Who is allowed to be confident? Psychiatry,humility,and epistemic double standards

Asymmetry and double standards

In psychiatric research and clinical practice, patients’ confidence, certainty and conviction are treated as clinically meaningful phenomena to be measured, interpreted and potentially corrected or treated, particularly in studies of metacognition, insight and overconfidence in the context of psychosis.^1–3 This literature may identify meaningful metacognitive phenomena. However, it also raises an unresolved epistemic question: why are patients’ confidence and conviction so readily available for clinical interpretation, while psychiatry’s own institutional confidence is less often subjected to comparable scrutiny?

This article therefore asks the following question: who is permitted to hold confidence, certainty or conviction without those states being transformed into clinical evidence of pathology, and who gets to draw the line between confidence, overconfidence and pathology? The argument is not that psychiatric expertise should be abandoned, nor that patients’ accounts are always correct. Rather, it is that professional confidence becomes ethically consequential when it shapes diagnosis, coercion, access to care and credibility, and should therefore be subject to reciprocal scrutiny.

The question matters because psychiatry is itself marked by uncertainty. Psychiatric classification systems remain contested, heterogeneous, culturally situated and historically unstable. This provisionality is central to Mullen’s discussion of Jaspers and phenomenology: phenomenological description should remain open to revision rather than becoming a self-authorising substitute for clinicopathological knowledge. ⁴ Contemporary critiques of the Diagnostic and Statistical Manual of Mental Disorders (DSM) also locate classification within power, institutional authority and global epistemic politics: Jackman et al. argue that DSM-VI revision risks reinforcing biomedical dominance unless it addresses plural epistemologies, rights-based care and social, cultural, spiritual, economic, commercial and political realities. ⁵ Cosgrove similarly argues that under-disclosed financial conflicts of interest in the development of diagnostic constructs within the DSM remain a persistent problem and may undermine public trust, reporting that 60% of eligible Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR) panel members received industry payments totalling $14.2 million. ⁶ Gureje’s account of psychiatric nosology, Solomon’s argument for conceptual humility, and Horien and Bommersbach’s reflections on humility in psychiatric training converge on the need to recognise the limits of psychiatric knowledge and resist overconfidence.^7–9 Brenner et al. extend this implication into practice by framing humility as an educational, moral, structural and therapeutic necessity. ¹⁰ If psychiatry asks patients to reflect on their confidence, certainty and insight, it must also acknowledge the provisionality of its own categories, explanations and clinical authority.

Acknowledging expertise does not remove the possibility of epistemic injustice. Sakakibara argues that epistemic injustice can arise within the professional-client, doctor-patient and psychiatrist-patient relationships precisely because professional roles shape what is treated as clinically relevant knowledge. ¹¹ Kidd, Spencer and Harris similarly argue that epistemic justice requires clinicians to scrutinise how doubt is applied, how credibility is assigned, and whether patients’ testimony is given meaningful uptake within clinical reasoning. ¹² Ballesteros further shows that psychiatric objectivity, when reduced to standardisation, may treat patients as passive informants rather than epistemic agents. ¹³ These arguments support a sharper claim: psychiatric authority is not illegitimate because it is authoritative, but it is ethically incomplete when its authority is exempted from scrutiny and critique.

These calls for humility sit uneasily beside psychiatric literatures that frame patients’ confidence as clinically significant. Balzan reviewed overconfidence in psychosis as a possible foundation of delusional conviction, Moritz et al. examined confidence in perceptual judgements among people with schizophrenia, and Hoven et al. framed abnormalities of confidence as a transdiagnostic psychiatric concern.^1–3 The issue is not whether such research has value. Rather, the asymmetry lies in the direction of scrutiny. Patients’ confidence is operationalised as data to be measured, interpreted and potentially corrected, while psychiatry’s confidence in its own categories, interpretations and interventions is more readily framed as expertise, certainty and clinical authority.

