I present this very long post with minimal revisions and surely with its fair share of mis-spelled words and editorial mistakes. But I just want to get it out at this point… Consider it a first draft of my ideas on the Research Domain Criteria (RDoC) approach that has been presented as the research-based contrast to the Diagnostic and Statistical Manual 5, the latest version of the “bible” for diagnosing mental illness.
I spend a long time in the first part of the post outlining what the RDoC is and the logic behind its approach to research on mental health. This post mixes together the anthropology of biomedicine with critical neuroanthropology, so I wanted to really get at where the RDoC approach comes from. Well, “really get at” might be an exaggeration, but I did go over some of the basic documents and reporting on the RDoC.
Then in the second half I engage in critiques of the RDoC. The RDoC minimizes the social and cultural dimensions of mental health and health care, promotes one vision of the brain over other potential visions, and has difficulties in how it has conceptualized the move from research to improved diagnosis. Whether you consider those flaws minor or fatal likely depends on your overall views about mental health, but that’s not really my point here. Rather, I think it’s important to consider the RDoC approach quite specifically, given that it’s presented as something open to adaptation and also as a guide to funding for years to come.
Tom Insel, the National Institute of Mental Health, and the Research Domain Criteria approach to Mental Illness
The New York Times provided a recent profile of Tom Insel, the innovative psychiatrist and neuroscientist and long-serving director of the National Institute of Mental Health, part of the US’s National Institutes of Health. It covers his career, from developing pharmacological treatments early in his career to his ground-breaking work with prairie voles and the biology of attachment. But I want to focus on the ending parts, and the vision that Insel has for research on mental health.
Benedict Carey writes that Insel recognizes that “the previous generation of biological research in psychiatry has been largely a disappointment, both in advancing basic science and in improving lives.” So Insel is doubling-down on neuroscience as the way to tackle mental health.
His second stubborn conviction is that the only way to build a real psychiatric science is from first principles — from genes and brain biology, as opposed to identifying symptom clusters. Some of the mental health institute’s largest outlays under Dr. Insel have been to support projects that, biologically speaking, are like mapping the ocean floor.
One is the Psychiatric Genomics Consortium, a far-flung group of top research centers that share data and analysis, based at the lab of Dr. Patrick F. Sullivan at the University of North Carolina. The other is the Human Connectome Project, a $40 million, five-year program to build a baseline database for brain structure and activity using M.R.I. imaging.
In April, when in a blog post Dr. Insel wrote that the D.S.M.-5 was “at best a dictionary” and lacked scientific validity, he wasn’t exaggerating for effect. He had to qualify his comments because he doesn’t yet have a replacement. But he is determined to remake psychiatric diagnosis entirely and has set up an alternative framework for doing so, called the Research Domain Criteria — RDoC, for short — to be built from the ground up, on genetic findings.
“My philosophy is really based on humility,” he said. “I don’t think we know enough to fix either diagnostics or therapeutics. The future of psychiatry is clinical neuroscience, based on a much deeper understanding of the brain.”
It’s fairer to say that Insel’s approach is really based on the distinction between symptoms and signs used in biomedicine. Symptoms are what the patient experiences as related to the disease, the sorts of things that the patient reports to the doctor. Signs are the objective indicators that doctors use to diagnose the patient and which play a major role in determining what is the best treatment for a specific problem. Insel generally views psychiatry as diagnosing and treating symptoms, when what is needed is a biomedical science of clinical signs for psychiatry.
This approach is readily apparent in his critique of the DSM 5, Director’s Blog: Transforming Diagnosis.
While DSM has been described as a “Bible” for the field, it is, at best, a dictionary, creating a set of labels and defining each. The strength of each of the editions of DSM has been “reliability” – each edition has ensured that clinicians use the same terms in the same ways. The weakness is its lack of validity. Unlike our definitions of ischemic heart disease, lymphoma, or AIDS, the DSM diagnoses are based on a consensus about clusters of clinical symptoms, not any objective laboratory measure. In the rest of medicine, this would be equivalent to creating diagnostic systems based on the nature of chest pain or the quality of fever.
Insel wants this approach to be completely replaced. He continued in his post:
Patients with mental disorders deserve better. NIMH has launched the Research Domain Criteria (RDoC) project to transform diagnosis by incorporating genetics, imaging, cognitive science, and other levels of information to lay the foundation for a new classification system. Through a series of workshops over the past 18 months, we have tried to define several major categories for a new nosology (see below). This approach began with several assumptions:
• A diagnostic approach based on the biology as well as the symptoms must not be constrained by the current DSM categories,
• Mental disorders are biological disorders involving brain circuits that implicate specific domains of cognition, emotion, or behavior,
• Each level of analysis needs to be understood across a dimension of function,
• Mapping the cognitive, circuit, and genetic aspects of mental disorders will yield new and better targets for treatment.
