Mental Sickness


Henrik Walter

Research Division of Mind and Brain, Department of Psychiatry and Psychotherapy, Charité Universitaetsmedizin Berlin, Berlin, Germany

2013 Sep 5.


As a first approximation, we can say that it ties psychiatry closely to the biology of the brain.

Under such a broad characterization today nearly everyone would qualify as a biological psychiatrist, as only very few would deny such a connection. However, there are stronger and more controversial claims, for example the ontological claim that psychiatric disorders are disorders of the brain, or, on the therapeutic level, that the best therapies are biological ones like medication or deep brain stimulation. However, many biological psychiatrists would not share these stronger claims, so this characterization seems too narrow.

To better understand the characteristics of the third wave, it will be helpful to take a short look at the first and second wave in the history of psychiatry. The first wave in the second half of the nineteenth century can be best understood as a new research agenda. It was not so much characterized by the idea that the mental and the nervous system are closely linked – this was already believed by ancient philosophers – but rather by the ambition to uncover the relation between mind and brain by doing systematic research linking neuropathology and mental disorder and by using the experimental method in animals and humans. Wilhelm Griesinger (1817–1868), one of the most important figures of this first wave, famously declared: mental disorders are disorders of the brain. Note, that this was not primarily intended as a reductionist claim, but rather as a statement intended to delineate his ideas against the two prevailing approaches of that time: the moral approach on the one hand, and the somatic approach, linking mental disorder to body processes in the lung, liver or other organs, on the other hand. Nevertheless, Griesingers claim was not at all uncontroversial as theorists felt that such a brain approach would not do justice to the intricate psychopathological phenomena psychiatrists dealt with. For example, Karl Jaspers, the philosopher-psychiatrist, called 1913 the localationist models of two main protagonists of the first wave, Theodor Meynert and Carl Wernicke, “brain mythologies.”

In the early twentieth century, there was a decline in the biological approaches through various developments. Emil Kraeplin, one of the most influential psychiatrists at his time, started as an opponent to biological psychiatry, and developed his diagnostic system on systematic observations of symptoms and course of mental disorders, laying the groundwork for the later DSM. Also, psychological models, inspired by psychoanalysis and behaviorism became increasingly fashionable and had a large impact on therapy.

The second wave of biological psychiatry started only in the second half of the twentieth century and was, according to Shorter, driven by two new discoveries. The first was genetics, which could show that severe mental disorders, in particular schizophrenia, have a strong genetic component. The second was the discovery of efficient medication for various mental disorders (1949 lithium, 1952 chlorpromazin, 1957 imipramin, 1958 haloperidol, 1963 diazepam). They quickly became a major pillar of psychiatric treatment and contributed strongly to the opening and later disappearance of the large mental asylums in the second half of the last century. Soon, the concept of a neurochemical imbalance of neurotransmitters became the favored explanatory model for psychiatric disorders. Interestingly, at the same time as psychiatry for the first time used effective medications, the movement of antipsychiatry emerged. It was part of a more general political protest against tradition starting in the 1960s and declared “mental illness as a myth” (Szasz, 1961). It also was quite effective in discrediting one of the most effective treatments for severe depression, electroconvulsive therapy, supported among other things by the impressive movie “One flew over the cockoo’s nest” (1975) by Milos Forman. So although the second wave was in effect quite successful there was always some opposition against it on one hand, but on the other hand those insights and practices that were helpful for patients are now integrated into daily practice.

So what is the third wave of biological psychiatry?

