Comfortably Numb p.95

The brain is the most complicated object in the universe. Nobel Prize – winning psychiatrist Eric Kandel has written, “In fact, we are only beginning to understand the simplest mental functions in biological terms; we are far from having a realist neurobiology of clinical syndromes.” Neuroscientist Torsten Wiesel, another Nobelist, scoffed at the hubris involved in naming the 1990’s “The Decade of the Brain,” by presidential proclamation. “Foolish,” he called it. “We need at least a century, maybe even a millennium” to comprehend the brain… In my travels in the neuro world, I have consistently found that the elite scientists are surprisingly modest about how little we know about the brain, despite spectacular progress in recent decades. It is the midlevel scientists who are prone to making exalted claims about the certainty and sophistication of our present knowledge.

To this day, no one knows exactly how the drugs work. The etiology of depression remains an enduring scientific mystery, with entirely new ways of understanding the disease – or diseases, as what we think of as “depression” now is probably dozens of discrete disease entities – emerging constantly. While serotonin has something to do with depression, the relationship is not a simple nor a well-understood one. No deficiencies in the serotonin system have consistently been reported among depressed people; in fact, no simple one-to-one relationship between any psychiatric disorder and a single neurotransmitter has ever been proven. While the SSRIs do indeed act on serotonin regulation in the brain, allowing the neurotransmitter to linger a little longer in the synapses, the changes that the drug ultimately exerts on the brain are entirely unclear. As an indicator of how little we know, it is striking that one of the more popular antidepressants in Europe, tianeptine, is a serotonin reuptake enhancer – it has the opposite effect of the SSRIs, allowing less serotonin to flow between the synapses. And yet it, too, can be an effective antidepressant!

The flimsiness of the entire enterprise was brought home to me in devastating fashion in a conversation with Elliot Valenstein, a leading neuroscientist at the University of Michigan and author of three highly regarded and influential books on psychopharmacology and the history of psychiatry. I was talking to Dr. Valenstein about why all psychiatric drugs address only a very small proportion of the neurotransmitters that are thought to exist. Virtually all psychiatric drugs deal with only 4 neurotransmitters: dopamine and serotonin, most commonly, and also norepinephrine, and GABA (gamma-aminobutyric acid). While no one knows exactly how many neurotransmitters there are in the human brain – indeed, even how a neurotransmitter is define exactly can be a matter of debate – there are at least 100, perhaps 125.

So I asked Dr. Valenstein, “Why do all the drugs all deal with the same brain chemicals? Is it because those four neurotransmitters are the ones understood to be most implicated with mood and thought regulation – i.e., the stuff of psychiatric disorders?”

“It’s entirely a historical accident, “ he said. “The first psychiatric drugs were stumbled upon in the dark, completely serendipitously. No one, least of all the people who discovered them, had any idea how they worked. It was only later that the science caught up and provided evidence that those drugs influence those particular neurotransmitters. After that, all subsequent drugs were ‘copycat’ of the originals – and all of them regulating only those same four neurotransmitters. There have not been any new radically different paradigms of drug action that have been developed.” Indeed, while by 1997 one hundred drugs had been designed to treat schizophrenia, all of them resembled the original, Thorazine, in their mechanism, of action.

“So,” I asked Dr. Valenstein, “if the first drugs that were discovered dealt with a different group of neurotransmitters, then all the drugs in use today would involve an entirely different set of neurotransmitters?”

“Yes”, he said.

“In other words, there are more than a hundred neurotransmitters, some of which could have vital impact on psychiatric syndromes. Yet to be explored?” I asked.

“Absolutely,” Dr. Valenstein said. “It’s all completely arbitrary.”

Indeed one of the basic tenets of biological psychiatry, that depression is a result of a deficit in serotonin (or the “monoamine theory of depression,” as it is known in the scientific literature), has proven prematurely seductive to psychiatric practitioners and patients alike. When the monoamine theory emerged in the 1960s, it gave the biologically minded practitioners of psychiatry what they had long been craving – a clean, decisive, scientific theory to help bring the field in line with the rest of medicine. For patients, too, the serotonin hypothesis was enormously appealing. It not only provided the soothing clarity of a physical explanation of their maladies, it absolves them of responsibility for their illness, and to some degree, their behavior. Because, after all, who’s responsible for a chemical imbalance?