It became immediately clear that we cannot design a system based on biomarkers or cognitive performance because we lack the data. In this sense, RDoC is a framework for collecting the data needed for a new nosology. But it is critical to realize that we cannot succeed if we use DSM categories as the “gold standard.” The diagnostic system has to be based on the emerging research data, not on the current symptom-based categories.
A lot of money and effort is being dedicated to this initiative and to the broad contours of how it sets funding for NIMH. A main question becomes, does this represent something more than simply pushing research beyond the previous generation of biological research and its acknowledged failings? Where does the empirical basis for such a vision come from? Is there even one?
Basic Brain Biology as the Basis for Studying Mental Illness
The Chronicle of Higher Education published an excellent piece A Revolution in Mental Illness on the Research Domain Criteria approach of NIMH last year. Journalist Paul Voosen provides a useful guide to understanding the NIMH’s view of mental illness as tied to problems in basic brain systems. I’ve linked together different parts of the essay here to illustrate this view, focusing on the fear response.
Davis soon delineated, following pioneering work by Eric Kandel into sea-slug neurons, a fear-response neural pathway in the rats, mediated by the amygdala. It was clear all mammals, including humans, shared a similar pathway.
As Davis’s work came out, the DSM-III had only recently appeared, codifying distinct anxiety disorders. Every month, there would be a new article on the psychophysiology of specific phobia or PTSD; it’s all about the amygdala, each proclaimed. These were all different disorders, yet somehow their physiology was the same. How could that be?
The patients suffering from specific fears or one-time traumatic events showed a heightened startle, similar to Davis’s early experiments. The patients who suffered from strong, prolonged distress, however, had blunted reactions, their startle barely a blip—a pattern that held true across the different anxiety diagnoses. Somehow, this dimension, the fear circuit, had gone dysfunctional or burned out. But it’s hard to say exactly how without getting these patients under an MRI machine.
“Now the big thing to do is see if these same patterns show up in the brain itself,” Lang said. And that’s exactly what he’ll be doing for RDoC.
Work by Lang, Davis, and many others on the brain’s fear circuit had reverberated into a broader “brain circuit” hypothesis of mental illness, where disorders are caused by malfunctions in long chains of neurons, often beginning in the brain’s early development. Discrete sections can’t be tied to depression, just as dopamine, the neurotransmitter, can’t be called a “love hormone.”
In other words, this approach is as much about overcoming the neurotransmitter-and-modularity view of the brain as anything else. Discrete localization and modulatory chemicals didn’t add up to a robust view of brain function and malfunction. More work is needed. As the NIMH puts it:
Building on new discoveries from genetics, neuroscience, and behavioral science, we are better poised to understand how the brain, behavior, and the environment interact to lead to mental disorders. Mental illnesses are now studied as brain disorders, specifically as disorders of brain circuits….
To further clarify how changes in neural activity contribute to mental disorders; it will be necessary to know more about the basic neuroscience of neural circuit formation and how these circuits interact to contribute to observable behaviors. These efforts will require teams to integrate findings from studies of genomics, neuroscience, behavior, and the environment. This research will serve as the foundation for translation to clinical studies.
It’s certainly compelling. Mental illness runs through the brain, and investing in the basic science will likely yield significant results over the long-term in our understanding of the biological bases of mental and behavioral health and what we can do to better manage brain health. But that statement is rather innocuous. NIMH has developed a specific view of how to go from basic neuroscience to mental illness. That’s the bet they are making, and where most of the contention lies.
Research Domain Criteria and Psychophysiology
What is fascinating to me is that the move from basic neuroscience to the full vision for Research Domain Criteria (RDoC) comes from a place I wouldn’t expect – psychophysiology. The Chronicle’s A Revolution in Mental Health outlines the research and logic behind Insel’s push for Research Domain Criteria and a brain-based view of mental health.
The psychophysiology approach is marked by measurement linked to underlying physiological and neural processes. For example, Peter Lang developed a ground-breaking approach to studying fear using the startle response, in particular eye blinks, and linking that to basic central nervous system responses:
Lang judged that emotion could be measured in three ways: There was behavior, what we do in a situation. And there was language, which dominated at the time. (“How frightened are you, from 1 to 9?”) But language is a distant event, filtered through consciousness’s confusion. And so Lang added a third dimension: the body’s response. Heartbeats, sweat, even the brain’s electrical firing—these would be data that he could then compare to language and behavior, building a construct of fear. (Such constructs lie at the heart of NIMH’s new system.) Along the way, he taught himself all the technical tricks needed for what only later came to be known as psychophysiology.
Lang and others eventually found two primal emotions: valence—how “good” or “bad” something is—and arousal, from calm to wired. At the same time, he also built standardized systems to elicit these two emotions, directing subjects to imagine themselves in a situation, or to look at photos of food, or erotica, or violence; this latter system, the International Affective Picture System, is widely used throughout psychology and neuroscience.
