Researchers Propose Dynamical Systems Framework for Complex Emotional Needs

A theoretical paper published in Frontiers in Human Neuroscience proposes reconceptualizing mental health conditions characterized by pervasive instability as dynamical disorders of multiscale synchronization rather than fixed categorical diagnostic entities. The authors, led by Franco Orsucci of the Norfolk and Suffolk NHS Foundation Trust and the University of Amsterdam alongside collaborators from eCampus University, the National Research Council of Italy, Clinic North at Psychiatric Hospital Bronderslev, and Aalborg University, argue that traditional categorical nosology struggles to accommodate the continuous, multidimensional nature of psychiatric distress. In particular, the authors focus on Complex Emotional Needs, a transdiagnostic presentation encompassing severe emotional dysregulation, unstable interpersonal relationships, identity disturbance, impulsivity, recurrent self-harm, and suicidality. Standard clinical interventions currently observe non-remission rates between 40 percent and 60 percent for common mental health conditions, an outcome that the authors attribute to a mismatch between rigid categorical diagnostic systems and the dynamic, multi-scale psycho-neurobiological processes underlying human distress.

The concept of Complex Emotional Needs emerged from work by the NIHR Mental Health Policy Research Unit to describe difficulties frequently observed in individuals diagnosed with personality disorders, complex post-traumatic stress disorder, or related conditions under ICD-11 and DSM-5-TR frameworks, while addressing persistent stigma associated with the personality disorder label. Globally, an estimated 8 percent of the population experiences difficulties consistent with Complex Emotional Needs, including an estimated prevalence of 4.4 percent in Great Britain. To account for this population, the authors construct a formalization grounded in complex adaptive systems theory, coordination dynamics, and coupled oscillator theory. They define dynamical disorders not as structural brain lesions or immutable traits, but as qualitative disturbances in physiological and psychological control dynamics across nested biological, psychological, and social scales, characterized by bifurcations, phase transitions, and loss of systemic flexibility.

Central to the proposed mathematical formulation is a stochastic master equation that models human functioning as a heterogeneous network of coupled oscillators spanning neural, autonomic, affective, cognitive, and interpersonal dimensions. Within this master equation, the state vector of each component evolves according to its autonomous intrinsic dynamics, directional coupling terms representing the magnitude and functional form of influence from other components, and a stochastic noise amplitude. The noise parameter incorporates both endogenous physiological fluctuations and externally modifiable elements accessible to intervention, such as environmental structuring, pharmacological agents, sensory management, or arousal regulation procedures. The authors stress that this formalization does not demand that neural firing rates and interpersonal conversations share identical physical units; rather, variables are normalized to their within-individual distributions so that coupling parameters represent rates and relative coordination dynamics across differing state spaces.

Operating within this mathematical structure, healthy human functioning is characterized by metastability, an optimal dynamic regime defined by partial synchronization where subsystems simultaneously balance functional integration and specialized segregation. Metastability prevents the system from falling into rigid phase-locking or chaotic decoupling, allowing flexible transitions between states as environmental contexts shift. The authors describe how intrinsic dynamics implicitly generate an effective potential energy landscape whose topology is determined by three core geometric parameters: basin curvature, energy barrier heights, and stability exponents corresponding to Lyapunov exponents. Curvature indicates how steeply an individual is drawn into a particular state, energy barrier heights determine the resistance or effort required to transition across different states, and the Lyapunov stability exponent dictates whether perturbations are adaptively dampened or amplified over time.

From the parameter space governing this multi-scale synchronization, the theoretical framework derives four distinct dynamical phenotypes that manifest as continuous topological regimes rather than discrete diagnostic categories. The Hypervigilant phenotype is defined by high basin curvature, low energy barriers, and unstable dynamics, resulting in rigid local stability accompanied by sudden, catastrophic transitions under stress. The Collapsed phenotype exhibits low curvature, exceptionally high energy barriers, and neutral stability, corresponding to severe systemic decoupling where high energy expenditure sustains behavioral and emotional shutdown. The Disorganized phenotype displays variable curvature, intermediate energy barriers, and chaotic trajectory instability, marked by rapid oscillations between hyper-synchronized and hypo-synchronized states without consolidated coordination. Finally, the Balanced phenotype features moderate curvature, traversable barriers, and negative Lyapunov exponents, supporting flexible, context-appropriate synchronization and robust recovery from emotional perturbations.

These derived topological landscapes align with foundational concepts in affective neuroscience, particularly Jaak Panksepp’s evolutionary model of seven primary-process subcortical emotional circuits: SEEKING, FEAR, RAGE, PANIC/GRIEF, CARE, PLAY, and LUST. In the Hypervigilant phenotype, the SEEKING drive becomes subordinated to threat detection alongside chronic upregulation of the amygdala to periaqueductal gray FEAR axis, paired with high precision weighting of threat-related expectations in predictive processing networks. The Collapsed phenotype is conceptualized as the neurobiological consequence of prolonged PANIC/GRIEF circuit activation followed by learned suppression, resulting in visceral attenuation of interoceptive prediction errors and emotional numbing. The Disorganized phenotype corresponds to impaired hierarchical integration between subcortical primary affective circuits, bodily interoceptive signals, and higher-order cortical regulatory processes, whereas the Balanced phenotype demonstrates successful cortical modulation of accessible primary emotional systems.

