Is psychoanalysis scientific?
The question is asked often — and just as often asked in the wrong place. Depending on the context, it means three different things: whether psychoanalytic treatments work, whether their theoretical concepts are sound, or whether they satisfy the criterion of a particular philosophy of science. This page differentiates what can be differentiated — and names what remains open.
What I mean by “psychoanalytically based treatment methods”
When this page speaks of psychoanalytically based treatment methods , what is meant are the three treatment formats of the German catalogue of guideline-approved treatment methods (Richtlinienverfahren) that are based on psychoanalytic theory. In my practice I work with all three:
Analytic psychotherapy (AP) — on the couch, at varying frequencies, from one session a week to high-frequency work. The analytic stance remains the same; what changes is the density of the encounter.
Modified analytic psychotherapy — seated rather than lying down, still with analytic method and stance. The setting is altered, the method the same.
Depth-psychologically founded psychotherapy (TP) — focally oriented, seated, low-frequency.
All three formats rest on the same theoretical foundation. They differ in setting and treatment focus, not in the underlying model. In the international research literature they run together under the term psychodynamic psychotherapy (PDT) — the designation under which the efficacy studies cited below are published, covering both short-term focused treatments (STPP) and longer-term, structurally oriented treatments (LTPP).
Do psychoanalytic treatments work?
That is an empirical question — and it can be answered. Since the 2000s there has been a substantial research base: meta-analyses, randomised controlled trials, umbrella reviews. The result is clearer than the public debate would suggest.
Efficacy is established
Several independent meta-analyses document consistent efficacy in depression, anxiety disorders, personality disorders and somatoform disorders. Effect sizes against waiting-list or placebo conditions lie in the medium to large range — clearly above the values documented for antidepressants in comparable trials (Turner et al., 2008). A pre-registered umbrella review awarded the highest level of recommendation under GRADE methodology.
Leichsenring et al., 2023, World Psychiatry; Abbass et al., 2020, Psychotherapy and PsychosomaticsEquivalence, not inferiority
In direct comparison with cognitive behavioural therapy, psychoanalytic treatments show no significant inferiority in any available meta-analysis. A formal equivalence test demonstrated parity across 23 randomised trials with over 2,700 patients. A superiority of CBT in certain anxiety disorders has been documented (Tolin, 2010), but has been considerably qualified by newer, methodologically stricter studies.
Steinert et al., 2017, American Journal of Psychiatry; Lilliengren, 2023, Psychoanalytic PsychotherapyImprovement after the end of therapy
A specific feature of psychoanalytic treatments: patients often continue to improve after treatment ends. This “sleeper effect” has been replicated in several studies, among them the Munich Psychotherapy Study with its three-year follow-up. It is theoretically consistent with an approach that aims at structural change — defence patterns, relational representations, mentalising capacity. More recent analyses estimate the effect size more conservatively than older studies.
Shedler, 2010, American Psychologist; Huber et al., 2013, Psychiatry; Woll & Schönbrodt, 2019, European PsychologistThe relationship works — but not without method
Across all treatment methods, the therapeutic alliance is one of the strongest predictors of treatment success (r = .278 across 295 studies, > 30,000 patients). From this, a short-circuited reading is often derived: “the method does not matter, as long as the relationship is good”. That is not what the research says: in every serious study, alliance operates as a relationship with a therapist who has command of her method — it arises from concrete work on transference, patterns, focus, and what the patient brings. Countertransference management and the repair of alliance ruptures — genuinely psychoanalytic concepts — have been officially recognised by the American Psychological Association as empirically supported factors of change.
Flückiger et al., 2018, Psychotherapy; Norcross & Lambert, 2019; Wampold, 2015, World PsychiatryWhat psychoanalytic treatments are particularly good at
Effect sizes show that the treatments work. They do not show what they are particularly made for. Whoever stops at the trials easily overlooks what distinguishes this therapeutic tradition clinically from others — and why precisely the areas in which randomised trials are hard to conduct are its true terrain.
The symptom in its history
What distinguishes psychoanalytic treatments from symptom-centred ones: symptoms are not treated in isolation but understood as the expression of a person — of her conflicts, relational patterns, history. Nowhere does this connection show more clearly than in personality disorders, where symptom and structure are inseparably interlocked. It is precisely for this that treatments such as TFP (Transference-Focused Psychotherapy, after Kernberg and Clarkin) and MBT (Mentalization-Based Treatment, after Bateman and Fonagy) were developed out of the psychoanalytic tradition — they are today among the best-evaluated treatments for borderline personality disorder. But the principle reaches further: wherever a symptom is embedded in a person — and clinically that is the rule — it is the psychoanalytic treatment that works on this embedding as well.