This approach forms the conceptual core of the RDoC:
The agency has identified five primary “domains” of mental function, which are subdivided into lower-level systems, like the fear response, that have some known tie to behavior and a brain circuit. These domains can be simply stated: one for keeping the brain up and running; one for social processes; one for storing and using information; one for moving toward positive rewards, like food and shelter; and one for avoiding harm, which is where the fear circuit resides.
This is rather abstract, so let’s take major depression, as defined by the DSM, as an example. As an RDoC-style researcher might describe it, multiple mechanisms are at play in a severely depressed person, and they vary from patient to patient. There may be dysfunction in the neuroendocrine system; in reward-seeking activities; in emotion regulation; in neurotransmitter systems; in cognition; in epigenetic controls. Many of these mechanisms are still poorly understood at a basic level, let alone in how they may connect to a circuit of dysfunction and illness. Cuthbert wants to give scientists permission to start exploring those connections.
What the NIMH has done is provided the framework they believe necessary for systematically studying those connections and connecting them to translational outcomes. I recommend going to the NIMH’s page that provides the RDoC Matrix that they have developed for the five domains – it is a concrete visualization and methodical laying-out of the sub-categories with each of the research domains – Negative Valence Systems, Positive Valence Systems, Cognitive Systems, Systems for Social Processes, and Arousal and Regulatory Systems.
For example, the domain of Negative Valence Systems is broken into several constructs that make up the domain; there are Acute Threat (“fear”), Potential Threat (“anxiety”), Sustained Threat, Loss, and Frustrative Nonreward. That’s on the vertical axis.
On the horizontal axis are the Units of Analysis: “Genes, Molecules, Cells, Circuits, Physiology, Behavior, Self-Reports, Paradigms.”
The research then links the constructs in a particular domain to underlying neurobiology and to specific measures. So, with Fear (opposite pole, – fearlessness), researchers might look at “amygdala, hippocampus, interactions with ventromedial PFC” and with Potential Threat, “HPA axis, BNST, hippocampus; CRF, cortisol.”
How might such an approach play out in an actual study? NIMH provides an example-driven guide:
A design to study fear circuitry might thus have as inclusion criteria all patients presenting at an anxiety disorders clinic. Classification variables: The construct of interest is Fear/extinction, in the domain of Negative Affect. The independent variable for grouping would be the extent of responding to fearful stimuli using a measure such as amygdala response (from fMRI) or fear-potentiated startle (i.e., a circuit-level variable). Dependent variables would be symptom measures on various fear and distress measures, in order to test hypotheses about mechanisms by which hyper-reactivity and hypo-reactivity to threat cues affect the nature and severity of presenting symptoms. As an outcome of such research, these results might generate predictive validity studies leading to improved treatment selection or new pharmacological targets for intervention.
What will the outcome of this approach be? My guess is that most neuroscience and mental health research will proceed apace. It already uses into a relatively compartmentalized approach, focusing on experimental studies that link brain function to cognitive processes with implications for mental and behavioral health. A good example of this work is highlighted in the DANA piece on the molecular neuroscience of memory and addiction, Solving the Mystery of Memory. Researchers will adapt their work to the new RDoC framework. For me, I’d likely slot this work into the Positive Valence System domain, under the construct of Reward Learning, using units of Genes, Molecules, and Cells linked to Circuits and Behavior.
I also think this approach will lead to the novel development of clinical tests linked to basic neural processes, and that this will be a major development. That’s the promise of the Lang approach of psychophysiology, with an emphasis on studying what he can measure. Insel wants biomedical signs for mental health diagnosis, and this approach should yield that sort of translational work.
A third development is the continued expansion of the circuitry view of mental illness, with increasing examination of how different combinations of brain circuitry and function underlie different aspects of mental illness. This approach was already on its way, but now it has official sanction.
A fourth development will be in basic neuroscience, part of an overall governmental push to invest in doing a Genome Project for neuroscience. To return to the NY Times article on Insel, this is an effort to “support projects that, biologically speaking, are like mapping the ocean floor.”
And the goal of reforming the DSM? Here’s where the political fight really lies, and where the NIMH officials have been the most vehement. For example, in The Chronicle piece, Bruce Cuthbert (“lead architect of NIMH’s radical restructuring”) is pretty pointed about the DSM:
“We’ve had a decade of the brain. We’ve had a decade of behavior and we’re ready to move toward translational models and thinking about our disorders in a different way. We’re ready to shift away from folk psychology.”
“If you think about it the way I think about it, actually the DSM is sloppy in both counts [biology and psychology]. There’s no particular biological test in it, but the psychology is also very weak psychology. It’s folk psychology without any quantification involved. What we really need to do is elevate both.”
