Psychiatry Tabulated

Fifteen major perspectives on psychiatric disorders, compared across ten dimensions. Rows are ordered from ontologically realist (top) to anti-realist (bottom); the left border of each row encodes its position on that continuum.

Perspective	Disorder ontology	Primary locus	Causal emphasis	Nosological stance	Treatment implication	Empirical grounding	Distress required?	Unit of intervention	Stance on comorbidity	Key figures
Biomedical	Disease entity; natural kind	Brain	Genetic, neurobiological	Categorical; natural kinds	Pharmacology, somatic	Strong (Rx); weak (etiology)	In principle no; in practice yes via clinical significance criterion	Individual biology	Genuine distinct entities	Kraepelin, Griesinger
RDoC	Circuit dysfunction; real but dimensional	Neural circuits	Biological / dimensional	Dimensional; sets DSM categories aside for research purposes — explicitly not offered as a clinical diagnostic system	Targeted bio + behavioral	Emerging	No	Neural circuit	Artifact of categorical carving; circuits cut across DSM boundaries	Insel, Cuthbert
Computational	Aberrant inference/computation; algorithm-level real	Inferential processes (Marr's algorithmic level)	Predictive coding, RL, Bayesian inference	Process-level; transdiagnostic mechanisms	Mechanism-targeted; computational-behavioral	Emerging; limited clinical translation so far	No; computational dysfunction sufficient	Inferential / computational process	Shared computational signatures across categories; transdiagnostic	Friston, Huys, Montague, Adams
Phenomenological	Disturbed structure of subjectivity; experientially real	Lived experience	Constitutional; experiential structure	Descriptive realism (esp. endogenous psychoses)	Experience-attuned description and care	Philosophical; qualitative; structured instruments (e.g. EASE)	Altered subjectivity is central; distress typical but not strictly constitutive	Whole experiencing person	Distinct syndromal Gestalts (Copenhagen school); existential strands more holistic	Jaspers, Binswanger, Parnas, Sass
Evolutionary	Miscalibrated adaptation (Nesse) or harmful dysfunction (Wakefield) — see note	Organism in environment	Evolutionary mismatch; failure of designed function	Functional/dimensional (Nesse); realist (Wakefield)	Remove trigger; context shift; restore function	Theoretical; uneven empirical	Yes (Wakefield: harm is constitutive in HDA); functional signal possibly adaptive (Nesse)	Organism–environment fit	Often artifacts; shared mechanisms or pleiotropy expected	Nesse, Wakefield, Brüne, Del Giudice
HiTOP	Empirical covariance structure	Symptoms / traits	Psychometric / structural	Hierarchical dimensional	Transdiagnostic	Strong psychometric	Agnostic; empirical question	Spectrum / dimension	Largely artifacts of arbitrary categorical cuts; explained by shared higher-order dimensions	Kotov, Krueger, Watson, Achenbach
Network theory	Causal symptom loops; no latent entity	Symptom network	Mutual reinforcement	Rejects latent variable model	Symptom-targeted	Theoretical; emerging	Agnostic; symptoms are the target regardless	Individual symptom	Expected; overlapping networks with shared bridge symptoms	Borsboom, Cramer
Psychodynamic	Meaningful conflict; real but psychological	Unconscious mind	Intrapsychic; developmental	Categorical, loosening	Long-term psychotherapy	Mixed; contested	Not strictly; unconscious conflict can be asymptomatic	Intrapsychic structure	Common; reflect shared underlying character structure	Freud, Klein, Kernberg
Biopsychosocial	Multifactorial; expands explanatory frame rather than denying disease ontology	Person-in-context	Mixed (bio + psycho + social)	Pragmatic / integrative	Integrative	Theoretical; weak operationalization	Agnostic	Person-in-context	Agnostic; pragmatic	Engel
Cognitive / CBT	Dysfunctional processing; real but cognitive	Cognition, behavior	Schemas, conditioning	Dimensional tendency	Structured psychotherapy	Large RCT base, but effect-size inflation, allegiance effects, and benchmarking debates well documented (Cuijpers et al.)	Typically yes; distress or impairment required	Cognition and behavior	Often artifacts; transdiagnostic protocols increasingly preferred	Beck, Ellis, Clark, Barlow
Interpersonal / relational	Disrupted relational patterns	Relationship patterns	Attachment, interpersonal	Varies; transdiagnostic	IPT, relational therapy	Moderate RCT base	Yes; relational suffering is central	Dyad or relational system	Common; shared interpersonal/attachment patterns underlie multiple presentations	Sullivan, Klerman, Weissman, Bowlby
Social & cultural	Socially produced; culturally variable; some strands favor formulation over diagnosis	Social context	Poverty, trauma, inequality	Cultural relativism	Social intervention	Strong epidemiology	Yes, but culturally defined	Social structure / community	Categories are culturally variable; Western comorbidity patterns may not generalize	Kleinman, Brown & Harris, Hopper
PTMF	Coherent response to adversity, power, threat, meaning; rejects diagnostic ontology	Person within power structures	Power dynamics, threat, meaning-making	Replaces diagnosis with formulation	Narrative reconstruction; trauma-informed; systemic	Conceptual; case-formulation evidence	Yes, but reframed as understandable response, not disorder	Personal narrative within social structure	Concept rejected; presentations reflect coherent response patterns	Johnstone, Boyle
Humanistic	Blocked growth; skeptical of disorder talk	Whole person	Existential, contextual	Skeptical of nosology	Person-centered therapy	Weak formal evidence	Yes; subjective experience is primary	Whole person	Skeptical; categories fragment the person	Rogers, Yalom, Frankl
Critical psychiatry	Social label / instrument of control	Society, power	Political, institutional	Rejects or radically critiques	Rights-based; social	Sociological; philosophical	Irrelevant; the category itself is the problem	Political / institutional	Comorbidity reflects diagnostic proliferation, not clinical reality	Szasz, Laing, Bentall, Moncrieff

