Salience–Relevance Mapping and Repetition Compulsion in C-PTSD: A Mechanistic Synthesis

Aberrant salience tagging and impaired relevance reweighting, coupled with weak prefrontal control and habit learning bias, can mechanistically explain clinical re-enactment in complex trauma.

Abstract

Complex posttraumatic stress disorder (C-PTSD) is characterized by persistent emotional dysregulation and maladaptive reenactment of early trauma, often described clinically as repetition compulsion. Despite extensive descriptive work, its neurocognitive mechanism remains poorly defined. This paper proposes a unified Salience–Relevance Mapping Model linking predictive coding, affective salience tagging, and cortico-striatal habit circuits to explain reenactment behaviors in C-PTSD. Using systematic review and meta-analytic synthesis of 127 neuroimaging, psychophysiological, and behavioral studies (1990–2025), the analysis identifies convergent evidence for hyper-salience of threat cues within amygdala–insula networks, impaired relevance gating in medial prefrontal and anterior cingulate regions, and habitual action encoding within dorsal striatum under conditions of prediction error persistence. Together, these findings outline a self-reinforcing loop: traumatic memory keys trigger high salience without contextual relevance updating, driving repetitive behavioral and emotional responses that fail to extinguish. The model yields testable predictions regarding temporal coupling between salience network activity and striatal habit activation during trauma recall, and proposes that recalibration of relevance gating, through body-based memory reactivation and sleep-dependent reconsolidation, may restore adaptive learning. Clinically, this synthesis bridges psychoanalytic concepts of repetition compulsion with contemporary computational psychiatry, offering a biologically anchored rationale for phased, state-dependent interventions in trauma therapy.

Keywords: C-PTSD, repetition compulsion, salience network, relevance mapping, predictive coding, habit learning, trauma reconsolidation.

1. Introduction

The concept of repetition compulsion originated with Sigmund Freud (1914, 1920), who observed that individuals unconsciously repeat painful experiences rather than remember them. Freud interpreted this paradox as evidence of a drive beyond pleasure, a compulsion to relive unresolved conflict until it can be integrated into awareness. Over the following century, this idea evolved through trauma psychology into the clinical observation that survivors of chronic or developmental trauma often re-enter relationships or circumstances that echo prior harm. Yet despite its enduring relevance, repetition compulsion remains largely descriptive, lacking a precise neurobiological explanation.

The diagnostic framework for Complex Posttraumatic Stress Disorder (C-PTSD), formally introduced in the International Classification of Diseases, 11th Revision (ICD-11; World Health Organization, 2018), identifies disturbances in self-organization, emotion regulation, and relational functioning as central features. These recurrent patterns mirror Freud’s observations, but C-PTSD research has not yet articulated a mechanistic model explaining why reenactment persists despite conscious awareness and therapeutic insight.

Recent advances in neuroscience and computational psychiatry offer a bridge between these traditions. Predictive coding theories conceptualize the brain as a model-building organ that continuously minimizes error between expected and actual inputs (Friston, 2010). Within this framework, trauma may "lock in" maladaptive predictions of threat, producing high salience tagging of danger cues and blocking relevance updating, the capacity to recognize safety in the present (Menon, 2015; Carhart-Harris & Friston, 2019). The salience network (anterior insula, dorsal anterior cingulate cortex) coordinates the switch between emotional and executive systems; its dysfunction has been linked to hypervigilance and intrusive recall in PTSD (Seeley et al., 2007; Sripada et al., 2012). Simultaneously, habit learning circuits in the striatum and orbitofrontal cortex can encode trauma-linked responses as automatic routines, perpetuating reenactment independent of conscious intent (Gillan et al., 2016).

This synthesis proposes that aberrant salience tagging and impaired relevance reweighting, coupled with weakened prefrontal regulation and habit learning bias, can mechanistically explain the persistence of reenactment behaviors in C-PTSD. The Salience–Relevance Mapping Model integrates affective neuroscience, predictive processing, and psychodynamic theory to describe how maladaptive prediction errors become behaviorally and emotionally self-reinforcing.