What’s Wrong with the DSM 5? The NIMH View
NIMH, in describing the Research Domain Criteria approach, is clear about the contrast between the DSM approach and the bottom-up biological approach they are advocating:
The current diagnostic system is not informed by recent breakthroughs in genetics; and molecular, cellular and systems neuroscience. Indeed, it would have been surprising if the clusters of complex behaviors identified clinically were to map on a one-to-one basis onto specific genes or neurobiological systems. As it turns out, most genetic findings and neural circuit maps appear either to link to many different currently recognized syndromes or to distinct subgroups within syndromes.
If we assume that the clinical syndromes based on subjective symptoms are unique and unitary disorders, we undercut the power of biology to identify illnesses linked to pathophysiology and we limit the development of more specific treatments. Imagine treating all chest pain as a single syndrome without the advantage of EKG, imaging, and plasma enzymes. In the diagnosis of mental disorders when all we had were subjective complaints (cf. chest pain), a diagnostic system limited to clinical presentation could confer reliability and consistency but not validity.
RDoC is intended as a framework to guide classification of patients for research studies, not as an immediately useful clinical tool. While the hope is that a new way forward for clinical diagnosis will emerge sooner rather than later, the initial steps must be to build a sufficient research foundation that can eventually inform the best approaches for clinical diagnosis and treatment. It is hoped that by creating a framework that interfaces directly with genomics, neuroscience, and behavioral science, progress in explicating etiology and suggesting new treatments will be markedly facilitated.
This critique is amplified in the long Chronicle piece, A Revolution in Mental Health, by the NIMH folks. First comes the critique of the artificial silos of mental illness that the DSM imposes – depression as separate from schizophrenia as separate from addiction.
Cuthbert found his new world to be constricted by the DSM’s categories, the research siloed in terms of “depression” or “bipolar disorder.” This was not like the work they were doing in Gainesville. “We weren’t getting at these fundamental mechanisms,” he says…
“As a result of the DSM,” [former NIMH director] Hyman says, “investigators failed to see all of the mixed symptoms.” Nature is a meatloaf, he knew, yet these “fictive diagnostic silos” were holding the science back…
“You can pretty easily see that risk alleles don’t respect diagnostic criteria at all,” says Michael Owen, a geneticist and dean of research at the Cardiff University School of Medicine, in Wales. His lab first found that schizophrenia and bipolar disorder shared deep genetic similarities. Then they found links to autism, ADHD, and even intellectual disabilities.
“This got us thinking, looking at the clinical data, that yeah, it’s possible to put people in these boxes, these categories,” Owen says. “But there’s no clear water between them. It made me realize these were categories we were imposing on people.”
Second is that within each silo, sub-categories proliferated as each group of experts fought for their piece of the pie. It was the Balkanization of mental illness.
[Lang] readily admits, as does Cuthbert, that the third edition of the DSM, released in 1980, helped save psychiatry, giving clinicians somewhat reliable diagnostic standards for disorders like agoraphobia, specific phobia, panic disorder, and others, and freeing them from their psychoanalytic shackles. But his work did not respect the arbitrary boundaries in the manual; he has always welcomed all kinds of anxiety patients at his labs, grouping them in the same projects. He watched with dismay as researchers Balkanized. “That was very constraining, because the clinical material is not like that,” he said.
In other words, it’s not just the silos that are wrong, but how the silo then gets sub-divided. The proliferation of diagnoses in the DSM 5 is a common critique, and NIMH is on board. As the NIMH document puts it:
In contrast to cancer and heart disease, where research has identified subtypes of common disease, it appears that the biological findings with mental disorders are relatively non-specific; could specificity in fact exist, but not for the currently recognized clinical categories? This question leads to a consideration of how current categories were derived.
Research Domain Criteria: Updating the Bio-Bio-Bio Approach
Tanya Luhrmann, in her piece on schizophrenia for The Wilson Quarterly, Beyond the Brain, writes a history of schizophrenia and its diagnosis. She describes the biomedical approach developed in the 1980s and 1990s as follows:
It is now clear that the simple biomedical approach to serious psychiatric illnesses has failed in turn. At least, the bold dream that these maladies would be understood as brain disorders with clearly identifiable genetic causes and clear, targeted pharmacological interventions (what some researchers call the bio-bio-bio model, for brain lesion, genetic cause, and pharmacological cure) has faded into the mist. To be sure, it would be too strong to say that we should no longer think of schizophrenia as a brain disease. One often has a profound sense, when confronted with a person diagnosed with schizophrenia, that something has gone badly wrong with the brain.
The Research Domain Criteria, in my view, represent an evolution of the bio-bio-bio model. It is codifying a paradigm shift in the biological study of mental illness, and doing so in the way of biomedical science – through reflection on the increasing empirical knowledge and conceptual tools available to biomedicine over the past two decades. In this sense, the Research Domain Criteria represent a positive development.