Bateman & Fonagy, 2009, American Journal of Psychiatry; Clarkin et al., 2007, American Journal of PsychiatryMore than goal clarification
Psychoanalytic treatments reckon with something that many other approaches pass over: that change cannot engage only what is conscious. Patients want to be rid of their symptom — and unconsciously sabotage that goal at the same time. They have conflicts that are not accessible to them. They use defence mechanisms whose function they do not see through. Symptoms often serve a compensatory function — they are not “the problem” but a failed solution. Whoever merely clarifies motives, formulates goals and proceeds step by step works at the surface of what people can make consciously accessible to themselves. Psychoanalytic treatments work on what eludes direct control — resistance, transference, unconscious conflicts — and take seriously the developmental time needed before something old can loosen.
Hoglend et al., 2008, American Journal of Psychiatry; Cramer, 2006, Protecting the SelfThe sleeper effect
The effect of psychoanalytic treatments frequently continues to grow after therapy ends — a sleeper effect described in several meta-analyses. Patients profit from what they have understood about themselves long after the last session. This finding speaks for the depth of the change, not merely its speed.
Shedler, 2010, American Psychologist; Abbass et al., 2014, Cochrane Database of Systematic Reviews; Huber et al., 2013, PsychiatryMore than symptom remission
The aim is not solely the disappearance of a symptom, but a change in experience, in the regulation of self and relationships. That makes the therapy more demanding to measure — and at the same time makes it what many patients are actually seeking when they commit to a longer psychoanalytic treatment. This became empirically tangible in studies that explicitly measure structural change — for instance via the OPD structure axis or via attachment representations and reflective functioning.
Grande et al., 2009, Psychotherapy Research (Heidelberg-Berlin Study); Levy et al., 2006, Journal of Consulting and Clinical PsychologyPut differently: where a symptom stands on its own and can be readily manualised, the world of randomised trials is full of good answers. As soon as the symptom is part of a history, of a relational structure, of a self-organisation — there begins the terrain on which psychoanalytic treatments have their strengths. It is precisely in this tension that I work: on symptoms that cannot be resolved without reading their history alongside them.
Are the theoretical concepts sound?
That a therapy works does not prove that it works for the reasons it postulates (Kazdin, 2007). That is a separate question — and one that can be answered independently of psychotherapy research, through experimental psychology, neuroscience and developmental psychology.
The unconscious
Unconscious information processing has been a consensus of experimental cognitive psychology since the 1980s. Implicit memory, automatic evaluation processes and unconscious goal activation have been replicated hundreds of times. Freud’s model has been described as the most specific and detailed model of the unconscious to this day.
Kihlstrom, 1987; Bargh & Morsella, 2008, Perspectives on Psychological Science; Westen, 1998, Psychological BulletinRepression
Anderson and Green (2001) demonstrated in Nature that active suppression of memories reduces their retrievability — dose-dependently, robust against alternative explanations, with explicit reference to Freud’s concept of repression. It is among the most-cited studies in experimental memory research.
Anderson & Green, 2001, NatureDefence mechanisms
Cramer’s longitudinal research documents a developmental hierarchy of defence mechanisms that correlates with psychopathology and changes over the course of therapy. Vaillant’s Harvard Grant Study shows across decades: maturity of defences predicts long-term mental health.
Cramer, 2006, Protecting the Self; Vaillant, 1977Transference
Early relational experiences form mental prototypes that are unconsciously transferred onto new relationships. Different significant others activate different cognitive, emotional and motivational patterns — the experimental foundation of the concept of transference.
Andersen & Cole, 1990; Westen, 1998, Psychological BulletinChildhood & personality
That early relational experiences shape personality is well established. Freud’s specific stage theory (oral, anal, phallic) is not. Modern attachment research provides the more valid framework and confirms the basic psychoanalytic idea in updated form.
Main et al., 1985; Westen, 1998, Psychological BulletinDrive theory & death drive
Freud’s metapsychological speculations — the hydraulic drive model, the death drive, the specific Oedipal mechanics — are not empirically supported. Modern psychoanalysis has itself revised them since the 1970s. Whoever criticises them today is criticising the discipline’s history, not its current state.
Westen, 1998, Psychological BulletinThe picture is different from what one might expect: the basic architecture of psychoanalytic theory — the unconscious, defence, relational schemas, childhood influences — is better supported experimentally than most comparable psychological theories (Westen, 1998). What has been refuted, psychoanalysis has itself revised. The critics frequently attack a theory that has not existed in that form for fifty years.
Can psychoanalysis be a science?
The philosophically most demanding question — and the most interesting. Popper, Grünbaum and others formulated serious objections. They deserve serious answers.