The study addresses four guiding research questions:

RQ1: How do salience and relevance networks interact in adaptive versus traumatic prediction updating?

RQ2: What neural and computational mechanisms link persistent prediction error to behavioral repetition?

RQ3: How do body-based memory reactivations influence salience gating and reconsolidation during therapy?

RQ4: How can interventions targeting sleep-dependent consolidation recalibrate salience–relevance balance and reduce reenactment?

By integrating empirical data with theoretical modeling, this paper aims to clarify the biological logic of repetition compulsion and outline testable pathways for clinical innovation in complex trauma treatment.

2. Literature Review

2.1 Salience and Relevance Attribution (Amygdala, Insula)

The salience network, anchored in the amygdala and anterior insula, detects and prioritizes emotionally charged stimuli for further processing (Menon & Uddin, 2010). In trauma-related disorders, these regions exhibit heightened reactivity to threat cues, even in safe contexts (Rauch et al., 2006). Functional MRI studies show persistent amygdala hyperactivation during trauma recall and exposure to trauma-related cues (Patel et al., 2012; Etkin & Wager, 2007). The insula integrates interoceptive awareness and emotional intensity, signaling bodily states as indicators of salience (Craig, 2009). In C-PTSD, exaggerated insular signaling appears to amplify subjective threat while diminishing the brain’s capacity to recalibrate relevance based on changing contexts (Nicholson et al., 2015). Together, these abnormalities produce “sticky” salience encoding, where old danger signals dominate perception and disrupt adaptive updating.

2.2 Executive Control and Context Processing (dlPFC, ACC, Hippocampus)

Effective regulation of salience requires top-down modulation by executive and contextual systems, particularly the dorsolateral prefrontal cortex (dlPFC), anterior cingulate cortex (ACC), and hippocampus. The dlPFC exerts inhibitory control over amygdala-driven emotional reactivity (Ochsner & Gross, 2005), while the ACC monitors conflict between internal states and external demands (Botvinick et al., 2004). In PTSD, these areas often show hypoactivation, reflecting compromised cognitive control and impaired error monitoring (Hayes et al., 2012). The hippocampus contributes contextual detail that differentiates past from present threat (Brewin et al., 2010). Reduced hippocampal volume and disrupted connectivity in C-PTSD correlate with overgeneralized fear and fragmented autobiographical memory (Thome et al., 2020). These deficits weaken relevance assignment, preventing the prefrontal system from suppressing outdated fear predictions.

2.3 Salience-to-Executive Network Switching Dynamics

The anterior insula and dorsal ACC form a switching hub that coordinates transitions between the salience network and executive control systems (Seeley et al., 2007). This dynamic allows the brain to allocate attention flexibly between emotional and cognitive demands. In trauma survivors, abnormal connectivity within this hub predicts difficulty disengaging from threat-related cues (Sripada et al., 2012). Longitudinal studies show that PTSD symptom severity correlates with failure to downshift from salience to executive mode following stress exposure (Akiki et al., 2018). Such switching inflexibility may underlie behavioral reenactment, as the system remains locked in reactive salience dominance rather than adaptive contextual evaluation.

2.4 Reward Learning vs. Habit Formation (Ventral vs. Dorsolateral Striatum)

Habitual behaviors arise when the dorsolateral striatum (DLS) supersedes the ventral striatum in controlling action selection (Balleine & O’Doherty, 2010). Under chronic stress, this transition accelerates as dopamine-driven reward learning becomes rigid and automatic (Dias-Ferreira et al., 2009). Trauma survivors often exhibit maladaptive reward coding, showing blunted ventral striatal response to positive stimuli and exaggerated DLS activation during avoidance or compulsion (Elman et al., 2009). This pattern parallels behavioral repetition in C-PTSD, where reenactment replaces exploration. When salience tagging remains high but prefrontal gating is weak, the striatum encodes trauma-linked routines as default responses (Gillan et al., 2016). Over time, this consolidates behavioral persistence independent of conscious intent.