The new bio-bio-bio model is Brain Circuits, Genomes, and Biological Nosology. Brain circuits have replaced an over-emphasis on localized malfunctions and on neurotransmitters-as-disease views (e.g., depression as a deficit in serotonin). Genomics highlights how the regulation of genes matters as much as the genetic code itself, and that for major mental illness, many many genes are involved, generally with quite small effects (Kenneth Kendler’s 2013 piece on psychiatric genetics is excellent here). And the Research Domain Criteria is premised on the promise of a new biologically-informed approach to psychiatric diagnosis, where the hope is that this in turn will lead to more tailored pharmacological and behavioral treatment. More specific nosology should, it is hoped, lead to better targeted interventions.
I agree with Luhrmann. It is hard to get away from the recognition that with serious mental and behavioral problems, something has seriously gone wrong with the brain. The Research Domain Criteria approach is a step forward in trying to understand just what has gone wrong.
The old model is wrong. Now there are five dimensions, each with sub-categories, and those can get combined in new ways to try to explain the complexity of mental illness, both within major categories (depression) and across categories (anxiety and depression). Is it a better model?
Yes, in the sense that it works with a biological framework where thinking about mental illness didn’t match up with the results coming from biological research. Insel clearly think this, as the NYT piece describes his “stubborn conviction” that “the previous generation of biological research in psychiatry has been largely a disappointment, both in advancing basic science and in improving lives.” Insel is betting that advancing basic science will result in improving lives. Given the overall track-record of science, technology, and biomedicine, it seems a pretty safe bet to make.
Research Domain Criteria: The Social Critique
As I alluded to above in discussing likely outcomes from the RDoC approach, the problem isn’t really on the basic science side – this work will continue on with or without the specific RDoC approach. The debate should center more on whether the RDoC is the best way to (a) conceive of mental illness and (b) improve people’s lives who suffer from mental health issues.
It is clear that the easier debate surrounds whether the NIMH approach helps people now. In their profile of Tom Insel, the New York Times provides the obligatory “other side” take on the NIMH approach.
“Instead of being an institute of mental health, he has made it almost exclusively a brain research institute,” Dr. Allen Frances, an emeritus professor of psychiatry at Duke and the author of the book “Saving Normal,” wrote in an email. “N.I.M.H. is betting the house on the long shot that neuroscience will come up with answers to help people with serious mental illness.” He added, “It does little or no psychosocial or health services research that might relieve the current suffering of patients.”
Another way to put this is that that for psychiatry, the Research Domain Criteria is the right bet. Psychiatrists engage in highly specialized care that depends on technical knowledge and expertise. But for mental health overall, given the resources that the NIMH employs, the Research Domain Criteria is likely the wrong bet.
First, mental health care is not going the way of specialized psychiatric care. The long-term trend is to reduce the number of beds. Drug companies are not investing in new drugs. And a major worldwide push today is to figure out more community-based approaches, which are low-cost and can target interventions where mental health patients live.
Second, mental health problems are rather like health problems. The Robert Wood Johnson Foundation is pushing for major investments in the health of children and communities as the major way to improve physical and mental health. As Zip Code Overrides DNA Code When It Comes to a Healthy Community puts it,
Targeted investment [can] yield the greatest improvements in overall health now and for generations to come. These recommendations urge us to fundamentally change our approach to improving health, recognizing that we need to do more than treat illness—we need to keep people healthy. The recommendations call for prioritizing investments in our youngest children, encouraging leaders in different sectors to work together to create communities where healthy decisions are possible, and broadening the mission of health care providers beyond treatment.
For mental health specifically, a different bet could be made on the links between biology, mental health, and improvement. Such a bet would look more concretely at the links between the social, the mental, and the biological. It too could be funded by a big investment in “ocean mapping” of our human lives. As Des Fitzgerald, Nikolas Rose, and Ilina Singh write in Urban life and mental health: re-visiting politics, society and biology:
[This] work helps us to understand how ingrained patterns of social-stress-processing might be an environmental risk that links city life to mental ill-health – showing us not only that city life becomes embodied, but also potential mechanisms for how it becomes embodied.
NIMH recognizes how much development and environment matters, discussing these additional dimensions of mental health after the RDoC criteria:
The RDoC concept is organized around basic neural circuits, their genetic and molecular/cellular building blocks, and the dimensions of functioning that they implement. There are two highly important areas of mental disorders research that are thus not represented in the matrix per se, but are considered to be critical elements in research fostered by RDoC. These two areas are developmental aspects and interactions with the environment. The intent is that the RDoC matrix will enhance the study of both areas by promoting a systematic focus on their relationship to specific circuits and functions.
But that’s not quite the same as making these dimensions the core research framework of an institute, with money that will fund ground-breaking basic science. There is also another danger, that this approach will once again lead to privileging the biological over the social. As I put it in my article Poverty Poisons the Brain, the focus will still remain on the brain, rather than the poverty. That is essentially what the RDoC approach does.
