is further modulated by different factors: occlusion [23], relative distance of observed action [8], reward value [6] and the view point of the observed action [7]. Umilta et al. [23] showed that 19/37 mirror neurons discharged even when the observed action was occluded or hidden from the observer, demonstrating that direct vision of the action was not necessary to elicit mirror neuron discharge. Caggiano et al. [7] showed that 149/201 mirror neurons discharged preferentially for one or more of three different views of the same action (at 0, 90 and 180 degrees). Sixty of these neurons showed a preference for only one view point.

Caggiano et al. [8] also found that F5 mirror neurons have a preference for whether an observed action occurred in peripersonal or extrapersonal space: 27/105 mirror neurons discharged preferentially when the observed action occurred in the monkeys extra-personal space, whereas 28/105 mirror neurons discharged preferentially when the observed action occurred in the monkey's peri-personal space. The remaining 50 mirror neurons showed no preference. Caggiano et al. [6] reported that mirror neuron discharge is modulated by the value of the reward associated with the action: they showed that 40/87 mirror neurons responded more when a rewarded object was grasped, while 11/87 responded more when observing an action to a non-rewarded action. The remaining mirror neurons showed no preference.

One study [17] recorded from 64 neurons in F5 that were identified as pyramidal tract neurons. Thirty-one of these neurons were classified as mirror neurons, with 14/31 mirror neurons showing the ‘classic’ facilitation response during the action observation condition. Compared with baseline, the activity of the remaining 17 mirror neurons was significantly suppressed during action observation. The inclusion of these ‘suppression mirror neurons’ [8, 17, 24, 25] clearly changes the overall proportion of neurons responsive during action observation.

In a recent study, Maranesi et al. [30] compared multiunit activity responses in areas F5, F4 (premotor regions) and F1 (primary motor cortex, M1). They reported a higher proportion of recording sites showing mirror type responses in area F5 (particularly in area F5c), compared with area F4 (caudal part of the ventral premotor cortex) and with F1. In addition, they reported that in penetration sites where they identified mirror responses, they were rarely able to evoke movement using intracortical microsimulation and argued that this might be due to presence of suppression mirror neurons, as first identified by Kraskov et al. [17].

One interesting study [27] looked at activity in premotor and parietal cortex neurons of the left hemisphere of a Japanese macaque monkey, either while it observed another monkey sitting opposite making reach-to-grasp movements for food rewards, or when it performed similar actions itself. Many neurons in both cortical areas were active during the other monkey’s movements, with the proportion varying across different actions (Table 1). Premotor cortex neurons showed a distinct preference for movements involving the observed monkey’s right arm and hand, and showed a similar preference for the monkey’s own right-sided actions.

Mirror Neurons in the Inferior Parietal Lobule

Four papers [5, 13, 20, 25] have reported neuronal activity recorded in the inferior parietal lobule that the authors have described as that of mirror neurons (Figure 1B). None of these papers explicitly specifies the percentage of neurons that were classified as mirror neurons; for three of these papers, however, we were able to estimate from the numbers in the papers that the average percentage of sampled neurons that were mirror neurons was 20% (41/165 Fogassi et al. [13]; 28/120 Bonnini et al. [5]; 51/423 Rozzi et al. [20]).

Two papers [5, 13] describe the modulation of mirror neuron activity in the inferior parietal lobule by the overall goal of the observed action. Here monkeys observed an experimenter reaching for and grasping an object and either placing it in the mouth (eating) or placing it in a container (placing). On average 53% of mirror neurons had a significantly greater firing rate when the monkey observed the ‘eating’ compared with the ‘placing’ condition, 17% had a significantly greater firing rate for ‘placing’ compared with ‘eating’. The remaining 30% showed no difference between the two conditions. Yamazaki et al. [25] reported examples of mirror neuron activity in macaque area inferior parietal lobe; these neurons responded to the same action carried out in rather different contexts, suggesting that they are involved in encoding the ‘semantic equivalence’ of actions carried out by different agents in different contexts.

Rozzi et al. [20] investigated the properties of mirror neurons in the IPL. They reported that 58% of mirror neurons were responsive to only one type of hand action, for example grasping, and 25% were responsive to two different hand actions. The remaining 17% were responsive to either observed mouth actions or mouth and hand actions. Furthermore, they reported that 29% of IPL mirror neurons were strictly congruent and 54% were broadly congruent.

Mirror Neurons in the Primary Motor Cortex

The first few papers [2, 18] that described mirror neurons in area F5 also reported that the authors found no evidence of mirror activity in M1. Indeed, Gallese et al. [2] argued that, because most neurons in M1 show activity during self-movement, the absence of detectable mirror activity in M1 was evidence against the idea that this activity might actually represent monkey’s making small, covert movements while they watched the experimenter. Similarly, a recent multiunit recording study [29] found only a low level of mirror activity within primary motor cortex. However, three papers [10, 22, 24] have reported mirror neuron-like responses in M1.

