Is Psychoanalysis Still Relevant to Psychiatry?

Psychoanalysis is a theory of psychopathology and a treatment for mental disorders. Fifty years ago, this paradigm had great influence on the teaching and practice of psychiatry. Today, psychoanalysis has been marginalized and is struggling to survive in a hostile academic and clinical environment.^1,2 This raises the question as to whether the paradigm is still relevant in psychiatric science and practice.

In a difficult climate for the theory and practice of psychoanalysis, several responses have emerged, either by attempting to bridge the gap with science or by redefining the field as lying outside of science. Thus, some analysts have supported revised paradigms, such as attachment theory, that are better supported by evidence.^3,4 Others have taken the view that Freud’s ideas concerning the unconscious mind are compatible with modern neuroscience.^5,6 Still others have moved in the opposite direction, arguing that it is sufficient to offer a coherent interpretation of psychological phenomena.⁷ This review will briefly examine all these attempts to revive psychoanalysis.

Revising the Paradigm

Almost no scientific theory or medical treatment that is a century old can be expected to survive without major changes. In fact, one of the main reasons for the decline of psychoanalysis is that the ideas of Freud and his followers have gained little empirical support.⁸ Freud’s theoretical model of the mind and of child development has been challenged and refuted by a wide range of evidence.⁹

The absence of solid and persuasive evidence for the theory may be the consequence of its self-imposed isolation from the empirical sciences. The philosopher Karl Popper considered psychoanalysis to be a pseudo-science because it has produced so many hypotheses that cannot be refuted empirically.¹⁰

Several psychoanalysts have attempted to update theory and practice in the light of current knowledge.¹¹ However, virtually all attempts to carry out revisions have fallen into the same trap as afflicted Freud, that is, basing theory on clinical experience rather on replicable scientific evidence.¹² Moreover, as acknowledged by one leader in the field,¹³ training in psychoanalysis has not encouraged research, while articles describing empirical findings remain a rarity in psychoanalytic journals

Attachment theory is a notable exception.¹⁴ This model now has an extensive scientific literature.⁴ Its originator, John Bowlby, unlike other analytic theorists, gave priority to research findings.² An American psychologist who worked with Bowlby, Mary Ainsworth, developed the “strange situation,” a method of evaluating attachment styles in children.¹⁵ These styles can also be measured in adults using interview or self-report measures.⁴ There are now thousands of studies showing a relationship between attachment styles and mental health.⁴ For many psychoanalysts, attachment theory, in one form or another, has become their primary model.¹⁴

However, as pointed out years ago by the British child psychiatrist Michael Rutter,¹⁶ the attachment model has limitations. Like previous models derived from psychoanalysis, attachment theory does not take into account the temperamental and biological vulnerabilities that make mental disorders more likely to develop.^17,18 Nor does the model, in its present form, take full account of gene-environment interactions in development.¹⁹ These problems are, of course, by no means specific to attachment theory. More specifically, the attachment model also tends not to consider that children can benefit as much from multiple attachment figures (“alloparenting”) as from a mother.²⁰ Finally, attachment theory runs the risk of being another psychodynamic model that, by emphasizing problems in early childhood, can be used to blame parents for psychopathology in their children.

Is Psychoanalytic Treatment Evidence-Based?

Modern medicine and psychiatry expect all forms of therapy to be supported by evidence.²¹ Peter Fonagy,^11(p77) a psychoanalyst who is also a respected researcher, has acknowledged that “the evidence base for psychoanalytic therapy remains thin.”

In a recent research update, Fonagy²² found support for psychodynamic therapy in a variety of conditions. However, almost all the studies he reviewed concerned short-term psychodynamic psychotherapy. This type of treatment has a good evidence base^23,24 and its efficacy is comparable to that of cognitive behavioral therapy (CBT).²⁴ But conflation of brief therapy with classical psychoanalysis is misleading.