Research Domain Criteria: The Brain Critique
It’s unlikely, however, that the NIMH will suddenly change course. But critical inquiry can also focus on how the Research Domain Criteria work within their given focus – on fostering better research on mental health and illness using an approach grounded in neuroscience. What potential problems are there here?
In short, the strength of the RDoC approach is also its weakness – the NIMH is promoting a mechanistic view of brain function, one that has worked well for other areas of biomedicine but still remains problematic for understanding the links between brain function and mental health.
Insel makes the distinction between symptoms (subjective experiences of illness) and signs (objective criteria for disease), and that the RDoC will provide the latter while the DSM has focused too much on the former. That might be a fair critique of the DSM, but Insel is also trying to sidestep the mind/brain dichotomy as well. Heart pain might be the subjective symptom of congestive heart failure, but such a neat dichotomy doesn’t work as well with the brain, which after all is what is constituting the subjective experience of heart pain or any other kind of pain for that matter. The brain is not like the heart – subjective “symptoms” is part of how it functions.
In other words, subjective experience is very likely a formative part of most mental illnesses, not simply a subjective layer on top. This point comes up in Tanya Luhrmann’s work on how hearing voices varies across cultures and my research on how the subjective experience of drugs plays a central role in why adolescents abuse drugs. In both cases, the research in no way denies the biological (in my case, I actively draw on neuroscience and theories of incentive salience based on animal research). The same integrative approach will also be necessary for better understanding how neuroscience links to mental illness, but the matrix completely overlooks this necessity. Rather, it’s compartmentalized, and looks largely as correlations with symptoms rather than direct relations. Take their example of “emotional challenges in patients with a particular mood or anxiety disorder” and how to study that:
Dependent variables would be symptom measures on various fear and distress measures, in order to test hypotheses about mechanisms by which hyper-reactivity and hypo-reactivity to threat cues affect the nature and severity of presenting symptoms.
But perhaps it’s okay to sidestep one of the oldest challenges in Western philosophy – the mind/body problem – when just trying to advance the neuroscience side. But even here I see three problems. These are more solvable problems, but all ones that come up in imagining the next steps to the new bio-bio-bio approach: circuits, genomes, and better knowledge & intervention.
On the circuits side, there is the same problem with localization approaches – is this really how the brain works? Certainly the circuits approach is a better approximation than the massive modularity and functional deficits combination. It’s also a very conservative take on neural function. For example, if the stronger versions of embodiment and extended mind continue to push their research forward, then a large gap will open up between research pushed by the NIMH and evolving models of cognition. More pointedly, if neural reuse theories of how the brain works prove more accurate than an approach that compartmentalizes function into domains and components, then the NIMH approach won’t have the predictive power it promises.
Neural reuse argues that “it is quite common for neural circuits established for one purpose to be exapted (exploited, recycled, redeployed) during evolution or normal development, and be put to different uses, often without losing their original functions (Anderson, 2010).” In this approach, the amygdala isn’t part of a fear circuit laid down to deal with negative valence and avoidance.
The traditional limbic system theory supports the idea that neural resources (e.g., physiological or somatic) were carried out by the evolutionarily old cortex (i.e., the so-called reptilian brain), whereas cognitive processes (i.e., higher-order functions) were subserved by the neocortex. The present view, however, integrates both the limbic system and the neocortex as separate but interacting brain systems functioning in parallel. The processes of long-term potentiation (LTP), long-term depression (LTD), and neural plasticity are just some of the ways that the brain can reorganize and change its pattern of activity across cortical regions (and across modalities) in response to experiences.
Following this logic, one can imagine that neural reuse, whether evolutionary old or new, also follows a similar trend whereby mental functions are mediated by separate but interdependent brain processes. In the context of emotional arousal, the domain highly implicated for such processing is the amygdala, interacting with the hippocampus, thereby playing a role in supporting the formation and storage of emotionally salient forms of declarative memories.
The brain state supporting such a process appears to occur primarily during the low-voltage, fast activity of rapid eye movement (REM) and stage II sleep (Anderson 2010: 273).
The neural reuse approach thus sits at an uncomfortable angle with a circuits/dimensions approach. The seemingly fixed nature of dimensions and circuits could prove illusory. That said, the Anderson article is a re-interpretation of lots of basic neuroscience and psychological research, so advancing basic science is still amenable to whatever type of model of brain function one subscribes to at this particular point of time.
A similar problem is represented in the shift from genes to genome. The simple mapping assumed before between genes and phenotypes has proven to be much complex than imagined. Kendler (2013) in his discussion of psychiatric genetics points that that many traits have proven to be heritable. What that actually means is harder to discern, particular for an approach premised on finding basic biological causes:
“Consider the heritable trait of ‘liking roller coasters’ that might be influenced by three high-level traits: risk for nausea, hedonic effects of rapid acceleration changes and thrill-seeking. Each of these traits is in turn influenced by a moderately coherent set of genes. But, because roller coasters are a human-constructed situation, there is no comprehensible underlying biology.