I want to suggest that this wave has started in the last two decades of the twentieth century and is now in full progress. Again, it has been driven by methodological and technological progress. Since the declaration of the last decade of the twentieth century as the decade of the brain by the president of the United States, neuroscience has developed into one of the largest research programs worldwide. According to my view, there were two developments particularly relevant in the transition of the second wave into the third wave. The first is the progress in the molecular neurosciences. The journal Molecular Psychiatry, founded in 1997, is now one of the fields most prestigious and most cited journals. It became increasingly clear that the effects of psychiatric drugs are not primarily exerted via the level of neurotransmitters in the synaptic cleft, but that there is up- and down-regulation of receptors, effects on intracellular cascades, and even regrowth of neurons in the hippocampus. The picture of the neurobiological changes underlying psychiatric disorders and treatment thus became much more complex and differentiated and it became apparent that different levels of brain organization are important which interact in a complex way. The second development was the birth of cognitive neuroscience and neuroimaging. This field studies information processing in the brain by combining the methods of experimental psychology with tools to record brain activity or to stimulate the brain. In fact, neuroimaging, in particular functional magnetic resonance imaging (fMRI) has contributed much to public “brain awareness,” by (although wrongly) suggesting that we can literally watch “the brain at work.” With the first human study published in 1991, fMRI has today become a major research tool in psychology as well as in psychiatry. This development could not have taken place without a large increase in computational power. In fact, computational neuroscience which tries to develop mathematical models of brain function, has become an important tool in explaining neurocognitive processes and recently the program of computational psychiatry has begun to evolve (Montague et al., 2012). Further methods and technologies have become available to investigate the interplay of genetics, experience and environment in the etiology and neural explanation of psychiatric disorders like imaging genetics, epigenetics, optogenetics, or deep brain stimulation. Also big science, combining large – omic datasets like the (epi)genom, metabolom, proteome, or connectom with clinical data is becoming more important in psychiatric research and allows for new ways of discovery. The underlying model is that of systems medicine, understood as an interdisciplinary field of study that looks at the dynamic systems of the human body as part of an integrated whole, incorporating biochemical, physiological, and environmental interactions that sustain organismic life. In brain science, the paradigm of localationist thinking is substituted increasingly by thinking in functional systems and brain connectivity patterns (Buckholtz and Meyer-Lindenberg, 2012).

At this moment, we are at a critical stage of the third wave. In fact, progress in the first decade of this century has been so impressive that researchers as well as media have been overenthusiastic with regard to the power of the new methods. In particular neuroimaging results, probably due to their seemingly simple and straightforward presentation, have ignited the imagination of researchers, lay people and the media. Results are reported, similar to genetic results, in a oversimplified causal language (“love is in the ACC,” “the God spot,” “gene for schizophrenia discovered,” etc.). Such oversimplified messages are well for drawing attention to headlines, but way over what really can be inferred from most studies. Consequently, neuroscience has recently been criticized for its overambitious claims, and the field of “critical neuroscience” has flourished in the last 5 years immensely with an increasing number of books, papers and blogs (for a respectable example compare Slaby and Choudhury, 2011). Actually, in neuroscience in general, as well as in cognitive neurosciences and neuroimaging in particular self-critical articles concerning methods have begun to be increasingly published (e.g., Kriegeskorte et al., 2009; Button et al., 2012) which is a healthy self-correcting development.

According to the third wave of biological psychiatry, mental disorders are relatively stable prototypical, dysfunctional patterns of experience and behavior that can be explained by dysfunctional neural systems at various levels. As with any understanding of disease in general the notion of a “dysfunction” inevitable involves normative judgments of what is regarded as normal, functional, healthy on the one hand, and as abnormal, dysfunctional, pathological on the other hand. Further below I will come back to normative issues. But before I do so, let’s look at the concept of mental disorder within biological psychiatry.


By now the general approach or framework of the third wave of biological psychiatry should have become clear. It is focusing on a research-inspired, multi-level approach to understand what psychiatric disorders are, what mechanisms underly signs and symptoms and how an understanding of those mechanisms might help in classification, diagnosis, prognosis, and treatment. Note, that the approach does not entail the claim that biological approaches in a narrow sense are the best therapeutic approaches. It is as such neutral to the question what intervention will prove best to treat whatever there is. For example it may very well be that psychotherapeutic approaches will emerge as the best way to treat certain types of disorders. In fact, psychotherapists see no general problem in integrating their approach into such a framework as psychological mechanisms and principles that are effective in psychotherapy can be conceptualized as part of cognitive neuroscience itself (Walter et al., 2009; Disner et al., 2011). Also, the role of psychosocial and cultural factors can be integrated effortlessly as the MPC approach by Kendler et al. (2011a),b makes clear: if social factors or societal and cultural mechanisms are part of the causal machinery that contributes to the instantiation of typical clusters of signs and symptoms that characterize psychiatric disorders they are part of the underlying explanatory structure.