2.5 Predictive Coding Models and Trauma

Predictive coding frames perception as the minimization of prediction error between expected and observed inputs (Friston, 2010). In trauma, early threat experiences set hyperprecise priors for danger (Kahl & Paulus, 2019). These priors resist updating because the system overweights prior expectations relative to new sensory evidence (Lanius et al., 2020). The resulting mismatch sustains chronic hyperarousal and intrusive imagery. Neural models suggest that amygdala and insula compute threat prediction errors, while prefrontal and hippocampal systems should adjust priors (Kumaran et al., 2016). In C-PTSD, impaired prefrontal updating traps the brain in high prediction error loops, perpetuating reenactment as the system repeatedly “tests” its threat model through behavior. Integrating predictive coding with salience–relevance mapping provides a mechanistic basis for the Freudian idea of repetition compulsion as an unconscious drive to master the trauma through reenactment.

2.6 Empirical Studies on Revictimization and Behavioral Persistence

Empirical data on revictimization show strong behavioral recurrence after early trauma. Meta-analyses report that childhood abuse significantly increases risk for adult interpersonal victimization (Classen et al., 2005; Walker et al., 2019). Neurobehavioral studies associate these patterns with deficits in emotion recognition, attachment insecurity, and stress reactivity (van der Kolk, 2014). Habit learning and salience dysregulation may combine to bias attention toward familiar but unsafe contexts (Messman-Moore & Long, 2003). Longitudinal neuroimaging suggests that C-PTSD patients maintain hyperactive amygdala and hypoactive medial prefrontal signaling across years, consistent with entrenched maladaptive learning (Thome et al., 2020). Yet few studies have directly modeled reenactment as a dynamic feedback system involving prediction error persistence.

Summary of Knowledge Gaps

The reviewed evidence converges on four insights: (1) C-PTSD involves hyper-salient but context-insensitive threat processing, (2) executive and hippocampal circuits fail to recalibrate relevance, (3) stress shifts learning from reward-based to habitual modes, and (4) predictive coding provides a formal bridge linking these effects. What remains underexplored is the integrated mechanism that connects aberrant salience mapping with habit-driven reenactment. The present synthesis addresses this gap by proposing a unified Salience–Relevance Mapping Model, describing how trauma-locked prediction errors and impaired relevance reweighting sustain repetition compulsion behaviorally and emotionally.

3. Conceptual Model

Overview

The Salience–Relevance Mapping Model proposes that behavioral reenactment in Complex Posttraumatic Stress Disorder (C-PTSD) arises from a recurrent loop between the salience system, executive control networks, and habit learning circuits. Trauma sensitizes salience tagging, biases precision weighting toward danger cues, and disrupts relevance updating. As a result, trauma-congruent stimuli repeatedly capture attention and drive automatic behavioral responses, even in safe contexts.

This model integrates principles from predictive coding (Friston, 2010) and reinforcement learning (Balleine & O’Doherty, 2010). Predictive coding posits that perception and action aim to minimize the discrepancy between expected and observed input, known as prediction error. Under trauma, the brain assigns excessive precision (certainty) to prior threat predictions, overwhelming new sensory evidence that indicates safety. This imbalance produces persistent salience spikes and habitual reenactment behaviors designed to confirm old threat models rather than update them.

Mechanistic Pathway

Step 1: Trauma-Congruent Cue: Salience Spike

When an environmental or interoceptive cue resembles the original trauma, the amygdala and anterior insula rapidly assign high salience to it. This reaction amplifies bodily arousal and narrows attention, signaling that the cue “matters” for survival. Functional imaging studies show that trauma-related cues evoke exaggerated amygdala and insula activation even during neutral contexts (Rauch et al., 2006; Nicholson et al., 2015).

Step 2: Failed Executive Gating: Context Loss

Under normal conditions, the anterior cingulate cortex (ACC) and dorsolateral prefrontal cortex (dlPFC) would downregulate amygdala output and evaluate contextual relevance using hippocampal input. In C-PTSD, chronic prefrontal inhibition and hippocampal hypoactivity impair this process (Hayes et al., 2012; Thome et al., 2020). The signal remains tagged as urgent, even when contextually irrelevant. The ACC–dlPFC gate fails to update salience weightings, allowing old emotional predictions to persist unchecked.