“Or, the construct of loving roller coasters exists at such a high level within the mind–brain system that it is currently out of reach for our science. In large GWAS or sequencing studies of the love of roller coasters, we would see these three gene sets turn up. But they would not be connected, nor with our current methods, would they point us to a coherent biological cause for love of roller coasters (Kendler 2013: 1062-1063).”
There is another option – the coherent cause around roller coasters doesn’t need to be biological. Amusement parks (like other human constructed situations) are what regularly bind together potential nausea, fun, and thrills. We create the coherence that through mechanisms like neural reuse, biological embedding, and neuroconstructivism results in variation in both the environmental distribution of liking roller coasters and in individuals who might be pathologically afraid or involved with roller coasters.
NIMH’S Cuthbert says he is giving scientists permission to explore connections between domains and units of analysis. But as neural reuse and psychiatric genetics highlight, “connections” might not be enough. There might be a need for a generative grammar to make sense of how the connections are created and maintained in ways that lead to multiple dimensions of mental health and illness.
To make an analogy, Insel is taking a behaviorist approach to mental health, much like Skinner took to language. But Chomsky’s theory of universal grammar built from a poverty of stimulus argument. There simply wasn’t enough time or exposure for children to learn language in the way Skinner proposed. Similarly, Insel might find that all the RDoC research ends up producing a load of new information, but still fall short. The new neurobiological research might simply not be enough to make sense of the grammar of mental health, precisely because it wasn’t conceived to do this kind of work.
Moreover, the move from a grammar to accounting for diversity is also theoretically difficult, but this is precisely the problem faced by the RDoC approach – what we can measure, basic features, the production of diversity. As Evans and Levinson (2009) write in The myth of language universals: Language diversity and its importance for cognitive science:
After surveying the various uses of “universal,” we illustrate the ways languages vary radically in sound, meaning, and syntactic organization, and then we examine in more detail the core grammatical machinery of recursion, constituency, and grammatical relations. Although there are significant recurrent patterns in organization, these are better explained as stable engineering solutions satisfying multiple design constraints, reflecting both cultural-historical factors and the constraints of human cognition.
Similarly, mental health varies radically in its symptoms and organization. There is likely some core grammatical machinery, some of which is surely represented in basic dimensions like positive valence and negative valence and arousal (and much of which is not). But explaining recurrent patterns cannot be just because of any underlying systems of inputs coupled to circuits – that is not enough, given what we already know about mental and behavioral health.
Finally, there is that other side – the developmental and environmental dimensions that the RDoC leaves to the end. One view of the brain is as a mechanistic engine, sure a bit of a Rube Goldberg machine because of evolution and with some wet wiring because of what it has to do, but still a self-contained computational machine, running on some biochemical alchemy that is increasingly getting hacked.
But there is another view that could be equally viable. One of the main problems with a neuroscience-driven view is that it won’t necessarily connect to life outside the lab or clinic, where mental health actually exists. Put differently, neuroscience might still be in a bit of a Ptolemy phase, where the universe revolves around it. Ptolemy’s earth-centric approach still provided useful sets of calculations; from his Tables, the positions of the Sun, Moon and planets could be calculated as well as eclipses of the Sun and Moon. But it turned out to be wrong.
The following excerpt from In mental illness, let’s go beyond nature v. nurture to look at what interferes with the brain’s function suggests an alternative view. This view is also likely incomplete, but certainly presents an alternative way to envision what happens with the brain, one that is also amenable to observation and measurement and probing.
An alternative model of the brain has emerged with research into the neurobiology of trauma as well as research into the different regions of the brain. With this model, the brain is understood more in terms of the functions of its components and adaptation to environmental conditions, especially the environments created through our relationships with significant people in our lives. This is an important change in scope from the biochemical model of mental illness that seems to rest on the “command, control, communicate” metaphor that has dominated information systems thinking since World War II.
In the end, the two views are likely two sides of the same coin. And without both sides, there isn’t currency. The NIMH is privileging one side, in line with supporting high-end neuroscience in the hope it translates into high-end psychiatry. This approach will definitely yield positive outcomes. Increasing knowledge of the brain, developing new diagnostic criteria for specific components of mental illness, and utilizing basic neuroscience and psychiatric genetics to understand varying patterns of mental illness that are often at contrast with present conceptualizations – these are all laudable goals.