However, probably many or at least some people will still view this approach skeptically. Indeed, there are several problems and limitations. To name just four of them: first, it could still be argued that the framework favors the neurobiological over other factors, as it entails the idea that psychiatric disorders are brain disorders. It will make no difference if you call psychiatric disorders “disorders of the brain” or “disorders of brain circuits” and thus do not justice to the mental within the concept of mental disorders. Second, the third wave does not include a solution to the normativity problem, namely the question when a constellation of psychological signs and symptoms is already a disorder or when it is still part of “normal experience,” so it will still promote a medicalization of life problems. Third, even if we somehow could solve the first two problems, it might be argued that a focus on the brain will lead to inefficient resource allocation because the outcome for patients is not worth the effort be put in. History has shown that all general claims that we will in the near future know “the” causes of mental disorders have failed, and the continuous failure of neurobiology (with some exceptions) to sufficiently explain or predict mental disorders shows that it cannot account for such complex phenomena. Therefore, we should rather focus on the well-known psychosocial factors contributing to the development or sustainment of psychiatric disorders which are much more relevant in practice.

A recent critique of the thesis that “addiction is a brain disease” can be interpreted as a condensed combination of these worries. It argues that addiction would only be a brain disease if it has (i) neural correlates, (ii) these correlates are pathological and (iii) that pathology is sufficient for the person to have a disease, in almost any accessible environment Levy, 2013. As addicts are able to quit in certain environments, addiction would not qualify as a brain disease. This is a very clever argument as it uses one feature of the multilevel approach, namely the role of environmental factors, to argue against the “disorder of brain circuits thesis.” Indeed, there is a grain of truth in this argument, but only insofar as it helps to distinguish “organic or neurological” from “mental or psychiatric” disorders. For example, neurodegenerative diseases like M. Huntington or Alzheimer will progress in almost any environment, whereas drinking might stop. However, there are two problems with this argument: first of all, it confuses behavior (drinking) with the disorder (alcohol addiction). It is well known that people suffering from alcohol addiction who manage to quit, still are addicted life-long and have a high propensity for relapse – exactly this might be explained by the brain disorder thesis. Secondly, the argument puts the stake much too high. Using the same kind of argument it could be argued that phenylketonuria, a genetically transmitted severe metabolic disorder is not really a metabolic disorder as it can be effectively treated by a diet, i.e., the pathology is not sufficient for a person to have a disease in almost any accessible environment.

Finally, some may argue, that also the third wave of biological psychiatry, like the preceding waves, will tend to devalue an approach to psychiatry that focues on the personal level. For example, the concept of MPC is based on the idea that regards minds as brains and brains as kind of machines that are causally effected by different levels. This approach, so the argument may go, ignores the personal level even if it may pay lip service to the subjective by for example including “subjective reports” in the RDoC grid.

There are several ways to response to these critiques from within, some of which I will mention here. First, admittedly, there is a common misunderstanding on the role of neurobiological findings in psychiatric disorders. Very often, it is either said, implicitly assumed, or implied that the mere fact that there is a neurobiological correlate of a mental dysfunction is already a proof that the “causes” of the respective disorder are biological in the same way as for neurological disorders. But this clearly is a misconception. Because every mental state has a correlate in the brain, we should be able to find at least in principle neurobiological correlates of any mental state, pathological or not. So the question is not, whether there is a neurocognitive correlate or mechanism, but whether it is pathological, how it came into being, whether it is persistent, whether and how it can be influenced, and so forth. In fact, the neurobiological misunderstanding even goes further in many cases as often it is wrongly concluded that the existence of a “brain signature” (to use a more neutral term) would already imply that the disorder cannot be controlled or changed by psychological means, or even that it is inborn or genetically caused, implications which clearly are non-sequitures, but widely believed.