Tkach et al. [22] reported that when monkeys either performed a visuomotor tracking task themselves, or watched the same target and cursor being operated by an experimenter, 70% (581/829) of recorded neurons in M1 showed stable preferred direction tuning during both execution and observation. These authors also reported that 60% (77/128) of neurons in dorsal premotor cortex were modulated in the same way.

Dushanova and Donoghue [10] recorded from neurons in M1 whilst the monkey either performed a point-to-point arm-reaching task or observed a human experimenter performing the same action. This study reported that 34.7% (105/303) of the neurons recorded in M1 were directionally tuned during both action execution and action observation. The mean firing rate during the observation condition was on average 46% of that during the execution condition. In addition, 38% of neurons retained the same directional tuning during both execution and observation conditions. It should be noted that these studies differ from those previously described that recorded from F5 and IPL.

All the studies on mirror neurons in F5 and IPL have employed tasks where the macaque monkey observed either a video or the experimenter performing simple reach and grasp actions. The two studies [10, 22] described above on mirror-like responses in M1 differed in that they used tasks in which the monkey had been extensively trained on the motor execution task. It is unclear whether the relatively high percentage of these mirror-like responses, compared with those in F5 and IPL, reflects differences between the task or real differences in the number of mirror neurons.

The final paper [24] on M1 mirror neurons recorded from 132 neurons that were identified as pyramidal tract neurons; 58% of these neurons (77/132) were classified as mirror neurons. As in F5, these authors found that these pyramidal tract neurons were either facilitation mirror neurons (58.5%) or suppression mirror neurons (41.5%) during the action observation condition. In contrast to F5, facilitation mirror neurons in M1 fired at significantly lower rates during action observation vs execution, with the former reported as “less than half of that when the monkey performed the grip”. It is noteworthy that these authors made simultaneous EMG recordings from up to 11 different arm, hand and digit muscles and confirmed complete absence of activity during action observation.
 

Mirror Neurons in Other Regions

Above, we have described the results of studies reporting mirror neurons in ventral premotor cortex, dorsal premotor cortex, primary motor cortex and inferior parietal lobule. Three further papers [14, 21, 26] have reported mirror neuron-like responses in two further areas. The first [14] recorded visuotactile bimodal neurons in the ventral intraparietal area (VIP). These are neurons that exhibit tactile receptive fields for a particular body part (for example, face or head) and also exhibit visual receptive fields in the congruent location. This study demonstrated that 48/541 bimodal neurons also exhibited visual receptive fields when observing the congruent area being touched on the experimenter. These neurons were not called mirror neurons but ‘body-matching bimodal neurons’.

Shepherd et al. [21] reported mirror neuron-like responses in the lateral intraparietal (LIP) area. These authors reported that 30/153 neurons in LIP responded not only when monkeys oriented attention towards the receptive field of those neurons, but also when they observed other monkeys orienting in the same direction.

Yoshida et al. [26] recently recorded from neurons in the medial frontal cortex, some of which selectively responded to self or observed actions within a social context. The neurons were recorded in one of two monkeys who, on alternate trials, chose a movement in order to earn a reward. Correct (or incorrect) choices rewarded (or punished: no reward) both monkeys. ‘Partner-type’ neurons were selectively responsive to the choices made by the other monkey, signalling the correct or incorrect choice made; interestingly around 19% of these ‘partner neurons’ showed decreased activity during self-movement.

Relating Human Neuroimaging Data to Mirror Neuron Activity

Of the over 800 papers returned when searching PubMed for ‘mirror neurons’ or ‘mirror neuron’, the vast majority report the results of experiments on human subjects. Of these, the results of human neuroimaging experiments, specifically fMRI [31], confirm a broad overlap between cortical areas active in humans during action observation and areas where mirror neurons have been reported in macaque monkeys (see above). Thus, changes in the BOLD signal during action observation seem to be consistent with the existence of a mirror neuron system in humans, but they cannot yet furnish conclusive proof. There has, however, also been a report of single neuron activity recorded from human neurosurgical patients that has demonstrated mirror neuron activity [32]. Recordings were focused on medial frontal cortex and temporal lobe structures, and show the extensive nature of the mirror neuron system. Unfortunately, neither of the premotor or posterior parietal areas so heavily investigated in monkeys were available for study in these patients.

Central to being able to interpret human fMRI studies of the mirror neuron system is understanding the relationship between the BOLD signal in human and mirror neuron activity in macaque monkey. To this end, monkey fMRI studies have now demonstrated significant activity during action observation in regions where mirror neurons have been previously reported [33, 34]. These monkey imaging studies have taken advantage of enhancing the neurovascular responses with an iron-based (MION) contrast agent. As with the vast majority of human fMRI studies, however, there is difficulty in relating these results to mirror neurons, in that they only employ an action observation condition and have no action execution condition. This makes it difficult to calibrate the activity changes in observation to those in execution, and also raises the possibility that sensory responses other than mirror responses contribute to the neurovascular changes (see Introduction).