Unfortunately, from the very start, there was a massive contradiction at the heart of the monoamine theory, whatever it is that SSRIs do to change brain chemistry, it happens almost immediately after they are ingested. The neurochemical changes are quick. However, SSRIs typically take weeks, even months, to have any therapeutic influence. Why the delay? No one had any explanation, until the late 1990s, when Ronald Duman, a researcher at Yale, showed that antidepressants actually grow brain cells in the hippocampus, a part of the brain associated with memory and mood regulation. Such a development would have been viewed as preposterous even a decade earlier; one of the central dogmas of brain science for more than a century has been that the adult brain is incapable of producing new neurons, a belief that has been disproved by Duman and a host of other well-regarded scientists. Duman believes that it takes weeks or months to build up a critical mass of the new brain cells in order to exert a healing process in the brain.

While Duman’s explanation for the mechanism of action of the SSRIs remains controversial, a consensus is building that most likely SSRIs initiate a series of complex changes, involving many neurotransmitters, that alter the functioning of the brain at the cellular and molecular levels. The emerging truth appears to be that the SSRIs may be only the necessary first step of a “cascade” of brain changes that occur long after, and well “downstream,” of serotonin alterations. The frustrating truth is that depression, and all mental illnesses, are incredible complicated and poorly understood diseases, involving many neurotransmitters, many genes, and an intricate, infinite, dialectical dance between experience and biology. One of the leading serotonin researchers, Jeffrey Meyer, of the University of Toronto, summed up the misplaced logic of the monoamine hypothesis: “There is a common misunderstanding that serotonin is low during clinical depression. It mostly comes from the fact that many antidepressants raise serotonin. This is a bit like saying pneumonia is an illness of low antibiotics because we treat pneumonia with antibiotics.” Correlation with serotonin is not necessarily causation by serotonin.

Furthermore, the monoamine system comprises only a small percent of the neurons in the brain. The largest regulatory systems in the brain are the glutamate and GABA (gamma-aminobutyric acid) systems. Glutamate excites neurons and induces activity, whereas GABA inhibits neurons.


Comfortably Numb: How Psychiatry Is Medicating a Nation p.95

The Brain and the Inner World p.168

A similar line of reasoning applies to infantile amnesia. The hippocampus is not fully functional in the first two years of life. This suggests that it is not possible for someone to encode episodic memories during this time period. Naturally, this does not imply that these early years are unimportant, or that we have no memory of the first two years of life. It implies only that the memories that we do encode during the very early years will take the form of habits and beliefs (procedural and semantic knowledge) rather than explicit, episodic memories. Infantile knowledge is stored as “bodily memory” and implicit knowledge about how the world works. We therefore have every reason to expect that early experience has a decisive impact on personality development (considering the evidence of “neuronal pruning” and the like).


Mark Solms – The Brain and the Inner World: An Introduction to the Neuroscience of the Subjective Experience p.168

The Brain and the Inner World p.164

A famous case of Claparede’s is often cited in this connection. Claparede concealed a pin in his hand when he greeted the patient, pricking her hand as he shook it. When he next attempted to greet the patient, she withdrew her hand, even though she had no conscious recollection of ever having met Claparede before. The event of the meeting had disappeared from her memory, but its effect remained. This is an example of the dissociation between episodic and procedural memory. When asked why she refused to take Claparede’s hand, the patient explained that “one has the right to withdraw one’s hand” thereby demonstrating the dissociation between episodic and semantic memory. She know what to do (procedural memory), and she recalled relevant abstract facts (semantic memory), but she was unable to bring the appropriate actual experience (episodic memory) back to mind.