Second, the normative problem indeed has to be addressed – not only by biological psychiatry, but also by any other approach to psychiatry, and not only for psychiatric but also for all concepts of disorders – and consequently it has been discussed in medicine in general. As it is in no way specific for psychiatry, let alone biological psychiatry, I will not discuss it here in detail but just make some remarks. It is clear that the sheer discovery of neural correlates or mechanisms of a disorder cannot prove a state as pathologically. This can be done only by spelling out a concept of normal functioning. If a biological approach claims to be able to define mental disorders without reference to norms it must fail. Normativity in the context of mental disorders comes at least in three guises, “statistical,” “biological design” or “value-preference laden” (Graham, 2013, p. 59). For example, most definitions of mental disorders include a criterion of suffering or of clinical relevance, that only can be spelled out with respect to a norm that cannot be read simply from biological facts. I will return to this issue later. Although it has to be admitted that the third wave of biological psychiatry does not take a specific stance to the normativity problem, it should be noted that this can be only used as a critique against variants of biological psychiatry that explicitly claim that normality can be inferred simply from biological measures.

Thirdly, why has neurobiology failed to deliver better results for explanation, diagnosis, prognosis or treatment? Some answers relating to methodological problems have been already discussed above (Kapur et al., 2012). However, a further explanation for only modest progress is often not mentioned. These are the ethical constraints under which biological psychiatric research has to operate which does make progress difficult. In contrast to other medical disciplines psychiatric research can access the “organ of the mind,” the brain, only indirectly. There is no known ethically justifiable way to directly access brain tissue to investigate assumed molecular mechanisms. In contrast, the heart, the liver, the kidney and many other organs can be accessed directly in therapy and research by taking biopsies or measuring metabolites in the blood. There are only a few exceptions to this barrier, for example the possibility to measure certain molecules non-invasively with magnetic resonance spectroscopy, or with research windows related to invasive therapeutic procedures in epilepsy surgery or deep brain stimulation. Direct access to the brain in animal research also has its problems, because rodents and humans differ in many respects and animal experiments are confronted with ethical issues, too. So the “failure” of biological psychiatry is not necessarily related to its concepts or theoretical approach, but partly may be explained by important and relevant ethical barriers we have implemented in human research for good reasons.

Fourthly, does a biological psychiatry approach imply disrespect for persons? First note, that this critic in its most general and radical form is not confined to biological psychiatry but to any psychiatric approach that claims that there are mental disorders in the first place. This antipsychiatric argument claims that mental illness in general is a myth by confusing sickness with life difficulties and by stigmatizing people with mental problems as having a disorder and thus not giving them the credit and responsibility for what they do and chose to be. In a more specific and much less radical, but more frequent variant (not claiming the non-reality of mental disorders) a biological approach of psychiatry is accused of resulting in an overenthusiastic reliance on medication and an insufficient use of understanding the life stories and real-world problems of patients. Without doubt, overmedication is a problem in certain strands of psychiatry and admittedly this may be due to the fact of an oversimplified picture of mental disorders (“For depression you need to substitute serotonine like insuline in diabetes”). However, many of these implications are not inherent to the concepts of the third wave of biological psychiatry but rather are based on older conceptions that postulated a close connection between etiology and therapy, that has been abandoned today in current practice. For depression for example there was a distinction between endogenous depression (from within, medication, no talking cure), neurotic depression (originating in childhood, talking cure, no medication) and reactive depression (understandable reaction after a life event).

The aforementioned responses to a critique to biological psychiatry were given from within psychiatry and psychiatric research in itself. Many of these issues revolve about the “disorder” part of mental disorders. However, I think that a more comprehensive way of assessing the prospects of biological psychiatry can only be found when we turn to the “mental” part of a theory of mental disorders. In order to do so we can turn to a rich resource that has reflected on the concept of the mental for a long time: philosophy of mind.