The problem is not that confidence should never be studied. It is that psychiatry often lacks reciprocal epistemic accountability. Crichton et al. described epistemic injustice in psychiatry as a process through which patient testimony is discounted, and Kidd, Spencer and Carel locate such injustice across psychiatric research, diagnostic classification and clinical practice.^14,15 Isern-Mas and Hannikainen provide empirical nuance, finding that psychiatric patients’ testimony can be judged less credible than comparable testimony from non-patients, although this effect should not be overstated as universal or uniform. ¹⁶ In such contexts, confidence is not merely assessed in an objective manner. It is governed and policed.

Diagnostic framing can also shape whose confidence is treated as meaningful, excessive, defensive, manipulative or pathological. Watts argues that borderline personality disorder can produce epistemic injustice by structuring how testimony is heard, while critical work on the diagnostic construct shows how psychiatric labels can discredit lived experience knowledge, justify care withdrawal and obscure sensory, cultural and structural determinants of distress.^17,18 The relevance here is not borderline personality disorder as a standalone case study, but the broader concern that psychiatric diagnostic language can alter the interpretive status of what a person says, asks for, refuses, needs, or insists upon.

This double standard also appears in psychiatry’s exclusionary language. Beale argues that mental health services can become organised around denying care while convincing themselves that exclusion is clinically appropriate. ¹⁹ Similarly, Elwyn et al. show how eating disorder discourses of terminality and futility can write people’s futures through clinical assumptions about recovery, prognosis and treatability. ²⁰ In both examples, professional confidence may determine whether distress is interpreted as legitimate need, risk, futility, manipulation, non-compliance or therapeutic inappropriateness.

Treatment discourse provides a further illustration. Leichsenring et al. found that effect sizes for psychotherapies and pharmacotherapies across adult mental disorders are often small and frequently affected by high risk of bias. ²¹ Levine and Fink’s critique is relevant here because they argue that evidence-based psychiatry rests on unstable foundations: diagnostic categories may overlap and shift rather than map onto discrete conditions, rating-scale outcomes may treat fractional symptom reductions as meaningful clinical improvement, and selective reporting or unpublished negative trials may distort the apparent strength of evidence. ²² Psychotherapy should not be treated as an inherently benign default: Curran et al.'s synthesis of qualitative research and service-user testimony identified adverse processes across the therapy journey, including unresolved therapeutic alliance rupture, rigidity, over-control, misuse of power, blaming and invalidation, boundary violations, lack of cultural validity, narrow options and clients feeling silenced, devalued or disempowered. ²³

Although sometimes conflated, treatment benefit or effectiveness and treatment safety are distinct questions, because symptom reduction can coexist with iatrogenic harm, and harms may remain poorly defined, inconsistently monitored and inadequately reported. ²⁴ Dropout and disengagement may also reflect relational rupture, emotional unsafety, coercion, invalidation, cultural incongruence, epistemic injustice or iatrogenic harm rather than poor motivation, manipulation, or non-compliance. ²⁵ Downs further argues that iatrogenic harm from mental healthcare can become a barrier to later treatment when it is not meaningfully acknowledged or responded to (e.g. apology and repair-oriented practice), leading to patients disengaging from and actively avoiding mental health care services. ²⁶

