Crisis of objectivity: using a personalized network model to understand maladaptive sensemaking in a patient with psychotic, affective, and obsessive-compulsive symptoms

The paper addresses the problem of psychiatric comorbidity and the limitations of categorical diagnostic systems (DSM/ICD) that obscure heterogeneous symptom presentations and overlap. While dimensional frameworks like RDoC help, the authors argue a phenomenological account of lived experience remains crucial. They set three goals: to demonstrate how personalized network models (PNMs) can be integrated into qualitative phenomenology; to evaluate whether PNMs and RDoC can assist diagnosis; and to introduce a novel phenomenological category—the “crisis of objectivity” (CoO), a pervasive mistrust of one’s own cognition and subjectivity—as the core pattern of sensemaking in the presented case. The work aims to refine PNM methodology and show how multi-level, person-centered modeling may clarify comorbid symptom clusters.

The authors review critiques of DSM/ICD categorical diagnoses regarding overlapping symptoms and biology, heterogeneity within disorders, and the diagnostic hierarchy concept. They discuss RDoC as a dimensional, multi-level framework emphasizing endophenotypes across domains (cognitive, positive/negative valence, arousal, social), while noting phenomenological critiques of weak conceptual grounding of disease entities. They introduce PNMs (De Haan, 2020) as depictions of salient variables and relations across biological, social, experiential, and existential domains, and summarize empirical use in dual case studies of psychosis and cannabis (Larsen et al., 2022). They also address critiques of network models (boundary problems and perspectivism; De Boer et al., 2022) and the concern about relevance of emergent patterns for stakeholders. The paper situates CoO within phenomenological psychopathology (self-disorders, hyperreflexivity, inferential confusion in OCD) and transdiagnostic findings relevant to RDoC domains (social cognition, sensorimotor markers, cognitive control).

Design: Longitudinal single-case study over ~2 years using mixed qualitative-quantitative methods to build a personalized network model (PNM) grounded in lived experience. Participants and context: One highly engaged adult (pseudonym Hernan), 27-year-old, Western European, living abroad with husband; history of psychosis-like experiences, OCS, anxiety, adverse childhood experiences (ACE). Remote (video) contact owing to recruitment via an online psychotherapy project; supportive psychotherapy provided during the study. Data collection: - Qualitative: 19 interviews over 18 months, including in-depth phenomenological interviewing informed by micro-phenomenology, interpretative phenomenological analysis (IPA), and constructivist grounded theory. Two sessions for EAWE and EASE. Medical history (two sessions) by psychiatry resident. EASE by trained clinician. Interviews transcribed verbatim. - Semi-structured symptom interviews: EASE (Examination of Anomalous Self Experience) and EAWE (Examination of Anomalous World Experience). Items scored 0/1 for absence/presence. - Quantitative assessments (mini-RDoC battery adaptation): - Positive valence: BIS/BAS drive, fun-seeking, reward responsiveness; PANAS positive. - Negative valence: BIS/BAS BIS; PANAS negative; SCL-90 phobic anxiety. - Social processes: WHODAS 2.0 getting along, participation; SCL-90 interpersonal sensitivity, anger/hostility. - Sensorimotor systems: SCL-90 somatization; WHODAS mobility. - Hot cognition: Emotion Regulation Questionnaire (ERQ: expressive suppression, cognitive reappraisal); WHODAS cognition; Barratt Impulsiveness Scale. - Cold cognition: Verbal 2-back task (letters, 2.0 s display; accuracy recorded). - Additional clinical scales: SPQ-32 subscales; Y-BOCS (obsessions/compulsions); PHQ-9 (depression); GAD-7 (anxiety); RRS (brooding, reflective pondering); SCL-90 subscales (OCD, depression, anxiety, psychoticism, paranoid ideation); CTS (Childhood Trauma Screener); WHODAS total. - Keyword analysis: Extracted transcripts from two prior qualitative datasets on sense of realness (normative and altered-experience samples). Searched Hernan’s transcripts for root morphemes: rationality, truth, objectivity, reality/fact/logic (five keywords as listed); removed interviewer usage; computed average occurrences per interview; compared groups via nonparametric tests (Shapiro-Wilk normality tests failed; Mann-Whitney; FDR correction). Determined Hernan’s percentile ranks within merged transdiagnostic sample. Analysis: - IPA in two phases: (1) inductive-deductive coding of transcripts using a codebook (pre-existing concepts plus emergent categories like CoO). (2) Construction of PNM by identifying relationships among experiential categories. Simplifying assumption: relationships coded as upregulation or downregulation; directed edges; rhizomatic possibility of connections. Only well-grounded links (appearing across multiple sessions) retained; single-episode links discarded. PNM divided into five domains: sensemaking, symptoms, developmental factors, biological factors, social factors. Given primarily phenomenological data, only experiential-level relations were specified. Additional keyword analysis conducted as above. Norms for scales derived from lab/clinical data and published sources as specified.