Time-limited dynamic psychotherapies, the only treatments derived from analytic theory that have been widely tested in research, may be efficacious independent of their theoretical position or brand name. This “dodo bird verdict” has been supported by a wide body of research and applies to all forms of psychotherapy.²⁵ Moreover, while meta-analyses of briefer psychodynamic treatments provide good evidence of efficacy,^23,24 these results cannot be generalized to long-term psychodynamic therapy, in which virtually no controlled trials have been conducted.

A few reports have attempted to examine the outcome of classical psychoanalysis.²⁶ However, a meta-analysis of 14 studies^{27 (p107)} concluded, “A limited number of mainly pre/post studies, presenting mostly completers analyses, provide empirical evidence for pre/post changes in psychoanalysis patients with complex mental disorders, but the lack of comparisons with control treatments is a serious limitation in interpreting the results.”

Fonagy²² claims efficacy for long-term psychodynamic therapy for personality disorders. However, the therapies that have been tested, such as his own “mentalization-based treatment,”²⁸ are not psychoanalysis but mixtures of psychodyamic and cognitive-behavioral interventions adapted for the treatment of severely ill patients.

The German psychoanalyst Falk Leichsenring has published meta-analyses of extended forms of psychoanalytic treatment,^29,30 claiming that there is sufficient evidence to support this treatment in complex mental disorders, and other recent reviews of this literature have made a similar argument.^31,32 However, these conclusions are not justified because of heterogeneous clinical presentations, small samples, and small effect sizes.³³ It would be possible, in principle, to conduct better studies that address all these issues. However, the cost of that kind of investigation would probably be prohibitive.

Given its cost, the lack of good evidence for the efficacy of psychoanalysis has reduced its market share. Consumers are now less likely to seek this lengthy and expensive form of treatment; for some time, it has been difficult to make a living as a full-time analyst.³⁴ Modern psychotherapists practice in a competitive market that includes a very large number of treatment methods. Psychotherapies of all kinds now tend to follow other paradigms and to be relatively brief.³⁵

Some analysts who rejected the classical model became prominent in the development of different approaches to treatment. For example, CBT, now the most influential form of psychotherapy, was originated by Aaron Beck, a psychoanalyst who had given up believing that Freudian methods were helpful for patients.³⁶ A new generation of clinicians, particularly those trained in clinical psychology, have tended to adopt this perspective.

Neuropsychoanalysis

In an era in which psychiatry is dominated by neuroscience-based models, psychological constructs tend to be neglected and may be taken seriously only when they have neural correlates.³⁷ Some psychoanalysts have sought to link their model with neurobiological research and to claim that newer methods of studying the brain can validate their theories.^5,6

Mark Solms, a South African neuropsychologist, is the founder of “neuropsychoanalysis.” This new field, with its own society and its own journal, proposes to use neuroimaging to confirm analytic theories. Its key idea is that subjective experience and the unconscious mind can be observed through neuroimaging.⁵ It is known that brain processes can be seen on brain imaging even before they have entered consciousness.³⁸ However, claims that neuroimaging validate Freud’s model of the unconscious can be based only on “cherry-picking” the literature. The observed correspondences are superficial and hardly support the complex edifice of psychoanalytic theory.

Solms³⁹ has also suggested that Freud’s ideas about dreams are consistent with neuroscience research based on rapid eye movement (REM) activity. This attempt to rescue a century-old theory met with opposition from dream researchers who consider Freud’s clinical speculations to be incompatible with empirical data.^40,41

The proposal to establish a discipline of neuropsychoanalysis also met with a mixed reception from traditional psychoanalysts, who did not want to dilute Freud’s wine with neuroscientific water.⁴² Neuroscientists, who are more likely to see links to psychology as lying in cognitive science,⁴³ have ignored this idea. In summary, neuropsychoanalysis is being used a way to justify long-standing models, without attempting to find something new or to develop an integration of perspectives on psychology.

However, Eric Kandel,⁴⁴ influential in the light of his Nobel Prize for the study of the neurochemistry of memory, has taken a sympathetic view of the use of biological methods to study psychoanalytic theory. Kandel had wanted to be an analyst before becoming a neuroscientist.⁴⁵ But Kandel, who does not actively practice psychiatry, may be caught in a time warp, unaware that psychoanalysis has been overtaken by competitors in the field of psychotherapy.