Color scheme — intellectual tradition

Blue Biomedical tradition

Teal Computational tradition

Purple Psychological traditions

Yellow Phenomenological tradition

Orange Evolutionary tradition

Green Social & cultural tradition

Red Critical tradition (Szasz and Laing wrote as anti-psychiatry; Bentall and Moncrieff reject that label)

Gray Integrative or meta-frameworks

Blue left border = stronger ontological realism

Red left border = stronger anti-realism / constructionism

On the continuum. The realism–anti-realism axis encodes whether a perspective treats psychiatric disorder categories as mind-independent natural kinds (realist) or as social, conventional, or constructed categories (anti-realist). Placement is approximate and reflects each tradition's central tendency rather than any individual theorist. The axis conflates several distinctions — realism about latent disease entities, categorical vs. dimensional structure, and social construction — which are not perfectly correlated. Network theory is the cleanest illustration: anti-realist about latent disease entities, yet realist about the causal structure among symptoms.

On the evolutionary row. The Wakefield/Nesse split is unusual in cutting across the very axis the table is sorted on. Wakefield's harmful-dysfunction analysis is among the most realist positions in philosophy of psychiatry (disorder = failure of an evolved function plus harm). Nesse's mismatch/functional framing treats many presentations as adaptive signals operating in evolutionarily novel environments — closer to dimensional and contextualist. A single placement misrepresents the tradition; #5 is a compromise.

On phenomenological psychiatry. Placed relatively high in realism. Jaspers founded much of descriptive realism; the Copenhagen school (Parnas, Sass) treats schizophrenia as a real ipseity disturbance with a specific phenomenological signature, operationalized in instruments such as EASE. Existential-phenomenological strands (Binswanger, early Laing) are less categorically realist; the row's central tendency lands toward realism about experiential structure.

On computational psychiatry vs. RDoC. Both are realist about mechanism, but at different Marr levels: RDoC commits to neural circuits (implementational level); computational psychiatry to inferential or algorithmic processes, often agnostic about the biological substrate that realizes them. The two are complementary rather than competing.

On PTMF. The Power Threat Meaning Framework (Johnstone & Boyle, 2018) is the most developed instance of a broader formulation-based move that also includes trauma-informed and liberation-psychology traditions. It sits between social/cultural realism about causes and critical-psychiatry rejection of diagnostic categories.

Citation

Persson, B. N. (2026). Psychiatry Tabulated [Comparative table]. https://bjorn-persson.github.io/visualizations/tabulated-psychiatry/

@misc{Persson2026psychiatrytabulated,
  author = {Björn N. Persson},
  year = {2026},
  title = {Psychiatry Tabulated},
  note = {Comparative table},
  url = {https://bjorn-persson.github.io/visualizations/tabulated-psychiatry/}}