The authors also map these four dynamical phenotypes to contemporary attachment theory, framing attachment styles as fluid, context-dependent adaptations that function as biological control systems regulating proximity and arousal under stress. The Hypervigilant phenotype corresponds to anxious-preoccupied attachment, displaying paradoxical oxytocin responses where endogenous neuropeptide release amplifies interpersonal threat perception. The Collapsed phenotype maps onto dismissive-avoidant attachment with blunted oxytocin release following social cues, while the Disorganized phenotype mirrors disorganized or disoriented attachment, which prior literature links to structural alterations in the amygdala, hippocampus, and temporal cortices. The Balanced phenotype maps to secure attachment, characterized by context-appropriate oxytocin release that supports affiliative behavior, emotional attunement, and flexible interpersonal repair.

To evaluate these coordination dynamics empirically, the paper outlines a layered measurement architecture that translates continuous bio-psycho-social time series into quantitative dynamical indices. Modalities examined include orthographic text series derived from speech analysis, motion-energy tracking of physical movement, and heart rate variability, which serves as a physiological biomarker of parasympathetic vagal tone. Nonlinear dynamics are quantified using recurrence quantification analysis and cross-recurrence quantification analysis, which calculate metrics such as recurrence rate, determinism, laminarity, trapping time, and Shannon entropy without requiring assumptions of data linearity or stationarity. Furthermore, surrogate synchrony analysis uses iterative amplitude-adjusted Fourier transform procedures to construct empirical null distributions, allowing researchers to distinguish genuine interpersonal coupling from chance associations across therapeutic dyads.

This dynamical framework offers specific translational implications for clinical interventions, including the analysis of interpersonal rupture and repair during psychotherapy sessions. In the Hypervigilant phenotype, high curvature and low energy barriers render repairs fragile, meaning minor subsequent strains easily destabilize newly negotiated collaboration. In the Collapsed phenotype, ruptures manifest primarily as subtle withdrawal and quiet compliance that clinicians may overlook, where overcoming high energy barriers requires active energetic engagement without triggering defensive withdrawal. In the Disorganized phenotype, patients experience rapid mid-session shifts between confrontation and withdrawal, and the stability of repair may paradoxically trigger distress. Grounding modular intervention sequences in phenotype topology explains why symptom-focused approaches often cause symptom substitution, and clarifies why treatments such as mindfulness, distress tolerance, and behavioral activation alter underlying landscape curvature and barrier heights rather than simply removing target symptoms.

The authors formulate six explicit, falsifiable predictions to test the validity and clinical utility of their theoretical model in future empirical investigations. The primary clinical prediction posits that randomized controlled trials comparing phenotype-matched modular treatment to treatment-as-usual will demonstrate superior outcomes on measures such as the Borderline Symptom List-23 or the Difficulties in Emotion Regulation Scale at six-month follow-up, predicting medium between-group effect sizes between Cohen’s d of 0.4 and 0.6. Additional hypotheses predict that baseline recurrence metrics and surrogate synchrony will forecast therapeutic alliance, that successful treatment will shift 40 to 50 percent of phenotype-matched patients into the Balanced classification compared to 20 to 30 percent in usual care, and that changes in dynamical dwell metrics like laminarity and trapping time will precede symptom reduction in cross-lagged modeling.

The framework establishes explicit falsification criteria across structural, measurement, and clinical tiers to ensure rigorous scientific testing. The structural model will be falsified if estimated parameter distributions show an absence of topological bifurcation structure, or if pure dimensional models lacking qualitative transitions fit clinical data better. The operationalization will be falsified if recurrence quantification metrics fail to recover underlying landscape parameters above chance in simulation, or if parameter estimates lack stability across treatment sessions. At the clinical level, the framework will be considered falsified if phenotype-based treatment allocation fails to outperform treatment-as-usual across two independent trials, or if no interaction between phenotypes and module efficacy is observed.

The investigators emphasize several significant limitations of the work in its current form, reiterating that the four dynamical phenotypes remain theoretically derived rather than empirically validated in clinical populations. Calibration procedures have thus far been evaluated against synthetic data, leaving parameter recovery from empirical physiological recordings unproven. In addition, the framework characterizes the geometric topology of coordination landscapes at a fixed point in time, omitting how developmental histories produce specific landscapes or how dynamics shift across the lifespan. Finally, because the foundational literature stems largely from Western healthcare settings, the authors highlight that the cultural universality of these attractor regimes remains an open empirical question requiring systematic validation across diverse populations.

Subject of Research: Theoretical neuroscience framework modeling Complex Emotional Needs through multiscale synchronization dynamics

Article Title: Dynamical disorders: synchronization dynamics in complex emotional conditions

Article References: Orsucci, F., Zimatore, G., Baines, J., & Høgenhaug, S. S. (2026). Dynamical disorders: synchronization dynamics in complex emotional conditions. Frontiers in Human Neuroscience, 20, Article 1878289. https://doi.org/10.3389/fnhum.2026.1878289

Image Credits: AI Generated

DOI: 10.3389/fnhum.2026.1878289

Keywords: affective neuroscience, attractor landscapes, coordination dynamics, complex emotional needs, dynamical disorders, metastability, recurrence quantification analysis, synchronization dynamics, Dynamical, disorders, synchronization, dynamics