If we want to understand what mental disorders are then we should take the question what “the mental” is more seriously. Traditionally, there has been a close link between philosophy in general and psychiatric theorizing. Here, I will restrict myself to recent philosophy of mind approaches, as they are targeting similar problems as biological psychiatry: what is the connection between mind and brain? The idea behind consulting philosophy is simple: if we better understand how mental states are related to brain states we might better understand how disordered mental states relate to disordered brain states. Take for example the thesis of identity theory that assumes that mental states are identical with brain states. If this is true, it seems to follow straightforwardly that disordered mental states simply are disordered brain states. Or take the problem of reductionism and mental causation: if we were really able to show that mental states can be reduced to brain states, this would leave us with only two possibilities: either we have to eliminate mental states, because they are nothing more than a convenient, folk psychological way to talk about hidden brain states or we have to conclude that mental states are epiphenomenal, i.e., have no causal powers. This seems like a conclusion only few people would like to embrace. Or take the idea of dualism. Do we have to assume a special substance that does all the work in explaining mentality that is in a separate ontological realm outside of physical reality?

However, if we dwell too deep into the heart of philosophy of mind, the danger is great, that we will end up with metaphysical debates that might too easily be dismissed as theoretical talk with no direct relevance for psychiatry. Instead, I will refer here to two examples of the relevance of philosophy of mind for psychiatry: one specific approach of a theory of mental disorder by a philosopher (George Graham) and one family of problems discussed in contemporary philosophy of mind, namely if mental states extend beyond the brain in a relevant sense.

A comprehensive and accessible version of linking philosophy of mind and mental disorders has been given by Graham (2013). In his theory he explains what mental disorders are, according to which (normative) criteria we classify them as clinically relevant and how they differ as mental disorders from proper brain (=neurological) disorders. According to Graham a mental disorder is a disability, incapacity or impairment in one or more basic or fundamental mental faculties of psychological capacities of a person that has harmful or likely harmful consequences for its subject. It is a disorder because it is harmful and undesirable for the subject, whether the subject himself appreciates this or not. In more concrete terms this means that the person is worse off with than without the disorder, that the disorder has a non-voluntary and personally uncontrollable nature and that the disorder cannot be excised or extirpated by the mere addition of other psychological resources. For example, the delusion of a paranoid person will not be alleviated by giving more information about the content of his delusion and the sadness of a depressed person will not be cured by cheering him up. Mental disorders are mental disorders because they are brought about by a mix of mental forces and brute a-rational neural mechanisms, or at least Graham argues so. The crucial point here naturally is what Graham means by mental forces. He explicitly states that he is not a dualist. Rather, he tries to argue what the “mental” in mental disorders refers to. The mark of the mental is that states of mind are constituted by either or both of two elements, i.e., consciousness and intentionality. Only if the causal mechanisms bringing about or sustaining a mental disorder work through conscious and/or intentional states, so Graham claims, they should be categorized as mental disorders. Mental symptoms that arise from brute brain affections (like stroke, neurodegeneration, or infection) are neurological disorders even if they present with (secondary) mental symptoms. Also, the mental is decisive for the criteria when a mental state of mind should be regarded as a disorder and not as a variant of normal mind life: namely when they impact a person’s reason-responsiveness or rationality considerably without totally destroying it.

In Graham’s theory the mental plays a prominent role in several respects: first, because the mechanisms causing or sustaining mental disorders are supposed to work through those brain mechanisms that implement mental (intentional and/or conscious) states and thus through mental qua mental. Second, the normative criteria for clinical relevance (and thereby the criteria for separating normality from disorder) rely on the impairment of intentionality and rationality, i.e., marks of the mental. Thirdly, he argues that mental disorders (like panic attacks, schizophrenia, depression) should and can be distinguished from proper neurological brain disorders (like stroke, Parkinson, Alzheimer) by the fact that the latter are brought about by pure mechanical, brute, a-rational affections of the brain that moreover are not sensitive to psychological (mental) treatment. In contrast, the “mental” in mental disorders has a double role: first it is characterized by an impairment of intentionality and rationality and second, because these marks of the mental are not totally absent but within the symptoms there is still a sense of rationality and intentionality preserved.