“Psychoanalysis is not falsifiable — for every behaviour there is an explanation after the fact.” (Popper, 1963)
That hits a real weakness — the tendency of clinical interpretation to arrange case material so that it confirms one’s own theory. But it hits a practice, not a logical necessity. Whoever operationalises defence mechanisms and tests them longitudinally (Cramer, 2006) is formulating falsifiable hypotheses. Whoever tests equivalence via a registered testing procedure (Steinert et al., 2017) works to the same standard as any other psychotherapy research. Popper’s critique describes an epistemic risk — one, however, that all clinical disciplines share.
“Therapeutic success does not prove the theory — it could all be suggestion.” (Grünbaum, 1984)
Logically correct — and philosophically the sharpest critique of psychoanalysis. Grünbaum demanded extra-clinical validation: evidence gained independently of the therapeutic situation. Psychoanalytic research has taken precisely this path. The experimental studies on repression (Anderson & Green, 2001), defence (Cramer, 2006) and the unconscious (Bargh & Morsella, 2008) are the answer to Grünbaum’s demand. Grünbaum’s critique has, paradoxically, strengthened the empirical tradition of psychoanalysis.
“CBT is evidence-based — psychoanalysis is not.”
This claim was once partly justified — it no longer is. Lilliengren (2023) identified 298 independent randomised controlled trials of psychoanalytic therapies, 41% of them from the past decade. The evidence base is smaller than that of CBT, but suffices for the highest GRADE level of recommendation (Leichsenring et al., 2023). In the German psychotherapy debate, the term “evidence-based” has been used politically as a synonym for CBT — that does not correspond to the scientific state of affairs.
“Psychoanalysis is an art of interpretation, not a natural science.”
The opposition is too crude. Psychoanalysis draws on quantitative efficacy research, experimental testing of its concepts, qualitative single-case analysis and clinical experience in the analytic situation. None of these sources suffices on its own. Warsitz and Küchenhoff (2015) speak of an epistemological pluralism: psychoanalytic knowledge cannot be fully reduced to a single paradigm — yet on precisely this basis they ground its claim to scientific standing rather than abandoning it.
What this means for your treatment
If you begin psychotherapy with me, we work within the psychoanalytically based treatment methods of the German catalogue of guideline-approved treatment methods (Richtlinienverfahren) — depending on indication and setting: depth-psychologically founded, modified analytic, or analytic. These are treatment formats whose efficacy is documented by hundreds of studies and recognised as empirically supported by the international research community (Leichsenring et al., 2023).
The particular strength of this approach lies in the fact that it does not only treat symptoms but understands the patterns underlying them: how you shape relationships, how you deal with distressing feelings, which convictions about yourself you carry from early experiences. This work takes time — and often unfolds its effect beyond the end of therapy (Shedler, 2010).
At the same time, scientific honesty means naming openly where the limits lie. Psychoanalytic treatments are not the best-researched treatment option for every disorder. In my practice I discuss this with you: what the research shows for your situation, what alternatives exist, and why I nonetheless — or precisely for that reason — work psychoanalytically.
The ease with which the sentence “psychoanalysis is unscientific” passes the lips reveals less about psychoanalysis than about those asking — about what we allow to count as knowledge, and about the effort it costs to withstand complexity.
— Florian LampersbergerThree questions, three answers
Does it work? Yes — comparably to other empirically supported treatments, with particular strengths in long-term effects and structural change. The evidence base has grown and has reached the highest international level of recommendation (Leichsenring et al., 2023).
Are the concepts sound? The core concepts — the unconscious, defence, relational schemas — are better supported experimentally than most psychological theories (Westen, 1998). What does not hold, psychoanalysis has itself revised.
Is it a science? It commands a substantial tradition of empirical research and an epistemological particularity that cannot be fully translated into experimental designs. Taking both seriously at once is the appropriate form of scientific honesty.
Selected sources for this page in alphabetical order (APA 7th ed.)
- Abbass, A. A., Kisely, S. R., Town, J. M., Leichsenring, F., Driessen, E., De Maat, S., Gerber, A., Dekker, J., Rabung, S., Rusalovska, S., & Crowe, E. (2014). Short-term psychodynamic psychotherapies for common mental disorders. Cochrane Database of Systematic Reviews, 7, CD004687.
- Abbass, A., Town, J., Holmes, H., Luyten, P., Cooper, A., Russell, L., Lumley, M. A., Schubiner, H., Allinson, J., Bernier, D., De Meulemeester, C., Kroenke, K., & Kisley, S. (2020). Short-term psychodynamic psychotherapy for functional somatic disorders: A meta-analysis of randomized controlled trials. Psychotherapy and Psychosomatics, 88(5), 265–275.
- Anderson, M. C., & Green, C. (2001). Suppressing unwanted memories by executive control. Nature, 410, 366–369.
- Andersen, S. M., & Cole, S. W. (1990). „Do I know you?“: The role of significant others in general social perception. Journal of Personality and Social Psychology, 59(3), 384–399.
- Bargh, J. A., & Morsella, E. (2008). The unconscious mind. Perspectives on Psychological Science, 3(1), 73–79.