Mark Solms – The Brain and the Inner World: An Introduction to the Neuroscience of the Subjective Experience p.164

The Brain and the Inner World p.132

There are also interesting clinical implications relating to modifications of this system in some children. This opioid system has been found to be overactive in some cases of autism. Consequently, such children experience far less “pain” on separation than their peers, and as a result they bond less well with caregivers and other people. Consistent with this, drugs that block the operation of opiate channels produce more positive social interactions in some cases of autism. But, importantly, the drug only appears to work (to the extent that it can) if it is combined with renewed, facilitating encouragement from the social environment. It is as if the drug opens a window, but by itself it cannot change the nature of the child’s object relationships (Panksepp, 1998).


Mark Solms – The Brain and the Inner World: An Introduction to the Neuroscience of the Subjective Experience p.132

The Brain and the Inner World p.122

The SEEKING system of a newborn baby is switched on when activated by a need, without the baby knowing what is needed. Left to its own devices, it is so helpless that it will never find the objects required to satisfy its needs and will therefore die. For this reason, it has caregivers that function as “intermediaries” between its needs (communicated by the expression of its emotions) and the objects in the outside world. The actions that these intermediaries perform on the baby’s behalf – and their effects – are then gradually learned (“internalized”) until the child can take care of itself. This, as we all know, is why parenting is so important. Early experiences of satisfaction form the templates of our understanding of how life works; for a child, learning how to adequately recognize its needs and meet them in the world is utterly bound up with the quality of parenting it receives. There are all sorts of subtle ways in which this process might be disrupted or distorted (for instance, if a baby’s needs are routinely neglected or misunderstood or even met too soon, before they can be felt). The foundations can thereby be laid for later psychopathology – in combination with a set of biological “risk factors,” such as variation in the inherent “setting” levels of the basic affective systems.


Mark Solms – The Brain and the Inner World: An Introduction to the Neuroscience of the Subjective Experience p.122


The Brain and the Inner World p.94

When one starts thinking about the problem of consciousness in the way that Damasio suggests, the question of whether or not a machine can be conscious begins to appear rather ridiculous. Some day this question might only be asked by people who are unfamiliar with the essential neuroscientific facts about consciousness. Consciousness has everything to do with being embodied, with awareness of one’s bodily state in relation to what is going on around one. Moreover, this mechanism seems to have evolved only because bodies have needs. Consciousness is therefore deeply rooted in a set of ancient biological values. These values are what feelings are, and consciousness is feeling. It is therefore very difficult to imagine how, why, and where a disembodied machine would generate consciousness. This does not rule out the possibility of an artificial system with self-monitoring properties. But the self that it monitors would have to be a body (and preferably one with a long evolutionary history) if it is really going to generate feelings.


Mark Solms – The Brain and the Inner World: An Introduction to the Neuroscience of the Subjective Experience p.94




The Brain and the Inner World p.47

David Chalmers – one of the philosophers participating in the interdisciplinary field of “cognitive science” – argues that one aspect of the mind-body problem is “easy” and the other “hard.” In this way, he divides the issue into two separate problems.

The easy problem is the one that most neuroscientists are concerned with, and it is the one discussed by Crick in his Scientific Search for the Soul. Crick attempts to solve the problem by neuroscientific means. His research strategy is to try to find the specific neural processes that are the correlates of our conscious awareness (he calls them “the neural correlates of consciousness,” or NCC for short). Finding the neural correlates of consciousness is a problem of the same general type as finding the neural correlates of anything – language or memory for instance. Neuroscience has made great progress in solving such problems in the past. Finding the brain regions and processes that correlate with consciousness is simply a matter of directing an existing research strategy from areas of previous success (language, memory) not a different aspect of mental functioning (consciousness).

We should not underestimate the difficulty of finding the neural correlates of consciousness, but Crick is only looking for which brain regions or processes correlate with consciousness and describing where they reside. He does not attempt to explain how that particular pattern of physiological events makes us conscious. This is the hard problem. The hard problem is a conundrum of a different magnitude – it raises the question of how consciousness (“you, your joys and your sorrows, your memories and your ambitions,…”) actually emerges from matter. Modern neuroscience is well equipped to solve the easy problem, but it is less clear whether is is capable of solving the hard problem. Science has few precedents for solving a problem that philosophers have deemed insoluble in principle.


Mark Solms – The Brain and the Inner World: An Introduction to the Neuroscience of the Subjective Experience p.47