Electroconvulsive therapy (ECT) shows how institutional confidence can organise the unbalanced and hyperbolic representation of interventions’ effectiveness and safety. Although ECT is often promoted as highly effective, acute response should not be confused with durable benefit: relapse after ECT is common, with meta-analytic evidence showing high relapse rates after ECT and newer follow-up evidence reporting relapse in more than half of participants by 6 months.^27,28 However, patients report not being informed about the high risk of relapse associated with ECT.^29–31 Cognitive risks require explicit monitoring, honest communication and meaningful informed consent, including discussion of the possibility of persistent autobiographical memory loss; patient information materials may inadequately represent uncertainty and long-term effects.^29,30 Shipwright and Murphy’s narrative analysis of people reporting long-term adverse effects after ECT further documents experiences of memory and cognitive worsening, loss of ability, lack of follow-up, and of being silenced and dismissed when expressing experiences of iatrogenic harm. ³¹ The ethical concern is not reducible to whether ECT is effective or ineffective in any absolute sense. It is how professional narratives may foreground acute response while backgrounding relapse, cognitive deterioration, coercion, uncertainty and lived experience accounts of iatrogenic harm, thereby undermining the transparency and honesty required for informed consent. This illustrates the broader epistemic asymmetry: psychiatry’s confidence can organise treatment and diagnostic legitimacy, while patients’ confidence in their own experiences of iatrogenic harm may struggle to be recognised as valid, credible and actionable.

Research about cultural and socio-political factors in mental health reinforces this concern. Jadhav critiqued cultural psychiatry for reproducing Euro-American assumptions while claiming global relevance. ³² Trinh et al., Kolade, and Konidaris and Petrakis each argue that cultural humility requires ongoing self-reflection, attention to power and a willingness to revise the expert role.^33–35 Hilton similarly warns that historical reflection should produce humility rather than present-day complacency. ³⁶ Tamatea, Ward and Day extend this point to cognitive behavioural therapy (CBT), arguing that CBT is not politically or culturally neutral and that work with Indigenous peoples in Aotearoa-New Zealand and Australia requires attention to systemic oppression, epistemic divergence and Indigenous ways of being. ³⁷ Jackman et al.'s global lived expertise perspective on DSM-VI revision further argues that psychiatric classification can obscure local, spiritual, relational, political and social understandings of distress when Western, Eurocentric biomedical categories are treated as universal. ⁵ Taken together, this literature shows that psychiatric confidence is not only a matter of evidence, but also of whose knowledge systems, cultural assumptions and political realities are permitted to define distress, recovery and appropriate care.

The corrective is not merely to include lived experience symbolically. Hawke et al. argue that mental health institutions and organisations must actively value researchers with lived experience and challenge stigma, while Banfield and Palmer warn that lived experience involvement remains vulnerable to tokenism when traditional powerholders retain control.^38,39 Hatchman et al. found that integration of lived experience workers requires role clarity, trust, accountability and attention to discipline-based defensiveness. ⁴⁰ Speyer and Ustrup add that dissensus within experiential knowledge should be treated as a source of rigour rather than a problem to be resolved, while Downs argues that mental health science must integrate lived experience with biological, clinical and translational evidence if research is to become meaningful in real life.^41,42 Jackman et al. similarly caution that institutional invitations to lived experience participation may remain insufficient when those most marginalised by poverty, racism, incarceration, colonial legacies and displacement are not meaningfully included. ⁵ Neurodiversity-affirming psychotherapy scholarship similarly argues that competence frameworks should explicitly incorporate iatrogenesis prevention, epistemic humility, relational safety and lived experience involvement. ⁴³ These contributions support an epistemically plural approach in which experiential knowledge is not treated as supplementary testimony to be filtered through professional authority, but as a valid source of interpretation, critique and knowledge production alongside clinical, biological, social, and cultural forms of knowledge.

Capacity, risk and clinical action under uncertainty

The argument presented in this article does not require claiming that patients’ views are always dismissed or that patients are always correct. Kious, Lewis and Kim argue that psychiatrists often rely on patient testimony and that clinical scepticism may sometimes be warranted. ⁴⁴ That concern should be taken seriously. However, capacity and credibility are not the same. A person may formally retain decision-making capacity while their testimony, reasons, refusals or accounts of iatrogenic harm are still discounted through diagnostic framing, assumptions about insight or risk discourse.^{11,13,17,29–31} The question is therefore not only whether patients are legally permitted to decide, but whether their decisions are treated as valid, credible and actionable within clinical interpretation.