One possible way of attributing the fMRI response to a single neuronal population, such as mirror neurons, is to use fMRI adaptation, or repetition suppression. This is a neuroimaging tool that has been adopted to identify neural populations that encode a particular stimulus feature [35]. The logic behind fMRI adaptation is that neurons decrease their firing rate with repeated presentations of the stimulus feature that those neurons encode. By extension it has been argued that the BOLD signal will also decrease with repeated presentations. It has been argued that areas of the cortex that contain mirror neurons should show fMRI adaptation both when an action is executed and subsequently observed, and when an action is observed and subsequently executed. This is because the stimulus feature encoded in mirror neurons is repeated irrespective of whether the action is observed or executed [36].

The results of such studies have produced mixed results. Of the five studies using this technique published to date [36, 37, 38, 39, 40], only three have demonstrated significant fMRI adaptation consistent with the presence of mirror neurons in the human brain [38, 39, 40]. One possible explanation for the mixed results is that humans do have mirror neurons, but that they do not alter their pattern of activation when stimuli that evoke their response are repeated. Indeed a recent study [9] has shown some evidence that mirror neurons may not alter their firing rate during repetitions of the same action; however, in this work the neuronal activity represented in the local field potential (LFP) did modulate with repetition. Further work is clearly required to determine why the BOLD signal in humans and the LFP in monkeys do adapt with repetition, while the evidence to date suggests that mirror neurons may not.

Great care must be taken when comparing the results from human and monkey studies. Specifically, readers must pay careful attention to the difference in the level of inference between the different modalities. The majority of human neuroimaging studies report significant results at the population level where the variance is estimated across subjects. This is in contrast to the studies reporting mirror neurons in macaque monkeys, where the aim is to test whether individual neurons show a consistent modulation of firing rate during periods of action observation and execution. Here the inference is closer to the analysis of fMRI at the single subject level. Therefore, when it is reported that 30% of neurons in any region were significantly modulated during both action observation and execution this does not mean that the remaining 70% do not modulate at all. Rather, it means there was not sufficient statistical evidence that these neurons displayed mirror activity. Indeed it is quite possible that when tested at the population level, the neurons that are non-significant at the single neuron level could be significantly modulated when observing an action.

The point here is that care must be taken when arguing that ‘only’ X% of neurons in any brain region are mirror neurons. The ‘only’ implies that the remaining neurons are not significantly modulated in any way during action observation. This is not a valid inference as to do so would be to accept the null hypothesis. This may be particularly problematic for cortical regions where responses in individual mirror neurons are relatively weak, such as in M1.

It is often assumed that mirror neuron activity during action observation is driven, bottom-up, by the visual (or auditory) input. The review of mirror neuron discharge presented here provides evidence that this is, at best, an incomplete description of mirror neuron firing. We now know that mirror neuron firing rates are modulated by view point [7], value [6] and that they even discharge in the absence of any visual input [23]. This suggests that mirror neurons can be driven or modulated top-down by backward connections from other neuronal populations. Indeed, the requirement for such top-down input to regions containing mirror neurons was realized by Jacob and Jeannerod [41], who argued that it was impossible for a mirror neuron system driven uniquely by the visual input to correctly infer an intention from an observed action if two or more different intentions would generate the same action. The fact that mirror neurons can be driven by backward connections is consistent with recent predictive coding models of mirror neuron function [42, 43, 44]. Within this framework, mirror neurons discharge during action observation not because they are driven by the visual input but because they are part of a generative model that is predicting the sensory input. This framework provides a theoretical account of mirror neuron activity that resolves the one-to-many mapping problem described by Jacob and Jeannerod [41] and is consistent with top-down modulation of mirror neuron firing rates.

Concluding Remarks

The discovery of mirror neurons has had a profound effect on the field of social cognition. Here we have reviewed what is currently known about mirror neurons in the different cortical areas in which they have been described. There is now evidence that mirror neurons are present throughout the motor system, including ventral and dorsal premotor cortices and primary motor cortex, as well as being present in different regions of the parietal cortex. The functional role(s) of mirror neurons and whether mirror neurons arise as a result of a functional adaptation and/or of associative learning during development are important questions that still remain to be solved. In answering these questions we will need to know more about the connectivity of mirror neurons and their comparative biology across different species.

Acknowledgements

J.K. and R.N.L. were both funded by the Wellcome Trust, London, UK. We would like to thank Alexander Kraskov for helpful comments on an earlier version.


References are available at the Current Biology page.