- Bateman, A., & Fonagy, P. (2009). Randomized controlled trial of outpatient mentalization-based treatment versus structured clinical management for borderline personality disorder. American Journal of Psychiatry, 166(12), 1355–1364.
- Clarkin, J. F., Levy, K. N., Lenzenweger, M. F., & Kernberg, O. F. (2007). Evaluating three treatments for borderline personality disorder: A multiwave study. American Journal of Psychiatry, 164(6), 922–928.
- Cramer, P. (2006). Protecting the self: Defense mechanisms in action. Guilford Press.
- Flückiger, C., Del Re, A. C., Wampold, B. E., & Horvath, A. O. (2018). The alliance in adult psychotherapy: A meta-analytic synthesis. Psychotherapy, 55(4), 316–340.
- Grande, T., Dilg, R., Jakobsen, T., Keller, W., Krawietz, B., Langer, M., Oberbracht, C., Stehle, S., Stennes, M., & Rudolf, G. (2009). Structural change as a predictor of long-term follow-up outcome. Psychotherapy Research, 19(3), 344–357.
- Grünbaum, A. (1984). The foundations of psychoanalysis: A philosophical critique. University of California Press.
- Høglend, P., Bøgwald, K. P., Amlo, S., Marble, A., Ulberg, R., Sjaastad, M. C., Sørbye, Ø., Heyerdahl, O., & Johansson, P. (2008). Transference interpretations in dynamic psychotherapy: Do they really yield sustained effects? American Journal of Psychiatry, 165(6), 763–771.
- Huber, D., Henrich, G., Clarkin, J., & Klug, G. (2013). Psychoanalytic versus psychodynamic therapy for depression: A three-year follow-up study. Psychiatry, 76(2), 132–149.
- Kazdin, A. E. (2007). Mediators and mechanisms of change in psychotherapy research. Annual Review of Clinical Psychology, 3, 1–27.
- Kihlstrom, J. F. (1987). The cognitive unconscious. Science, 237(4821), 1445–1452.
- Leichsenring, F., Abbass, A., Hilsenroth, M. J., Luyten, P., Munder, T., Rabung, S., & Steinert, C. (2023). The status of psychodynamic psychotherapy as an empirically supported treatment for common mental disorders — an umbrella review based on updated criteria. World Psychiatry, 22(2), 286–304.
- Levy, K. N., Meehan, K. B., Kelly, K. M., Reynoso, J. S., Weber, M., Clarkin, J. F., & Kernberg, O. F. (2006). Change in attachment patterns and reflective function in a randomized control trial of transference-focused psychotherapy for borderline personality disorder. Journal of Consulting and Clinical Psychology, 74(6), 1027–1040.
- Lilliengren, P. (2023). A comprehensive overview of randomized controlled trials of psychodynamic psychotherapies. Psychoanalytic Psychotherapy, 37(2), 117–140.
- Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and adulthood: A move to the level of representation. Monographs of the Society for Research in Child Development, 50(1–2), 66–104.
- Norcross, J. C., & Lambert, M. J. (Eds.). (2019). Psychotherapy relationships that work: Vol. 1. Evidence-based therapist contributions (3rd ed.). Oxford University Press.
- Popper, K. R. (1963). Conjectures and refutations: The growth of scientific knowledge. Routledge.
- Shedler, J. (2010). The efficacy of psychodynamic psychotherapy. American Psychologist, 65(2), 98–109.
- Steinert, C., Munder, T., Rabung, S., Hoyer, J., & Leichsenring, F. (2017). Psychodynamic therapy: As efficacious as other empirically supported treatments? A meta-analysis testing equivalence. American Journal of Psychiatry, 174(10), 943–953.
- Tolin, D. F. (2010). Is cognitive-behavioral therapy more effective than other therapies? A meta-analytic review. Clinical Psychology Review, 30(6), 710–720.
- Turner, E. H., Matthews, A. M., Linardatos, E., Tell, R. A., & Rosenthal, R. (2008). Selective publication of antidepressant trials and its influence on apparent efficacy. New England Journal of Medicine, 358, 252–260.
- Vaillant, G. E. (1977). Adaptation to life. Little, Brown.
- Wampold, B. E. (2015). How important are the common factors in psychotherapy? An update. World Psychiatry, 14(3), 270–277.
- Warsitz, R.-P., & Küchenhoff, J. (2015). Psychoanalyse als Erkenntnistheorie. Kohlhammer.
- Westen, D. (1998). The scientific legacy of Sigmund Freud: Toward a psychodynamically informed psychological science. Psychological Bulletin, 124(3), 333–371.
- Woll, C. F., & Schönbrodt, F. D. (2019). A series of meta-analytic tests of the efficacy of long-term psychoanalytic psychotherapy. European Psychologist, 25(1), 51–72.