Paradigms lost and the structure of psychiatric revolutions

The release of the fifth edition of the Diagnostic and Statistical Manual of the American Psychiatric Association in May 2013 represents a missed opportunity to move to a new psychiatric paradigm. A bolder vision would have incorporated the advances in neuroscience and psychotherapeutics that I believe form the basis for the next psychiatric revolution.

DSM 5 fails in this respect (Malhi, 2013). It continues the objective-descriptive approach that emerged from the work of the German psychiatrist Emil Kraepelin with his detailed and methodical delineation of psychiatric syndromes. This descriptive-phenomenological method was the dominant paradigm from the 1880s until the 1970s. Freud’s publication of the Interpretation of dreams in 1899 ushered in the second major psychiatric school of thought, psychoanalysis. The third wave in psychiatry, the psychopharmacological revolution, occurred in the early 1950s with the serendipitous discovery that Chlorpromazine had a calming effect in certain patients. This led to the overriding influence of pharmacotherapy that has largely persisted to the present time.

Limits of the current paradigms

Successive editions of DSM have had a profound influence in shaping the intellectual evolution of psychiatry. DSM III published in 1980 signalled a radical rupture with the preexisting taxonomy of DSM II. In DSM II diagnoses were defined loosely in narrative terms, strongly influenced by psychoanalytic theory. DSM III represented an important advance for psychiatric practice and research. Diagnosis became more rigorous and systemized and this raised the general standards of clinical practice and academic performance. DSM III introduced a multiaxial classification which was a concession to the biopsychosocial model. It was opportune to acknowledge that we must consider biological, psychological and social aspects of the patients’ predicament. But we did not have a paradigm to integrate these dimensions into a coherent whole. In reality the biopsychosocial model was a political response to a critique of psychiatry that saw it as too narrowly pre-occupied with psychopharmacologic treatments.

An objective-descriptive system has intrinsic limitations. A classification based on surface features without an explication of the underlying pathophysiology lacks explanatory power and construct validity. A criterion based system leads to overlap of syndromes and lack of sharp boundaries between diagnostic entities. This leads to the artificial over diagnosis of “comorbidities”. For example a patient diagnosed with posttraumatic stress disorder is also very likely to meet criteria for major depression because both conditions share many of the same symptoms. This results in the frequent diagnosis of “comorbid” major depression and PTSD. The patient in reality may have only one disorder not two. This can only be determined by an understanding of the pathophysiological basis for both disorders. There are similar problems with diagnoses of comorbid depression and general anxiety and comorbid PTSD and borderline personality disorder just to name a few examples. In fact it is estimated that comorbidity occurs in 35 to 45% of patients (van Loo et al., 2012). In a large proportion of these it is likely that this is in fact pseudo-comorbidity based on this artifact of overlapping diagnostic criteria. A model which generates such a high rate of false comorbidity creates difficulty both for research and clinical practice. Research subjects lack homogeneity and clinical practice is hampered by inconsistent results based on categories with unclear boundaries.

Over the past five decades there has been a shifting emphasis between psychopharmacology and psychotherapy in psychiatric education and clinical practice: a dialectic between a fundamentally biological viewpoint utilizing medications versus a primarily psychologically based conception. This dialectic is consistent with Kuhn’s view that it is the nature of scientific revolutions that prior to the emergence of a new paradigm there is a prodromal period when differing schools of thought compete for domination of the field (Kuhn, 1970).

The dominant position of the psychopharmacological paradigm has come under intense scrutiny (Angell, 2011). It is clear that theories of etiology based on single neurotransmitters are a simplification no less metaphorical than psychoanalysis. These theories have resulted in a narrow focus of pharmaceutical research with production of “look alike” molecules but no substantive discoveries of novel therapeutic drug actions for more than a decade.

Within the psychotherapeutic field a contest has raged for clinical and research primacy between psychodynamic and cognitive psychotherapies.

The psychodynamic therapies focus on emotional and unconscious processes based on a model of childhood development. From a neuroscientific vantage point this model reflects activity in the neurocircuits involved in emotion control located in subcortical brain areas. Cognitive behavioral psychotherapy is based on conscious thought and voluntary behavior. Its neurobiological locus is the cerebral cortex.

Are we now at the point where we have reached the limits of the objective-descriptive, psychoanalytic and psychopharmacolgical paradigms? Is it time to accept that these models are no longer able “to define the legitimate problems and methods of a research field” (Kuhn, 1970, p. 10). Can these divergent lines of thought be now brought together in a new unified paradigm?

Foundations for a new paradigm

Over the past 20 years, there has been a revolution in neuroscience research in psychiatry based on advances in neuroimaging. Application of fMRI and PET scan technology has led to thousands of publications (Bandettini, 2012; Mayberg 2007; Ressler and Mayberg, 2007; Sporns et al., 2005; Strakowski, 2012). Cardiology was transformed by echocardiography. The field of infectious diseases was revolutionized by the invention of the microscope. Psychiatry is at the point where the new modes of observation and investigation can revolutionize the data we collect and how we make sense of it. This has the potential to result in quantum advances.

What are the bases for a new paradigm?