Another attempt to reconcile psychoanalysis with science has come from the literature on neuroplasticity.⁴⁶ It is now known that neurogenesis occurs in some brain regions (particularly the hippocampus) during adulthood and that neural connections undergo modification in all parts of the brain. There is also evidence that CBT can produce brain changes that are visible using imaging.⁴⁷ These findings have not been confirmed in psychoanalytic therapies. However, Norman Doidge, a Canadian psychoanalyst, has argued that psychoanalysis can change the brain.⁴⁸ This may be the case for all psychotherapies. However, more recently, Doidge⁴⁹ has claimed that mental exercises can reverse the course of severe neurological and psychiatric problems, including chronic pain, stroke, multiple sclerosis, Parkinson’s disease, and autism. While these books have been best-sellers, most of their ideas in the second volume,⁴⁹ based on anecdotes rather than on clinical trials, have had little impact in medicine. This story underscores the difficulty of reconciling the perspectives and methods of psychoanalysis with scientific methods based on empirical testing.

Psychoanalysis and the Humanities

Psychoanalysis claimed to be a science but did not function like one. It failed to operationalize its hypotheses, to test them with empirical methods, or to remove constructs that failed to gain scientific support.¹ In this way, the intellectual world of psychoanalysis more closely resembles the humanities. Today, with few psychiatrists or clinical psychologists entering psychoanalytic training, the door has been opened to practitioners with backgrounds in other disciplines, including the humanities.

This trend is related to a hermeneutic mode of thought,⁵⁰ which focuses on meaningful interpretations of phenomena, rather than on empirical testing of hypotheses and observations. Since the time of Freud, the typical psychoanalytic paper has consisted of speculations backed up with illustrations, similar to the methods of literary theory and criticism.

One model currently popular in the humanities is “critical theory.”⁵¹ This postmodernist approach uses Marxist concepts to explain phenomena ranging from literature to politics. It proposes that truth is entirely relative and often governed by hidden social forces. In its most radical form, in the work of Michel Foucault,⁵² critical theory and postmodernism take an antiscience position, denying the existence of objective truth and viewing scientific findings as ways of defending the “hegemony” of those in power.

Some humanist scholars have adopted the ideas of Jacques Lacan, a French psychoanalyst who created his own movement and whose eccentric clinical practice resembled that of a cult leader.⁵³ Moreover, recruitment of professionals and academics with no training in science could lead to an increasing isolation of the discipline. While only a few contemporary psychoanalysts have embraced postmodernism, the humanities have made use of psychoanalytical concepts for their own purposes as a way of understanding literature and history.

Conclusions

In 2009, the British Journal of Psychiatry published a debate about whether the journal should accept psychoanalytic case reports.⁵⁴ The debate pitted a biologist, Lewis Wolpert, against a psychoanalyst-researcher, Peter Fonagy. Wolpert argued that psychoanalytic case reports should be excluded because they are in no way scientific. Fonagy, while conceding some of his opponent’s points, defended analysis by pointing out that research is possible and is now beginning to be conducted. But while Fonagy himself is committed to empiricism, he represents a very small minority in a field that lacks that commitment.

Unfortunately, the modern revisions of psychoanalysis do not offer a coherent response to critics. It is difficult to see how any of the current responses to criticism can save psychoanalysis from a continued and lingering decline. Analysis has separated itself from psychiatry and psychology by teaching its method in stand-alone institutes. The field may only survive if it is prepared to dismantle its structure as a separate discipline and rejoin academia and clinical science.

Whatever its limitations, psychoanalysis left an important legacy to psychiatry. It taught a generation of psychiatrists how to understand life histories and to listen attentively to what patients say. In an era dominated by neuroscience, diagnostic checklists, and psychopharmacology, we need to find a way to retain psychotherapy, whose basic concepts can be traced back to the work of Freud, as part of psychiatry.⁵⁵