Furthermore, epistemic reciprocity is not role equivalence. Psychiatrists and patients do not occupy identical institutional positions, and psychiatrists may have duties concerning care, risk and public safety that patients do not have.^11,44 However, those duties do not make professional confidence epistemically neutral. Where psychiatric judgement can shape detention, coercion, treatment access, diagnosis and credibility, the need to examine the evidential basis and limits of that judgement is heightened rather than reduced.^11,12 This is especially important in coercive contexts: Brayley et al. found that only 40% of South Australian involuntary admission forms explicitly addressed all required legal criteria for psychiatric detention, illustrating the need for transparent, legally grounded and reviewable reasoning when liberty is at stake. ⁴⁵

Acknowledging uncertainty does not require clinical paralysis. Clinicians may sometimes need to act despite incomplete knowledge, particularly where immediate risk is present.^12,44 The point is that uncertainty should not be hidden by the authority of practice. If an intervention is recommended because it may reduce risk despite limited, indirect or contested evidence, that rationale should be transparent and honestly communicated as part of informed consent, and the recommendation should be weighed against coercion, iatrogenic harm, therapeutic alliance rupture, false-positive risk, future avoidance of care and the person’s own account of what has helped or harmed them.^23,24,26 Risk-reduction claims also require specificity: in the context of ECT, reductions in depressive symptoms or suicidal ideation should not be overstated as established evidence that ECT prevents suicide death.^46,47 Sandsten et al.'s study of general psychiatric inpatient care for anorexia nervosa is especially instructive: participants described care as life-preserving, but not necessarily supportive of living, recovery, meaning or personhood. ⁴⁸ In addition, patients reported being dismissed and labelled as ‘difficult’ or ‘manipulative’ when they expressed frustration or disagreement with the treatment. ⁴⁸

Finally, the claim that something is better than nothing itself requires scrutiny. Possible benefit does not establish safety, and psychological or psychiatric intervention should not be treated as inherently benign when iatrogenic harms are underdefined, undermeasured, underreported or disbelieved.^23,24,43 Apparent compliance or early behavioural improvement may also rerflect masking, appeasement or pressure to meet clinician expectations rather than authentic participation or sustainable wellbeing. ⁴³ A psychiatry committed to humility should therefore not ask only whether an intervention might help, but also whether it might harm, whether informed consent has included clear and honest discussion of uncertainty, limitations, risks, and alternatives, whose conceptualisations of healing and wellbeing are being measured and prioritised, and what forms of repair and accountability follow when iatrogenic harm occurs.

Practical applications

Epistemic reciprocity has practical implications for individual clinicians, services, legal safeguards, research and education. At the individual level, clinicians should distinguish evidence, inference, clinical judgement, local convention and speculation when making recommendations, and should state when evidence is uncertain, indirect, contested or extrapolated.^11,12 This is particularly important where patients express disagreement, mistrust, refusal or self-advocacy. Such responses should not automatically be interpreted as lack of insight, resistance or non-compliance before considering whether they reflect prior harm, treatment mismatch, cultural unsafety, communication needs or reasonable disagreement with the proposed intervention.^24–26 A practical reflective question follows: what evidence would make me revise my view?

Assessment, diagnosis and documentation also require practical safeguards. Standardised tools can support consistency, but they should not replace idiographic understanding or patient meaning-making.^13,43 Diagnostic labels should be recognised as credibility-altering as well as descriptive, particularly where disagreement, distress or requests for accommodation are read through assumptions of manipulation, instability or lack of insight.^17,18 Clinical records should therefore avoid credibility-damaging terms such as non-compliant, manipulative, treatment resistant or lacks insight unless these terms are clearly evidenced, necessary and open to review.^17,25,45