I believe there are two foundational fields of inquiry that can provide the elements of a coherent explanatory paradigm. The first of these derives from the work of animal researcher and neuroscientist Jaak Panksepp and the publication of Affective neuroscience (Panksepp, 1998). Panksepp’s work (Panksepp, 1998, 2004; Panksepp and Biven, 2012) has elucidated the workings of the mammalian affective system. He has identified seven basic emotional operating systems and their corresponding neurocircuits. These neurocircuits underpin social bonding and separation distress, fear, anger, exploration, maternal care, sexual behaviour, and play. The concepts of affective neuroscience are highly compatible with the Research Domain Criteria Project of the National Institute of Mental Health (Insel et al., 2010). The National Institute of Mental Health has recognized the limitations of the current diagnostic systems and in response has launched the Research Domain Criteria project. The purpose of this latter project is to develop classifications built from the ground up based upon research in neuroscience, genomics, brain imaging and physiology, molecular biology, and neural circuits. NIMH has identified key domains related to fear, approach motivation and reward, cognitive systems, social processes including attachment separation and social dominance, and arousal and regulatory systems. The fundamental premise of this approach is that psychiatric disorders are disorders of neurocircuits. This is a long-term open-ended project that by its very nature does not propose a unifying heuristic conceptual model. In contrast, what I am proposing is something that has immediate application. Affective neuroscience is based firmly in the study of animal neurocircuits and rooted in evolutionary biology. Findings from neuroimaging studies provide a parallel body of knowledge that confirms the importance of subcortical brain regions in most of the major psychiatric disorders: depression, bipolar disorder, anxiety disorders, and post-traumatic stress disorder. Affective neuroscience can potentially allow for regrouping and reclassification of these disorders based on an understanding of the seven primary emotional operating systems of the brain described by Panksepp.

The second foundational component of the new paradigm is attachment theory, developed by John Bowlby in the 1950s. Attachment theory has become one of the most fruitful sources of testable hypotheses in childhood development, personality theory, and interpersonal relationships. A large body of empirical data has confirmed that the quality of infant maternal attachment has a major impact on childhood development, emotion regulation, interpersonal functioning, and susceptibility to psychiatric disorder (Cassidy and Shaver, 2008).

Fundamental to these approaches is the application of the attachment model to the psychotherapeutic relationship. The therapeutic relationship is based on some of the same essential parameters that arise in mother–infant interactions: secure base, responsivity, attunement, empathy, and synchrony (Fosha, 2000). The application of these ideas has been incorporated into many of the newer experiential and emotion-focused psychotherapies: experiential dynamic psychotherapy, relational psychoanalysis, and mentalization-based psychotherapy.

The link between attachment theory and affective neuroscience is provided by Panksepp’s work. He has elucidated the neurobiological basis for the attachment response in limbic neurocircuits mediated by various neuropeptides. Thus a basis for a paradigm uniting neuroscientific and interpersonal processes now exists.

Because we now know that there are reciprocal interactions between limbic and cortical regions, the current conceptualizations of psychotherapeutic interventions can be unified within the new paradigm. Emotional experience and regulation involve both top-down cortical conscious appraisal as emphasized in cognitive psychotherapy and bottom-up subcortical limbic dimensions which are central to the techniques of dynamic and experiential psychotherapies and of treatment with medications.

What will the new paradigm look like? I will take as an example the diagnostic entity of borderline personality disorder. DSM IV defined this by the presence of at least five out of a list of nine symptoms which include self-harm, fear of abandonment, affective instability, and identity disturbance. These criteria yield a potential of 126 symptom combinations. At the very least, this implies the disorder embraces a heterogeneous collection of phenotypes. This classification leads to diagnostic overlap with other DSM diagnoses such as dysthymia and post-traumatic stress disorder. Borderline personality defined in this way does not accurately predict treatment response and undermines basic research because of inadequate construct validity. DSM 5, although substantially revised in approach, continues to rely on a trait-based method subject to the same limitations.

In the affective neuroscience-attachment paradigm that I am proposing here, borderline personality would be defined as a disorder of neurocircuit function involving the prefrontal cortex and the amygdala, aetiologically related to insecure attachment (New et al., 2007; Eppel, 2005 ). This understanding would promote an agenda for research and rationally developed treatment strategies. Research could focus on neuroimaging studies of attachment and emotion-regulating neurocircuits, elucidation of the neurotransmitters and neuromodulators involved, and studies of the antecedents of insecure attachment in this population. Treatment interventions and prevention strategies would be guided by attachment-based psychotherapeutic principles and mood-regulating pharmacotherapies.

The affective neuroscience–attachment paradigm provides a unifying framework that integrates the biological and psychotherapeutic developmental perspectives. This model derived from basic science would place research questions within an explanatory framework. The new paradigm would have enormous heuristic and integrating potential for students and allow for a clinical practice that combines biological and psychotherapeutic endeavours. The affective neuroscience–attachment paradigm can be applied similarly across the spectrum of psychiatric disorders and holds the potential for revolutionary research breakthroughs and an inspiring transformation for the future of psychiatric practice. Affective neuroscience and attachment theory complement each other. The former describes the basic emotional operating systems, their neuroanatomy, and neurochemistry; the latter provides the developmental and interpersonal dimensions of the paradigm. It is the interaction of the neurobiological inheritance with the social and interpersonal environment that forms the basis for human psychological and social development. It is derangements of these processes that lead to psychiatric disorder.