Yet some of the limitations of this approach are also apparent at this early stage, and are likely to have direct bearing on how useful the research funded by the RDoC approach becomes in the long run. The brain isn’t like other organs, because subjective experience, social relations, and cultural processes are formative to how it works. The RDoC matrix also avoids the hard question of how basic neural processes combine in generative ways to pattern the diversity of mental and behavioral health continuums we see. And the NIMH isn’t investing in better understanding of what we recognize as major drivers of mental health in the community, and how to elaborate cost-effective mental health services that can reach a greater proportion of the population.
Research Domain Criteria: The Diagnosis Critique
A final problematic area for RDoC is its nebulous relation to diagnosis. It’s against the DSM 5, but not really, it’s only for research – that seems to be one of the main messages. The odd thing, though, in reading the NIMH RDoC documents is how much the major labels are lurking there in the background. Concepts like depression, anxiety, and trauma haven’t really gone away. They are rather like Voldemort; the thing that must not be named but everyone still thinks about.
It’s also clear that the NIMH thinks that a brain-based approach will usher in an era of clarity in communicating about mental health with the public. The National Institute on Drug Abuse has taken that same tack for a long time, promoting a brain-based view of addiction. The latest research on public and clinicians alike (see Meurk et al. 2013, Hammer et al. 2013, and Garriott 2011) shows a much more ambivalent impact, with stake holders using the brain-based model to various ends, often to the detriment of people suffering from substance abuse.
But both of these don’t focus specifically on the RDoC initiative and its plan that basic science will one day yield better diagnoses and more targeted treatments. In one way, it seems like a really long way around the main problem that is being faced – the basic neuroscience and clinical science focus to much more overlap between various mental illnesses than previously recognized, and that the underlying biology isn’t a good guide to anchoring present diagnoses (e.g., depression isn’t just about serotonin, and addiction isn’t just about dopamine). An equally viable approach would be to do a top-down analysis of this problem, of moving from the quite amazing diversity of mental health problems and experiences to identifying core dimensions. In other words, it’s a “big data” sort of problem that could profitably be tackled by the sorts of statistical and algorithmic approaches that can pick out patterns that might not be visible in the clinical setting or one particular line of research.
This approach would also require major funding and the collection of new data to create a better data base of clinical symptomatology and phenomenological experience, likely facilitated by the many patient and advocacy groups that have sprung into existence. But it could be done (see, for example, this new research project on the Experiences of Addiction). It would require a very different commitment of resources, but would yield results much sooner, in a period of years rather than decades.
Another problem that the NIMH approach is that the strict division between symptoms (subjective) and signs (objective) that work elsewhere in medicine might not as viable for mental health. If, as I argued above, the symptoms themselves are constitutive of mental illness, then the distinction loses its significance. A more useful conceptual approach might be to think of indications and counter-indications. Wikipedia defines indications as “a valid reason to use a certain test, medication, procedure, or surgery.” Indeed, a major concern of NIMH is developing more validity for diagnostic approaches, and here the neuroscience is likely to yield results much more quickly. It’s likely to be able to point to something like “this type of arousal pattern to a startle test indicates this type of anxiety disorder, different from other types of potentially anxious attending.”
The RDOC might also do better with contraindications, of where treatments aren’t likely to be viable. Certainly it could do so with the pharmaceuticals we already have in hand, but it would also be a way to expand the RDoC approach out from its core to the developmental and environmental dimensions the NIMH also values. The NIMH, using neuroscience, can definitely say, these developmental and environmental patterns and experiences and measures are contra-indicated given what we know about healthy brain function.
But that will require the NIMH expanding its overall vision about the RDoC. One of the things that I find most fascinating about its present conceptualization is that its proponents critique the DSM for creating these diagnostic silos and the subsequent balkanization of diagnoses within each category. Yet then they go out and do the same thing, creating new silos (positive valence, negative valence, etc.) and new means to balkanize with the different constructs in each domain. It says as much about how the anthropology of knowledge works in Western settings as anything else – a logical hierarchical approach that divides and sub-divides, and which will inevitably become unwieldy under its own weight. Perhaps they hope as the connections between domains are explored, the basic dimensions will fade into the background. It might be rather like the shift from the 0/1 of machine language to the programming language of C++, which can be both procedural and object-oriented – with the analogy that in the future, procedural treatments can be targeted at specific types of mental health problems (the objects) that research has revealed.
But the other possibility is already viable – that the RDoC matrix has already boxed the NIMH and its researchers into a particular corner, and will recreate much of the same history of the DSM without sufficiently addressing some of the major issues facing mental health patients and providers today.
These are both imagined histories, and perhaps the imagined vision of NIMH is the one that will emerge triumphant. Certainly it’s going to yield positive benefits; there’s little doubt about that. But given the diversity of causes of mental health and the needs of patients, their families, and providers, it is certainly important to reflect on whether the proposed Research Dimension Criteria approach is the best way to guide research and practice forward. That’s the whole point of this long exercise – a reflection on whether an idealized model for research will intersect with the reality of mental illness.