A problem in Graham’s theory is his explication of mental forces. Sometimes, he seems to imply that rationality or intentionality have a causal power of their own, although he denies that. But the worth of his approach for biological psychiatry seems for me that he insists on the relevance of the role of the mental in understanding, explaining and identifying mental disorders against pure brain disorders and non-pathological mind states on the other. In fact, many proponents of biological psychiatry now accept an interplay of neurobiological and psychological (mental) factors. However, if the mental is identical with the neural what does this claim of interaction amounts to? So let us turn then to the important question, if the mental can really be reduced to the neural.

In philosophy and in cognitive sciences there exist a number of proposals that doubt that cognitive processes (for our purpose: mental states) are best understood as only internal processes that happen within a cognitive system (in our case: the brain). Internal approaches, so the basic idea, ignore that cognitive processes are situated, i.e., that they essentially depend on (weak version), or even may be constituted by (strong version), their embodiment and the interaction with the natural, technological and social environment. There is not yet a consistent or complete theory of situatedness, rather there are several strands of research and theorizing that can be subsumed under the catchword “the 4Es”: the embodied, extended, embedded and enacted mind (Lyre andWalter, 2013). The main idea is that in order to understand what cognition (the mental) is, it is necessary to take into account that cognitive capacities of a system may depend on the fact that those systems (our brains) are (i) embodied, i.e., coupled to our bodily constitution and that it therefore is necessary to regard the bodily realization of cognitive abilities as an integral part of the cognitive architecture; (ii) situationally embedded, i.e. are dependent in a specific way on their environment, i.e., cognitive systems exploit the specific circumstances of their environmental context in order to increase their performative abilities, (iii) extended, i.e., extend over the boundaries of our body into the technological or social environment and thus are constituted not only by internal factors but also by external, environmental factors and (iv) enacted, i.e., arise only by the active interaction of an autonomous systems with its environment (Walter, 2010).

The thesis of embodiment has a long tradition in phenomenological philosophy, e.g., in the writings of Merleau-Ponty. The thesis of the extended mind has more recently been introduced into the debate by a paper published in 1998 (Clark and Chalmers, 1998). They introduce an example of an external device for memory (a cognitive process) that has since then been discussed extensively in the literature. The example refers to the notebook of Otto, an Alzheimer patient with memory problems who uses his notebook instead of his normal physiological memory in order to remember certain things. The argument is that if the entries into the notebook play the same role in Ottos life and in the explanation of his behavior as neurally implemented memory contents in healthy adults, it would be arbitrary or neural chauvinism if we would not regard them in the same way as genuine parts of the material substrates of his normal memories and beliefs. The general form of the argument inherent in this example is called the parity principle: if a part of the world functions in a way that, would it happen in our brain, we would have no hesitation in recognizing as part of a cognitive process, then that part of the world is part of the cognitive process. To make this part of the process more plausible it is easy to modify the example such that the notebook is constructed as a brain-computer-interface, e.g., as a digital device coupled more directly to the brain, for example in a technological advanced form of the actually existing google glasses.

Why could the 4E thesis be relevant to understand the nature of mental disorders?

Because they regard processes external to the brain as constitutive for mental processes and thus also as constitutive for disordered, pathological mental processes. An example, where this might be relevant is ADHD. ADHD might be only correctly diagnosed as a mental disorder if the external world is such that adolescents grow up in an environment that favors attentional distraction and punishes hyperactivity. In a similar vein, anorexia nervosa, a severe and often deadly mental disorder in Western countries seems to be much less frequent or even non-existent in environments in which a slim figure and control of eating and weight is not promoted, like in poor countries in Africa. These facts seem to draw into doubt that every currently acknowledged mental disorder is best categorized as a pure brain disorder – which is not to deny that internal processes of the brain play an important role if specific circumstances hold.

The main point which I would like to make here is that biological psychiatry has to take into account theories about how the mental is constituted. The new wave of biological psychiatry might be able to incorporate these issues into its conceptualization of mental disorders – but only if it comes along with a consistent theory of the mental that should take into account arguments and insights of philosophy of mind.