At the systemic level, practical humility requires scrutiny of how legal concepts are operationalised. Services should regularly audit how risk, capacity, insight, treatment necessity and least restrictive care are documented in involuntary treatment decisions, and whether patients can access and challenge the reasons given for detention, exclusion or treatment refusal. ⁴⁵ Public or service-level review of mental health law interpretation should include people with lived experience of coercion, diagnostic disbelief and treatment-related harm, because legal safeguards cannot meaningfully constrain professional confidence if the reasoning that activates them remains vague or unexamined.^38,39,45 Classification and policy processes should also address commercial and institutional sources of authority, because financial conflicts of interest in DSM revision may undermine trust even when they do not establish wrongdoing. ⁶

Treatment recommendations should include safety as well as efficacy. Clinicians and services should support meaningful informed consent by discussing potential benefit, uncertainty, alternatives and harms transparently and honestly, and should monitor adverse effects beyond the end of treatment.^23,24,31 Patient-reported outcomes, open-ended feedback and qualitative methods are necessary because clinician-defined outcomes may miss relational, epistemic, identity-based and trauma-related harms.^23,24 When harm is reported, the ethical response should not be defensive reinterpretation, but acknowledgement, investigation, repair and service learning.^24,26 This also means treating dropout or disengagement as a signal for reflective review rather than automatically as patient failure. ²⁵

Practical applications must also include cultural humility. Treatment models should not be presented as culturally neutral simply because they are described as evidence-based.^32,37 In relation to CBT, Tamatea, Ward and Day argue that clinical work with Indigenous peoples in Aotearoa-New Zealand and Australia requires explicit attention to the values underpinning CBT, the limits of individualising frameworks, systemic oppression, and points of epistemic and normative divergence with Indigenous ways of being. ³⁷ Clinicians should therefore ask whether a recommendation’s assumptions about agency, responsibility, rationality, distress and recovery fit the person’s cultural, social and political world.^33–35,37

Research and education should also change. Psychiatric education should teach epistemic humility as a clinical competence, including how to recognise iatrogenic harm, credibility deficits, diagnostic overreach, coercive reasoning and culturally situated assumptions.^10,11,13 Lived experience leadership should shape research questions, outcome selection, harms monitoring, interpretation, service review and dissemination, rather than being limited to illustrative consultation.^38–42 Classification reform should also be accountable to plural epistemologies and rights-based approaches, not only technical revisions within existing diagnostic frameworks. ⁵ Where standardised competence frameworks are presented as widely applicable, they should incorporate co-production, harm prevention, neurodiversity-affirming adaptation, relational safety, collaborative pacing and accessible outcome monitoring. ⁴³ These are not optional additions to clinical rigour. They are safeguards against the unexamined confidence of systems that can diagnose, detain, treat, exclude and disbelieve.

The ethical task, then, is epistemic reciprocity. If patients’ confidence can be clinically interrogated, psychiatry’s confidence must also be institutionally interrogated. If patients are asked to develop insight, psychiatry must develop accountable self-insight. If patient conviction can be treated as clinically meaningful, professional conviction must also be treated as ethically consequential, especially when backed by coercive authority, diagnostic power and control over access to care.

A humble psychiatry would not abandon expertise. It would refuse to confuse expertise with immunity from scrutiny and critique. It would distinguish uncertainty from incompetence, dissent from pathology, mistrust from lack of insight, and disengagement from simple non-compliance. It would ask not only how patients hold confidence, but how psychiatric systems authorise and assert their own confidence, distribute credibility and determine whose accounts are treated as valid, credible and actionable.

The issue, then, is not confidence itself, but the unequal interpretive conditions under which confidence is received. It is whose confidence is granted interpretive authority, and whose confidence is transformed into clinical evidence of pathology. Until psychiatry subjects its own institutional confidence to the same scrutiny it applies to patients, and until experiential knowledge is treated as valid, credible and actionable rather than merely anecdotal or clinically interpretable, its calls for insight, humility and metacognitive correction will remain ethically incomplete. ¹⁸

The task is not to make psychiatry less capable of acting, but to make its action more transparent and accountable. A psychiatry committed to humility would not abandon expertise; it would refuse to confuse expertise with exemption from scrutiny and critique.