- Core experiential category: Crisis of Objectivity (CoO) identified as central node in the PNM, characterized by pervasive mistrust of one’s cognition and subjectivity. Developmentally linked to adverse childhood experiences (mother with schizophrenia, familial rejection and homelessness) and to Hernan’s own psychotic episode(s). CoO is reinforced by parasomnias and anxiety, contributes to OCS, and paradoxically is experienced as helpful for structuring against anxiety. - Personalized Network Model (PNM): Rhizomatic network with CoO having the most connections (N=7). Key relations: ACE and psychotic experiences upregulate CoO; CoO upregulates OCS; OCS in turn reinforces CoO via perceived yielding to irrationality; parasomnias and anxiety reinforce CoO. Ruminative introspection (RI) downregulates anxiety and the dissociative “Falling into the Sunken Place.” - Symptom scales (selected): - Y-BOCS: mild OCS. - PHQ-9: moderate depression. - GAD-7: severe anxiety. - RRS: reflective pondering ~74–75th percentile; brooding ~90th percentile. - SCL-90: Depression 96.2nd percentile; Psychoticism 97.8th; OCD 84th; Anxiety 67.8th; Paranoid ideation 55.8th. - SPQ-32: Unusual perception 100th percentile; Eccentric behavior 99th; No close friends 93rd; Constricted affect 87th; others lower to mid-range. - CTS: very high, indicating significant ACE (reported as 112; 99th percentile). - WHODAS: significant functional impairment (sum score 98; 91.6th percentile). - EASE/EAWE: EASE total 14; EAWE total 59 (28 excluding schizophrenia-nonexclusive items). Disturbances in minimal self and world experience present but not uniformly across domains; consistent with milder end of schizophreniform spectrum and prior schizoaffective diagnosis. - Cognitive task: 2-back accuracy ~83rd percentile; despite good performance, the patient expressed continual doubt about performance, aligning with CoO. - RDoC profile: Middle-range sensorimotor, negative, and positive valence; high behavioral approach; consistently low social processes (alienation/isolation); high expressive suppression and lower cognitive reappraisal; moderately low impulsivity yet everyday cognitive challenges. - Language/keyword analysis: Marked preoccupation with epistemic terms. Compared to normative and altered-experience samples, Hernan showed extremely high usage and high percentile ranks: objectivity (100th; p<0.005), rationality (98th; p<0.005), fact (93rd; p<0.005), with similarly elevated logic and truth (non-significant after correction). Speech showed divergent thinking and occasional disorganization; strong metacognitive focus and axiomatic/analytic phrasing consistent with CoO. - Phenomenology: Detailed accounts of dissociative social episodes (“Sunken Place”), parasomniac hallucinations with veridical, threat-congruent content leading to nocturnal aggression, imperative auditory hallucinations prompting self-injury, visual pseudo-hallucinations, and OCS centered on danger checking and distrust of memory/perception. RI described as multi-step, ritualized analysis linking subjective events to objective facts and formal logic, culminating in reified conclusions that temporarily reduce anxiety/dissociation.

The findings demonstrate that a phenomenology-informed PNM can identify a core pattern of maladaptive sensemaking—CoO—as organizing multiple comorbid symptom clusters. From a phenomenological psychopathology perspective, Hernan exhibits disturbances of the minimal self (EASE=14), consistent with a milder schizophreniform profile (aligned with schizoaffective diagnosis). CoO, marked by chronic doubt regarding subpersonal processes (intuition, memory, perception), resonates with concepts of hyperreflexivity and loss of natural evidence in self-disorders, and with inferential confusion in OCD (distrust of senses leading to overvalued hypothetical possibilities). The patient’s compulsive adherence to doubt in OCS may functionally parallel delusional certainty in psychosis, situating CoO at the interface of obsessive-compulsive and psychotic phenomena. RDoC mapping clarifies domain-level dysfunctions: social processes (eye contact avoidance, impaired getting along/participation), sensorimotor anomalies (twitches, scratching), cognitive systems (good cold cognition on 2-back yet hot-cognition regulation biased to suppression over reappraisal; metacognitive doubt), and negative/positive valence system alterations (threat sensitivity; approach motivation). Language features suggest positive thought disorder and epistemic preoccupations. Developmental and sociocultural modifiers (ACE, discrimination related to sexuality/immigration) likely shape domain deficits via chronic stress and regulation load. Methodologically, using PNMs as a qualitative scaffold aided integration of lived-experience data with quantitative corroboration, highlighting how core sensemaking patterns can drive comorbid presentations and feedback loops.

This case demonstrates that integrating personalized network models with qualitative phenomenology and RDoC perspectives can uncover a unifying experiential core—the crisis of objectivity—organizing psychotic, obsessive-compulsive, and affective symptoms. The PNM illuminated how ACE and psychotic experiences feed CoO, how CoO contributes to OCS and anxiety, and how ruminative introspection can transiently downregulate anxiety and dissociative episodes. Contributions include: (1) a proof of principle for embedding PNMs in qualitative analysis; (2) an evaluation of RDoC/PNM utility for clarifying diagnosis and comorbidity; and (3) the articulation of CoO as a novel phenomenological category. Future work should integrate quantitative metrics and graph-theoretic modeling directly into PNMs, expand beyond single cases to assess generalizability and treatment personalization, and test whether targeting CoO-related processes improves outcomes across comorbid spectra.

- Single-case design limits generalizability; PNMs and inferences require replication across larger samples. - PNMs here were constructed primarily from phenomenological data; quantitative and biological measures were used for validation but not integrated into the network. Future PNMs should incorporate objective metrics and graph-theoretic methods. - Remote (video) assessments constrained clinical examinations (e.g., no bloodwork, neuroimaging) and may reduce interview validity due to reduced intersubjective attunement. - Methodological simplifications: relationships restricted to up/downregulation; only connections recurring across sessions retained; substantial data burden (≈14+ interviews) needed to ground nodes/edges; some potentially relevant single-episode links were excluded.