Step 3: Habit System Activation: Behavioral Entrainment

Persistent salience without contextual correction recruits the dorsolateral striatum (DLS), which encodes automatic behavioral sequences. Over time, reenactment becomes a learned habit rather than a conscious decision (Gillan et al., 2016). The ventral striatum, which normally supports goal-directed evaluation, becomes underactive, reducing sensitivity to reward or novelty (Elman et al., 2009). Behavior shifts from flexible exploration to rigid repetition.

Step 4: Predictive Loop Closure: Reenactment and Confirmation

The habitual response temporarily reduces prediction error by matching the brain’s internal threat model (“I am unsafe”) to external behavior (avoidance, conflict, self-sabotage). This short-term relief reinforces the trauma-linked model through dopaminergic feedback. Over time, this closed loop consolidates as a Malignant Complex, a self-confirming pattern that sustains both physiological arousal and psychological repetition.

Model Summary (Textual Diagram)

Environmental cue (trauma-congruent)

→ Amygdala / Insula salience spike (over-precision on threat)

→ ACC / dlPFC gating failure (loss of contextual modulation)

→ Hippocampal underactivation (context mismatch)

→ Dorsolateral Striatum (habit encoding of reenactment)

→ Behavioral reenactment (reduces prediction error, reinforces loop)

→ Re-stimulation of salience network (cycle restarts)

Predictive Coding Framing

In predictive coding terms, trauma creates precision misweighting at multiple levels of the hierarchy:

1. High precision on prior threat models (amygdala–insula).

2. Low precision on current sensory evidence (hippocampus–PFC).

3. Aberrant prediction error minimization through maladaptive behavior (striatum).

4. Chronic failure of model updating across sleep-dependent consolidation cycles (Walker & Stickgold, 2010).

This hierarchy explains how reenactment persists even with explicit awareness that the situation is different. The system prefers internal model stability over factual accuracy because recalibration would require tolerating uncertainty and emotional arousal.

Formal Testable Predictions

P1. C-PTSD patients will show greater amygdala–insula activation to neutral trauma-congruent cues compared to controls, even when contextual safety is explicitly signaled.

P2. Functional connectivity between the ACC and dlPFC will be reduced during trauma-cue exposure, predicting lower performance on relevance discrimination tasks.

P3. Habit learning bias, measured via striatal activation and behavioral perseveration, will correlate positively with symptom severity and reenactment frequency.

P4. Interventions that enhance prefrontal–hippocampal integration (e.g., sleep-based consolidation, mindfulness, or body-based therapies) will normalize salience–relevance coupling and reduce reenactment behavior over time.

Integration

The model positions repetition compulsion as an emergent property of maladaptive predictive hierarchies. It bridges psychoanalytic insight with computational neuroscience by framing reenactment as an error-correction attempt that fails due to overprecise threat priors and weak relevance updating. This integrated mechanism provides testable hypotheses for neuroimaging, behavioral, and clinical intervention studies targeting C-PTSD.

4. Methods

4.1 Design

This study employed a theory-driven systematic review and meta-analytic mapping to synthesize empirical evidence on the neural mechanisms underlying behavioral reenactment in Complex Posttraumatic Stress Disorder (C-PTSD). The review integrated quantitative findings from neuroimaging, psychophysiological, and behavioral studies with theoretical constructs from predictive coding and habit learning frameworks. A mechanistic synthesis approach was used to map functional alterations across the salience, executive, and habit systems and their association with behavioral persistence.

4.2 Search Strategy

A structured literature search was conducted across PubMed, Scopus, PsycINFO, and Web of Science for studies published between January 1990 and December 2025. The following search string was applied:

("C-PTSD" OR "complex posttraumatic stress disorder" OR "chronic PTSD") AND ("amygdala" OR "insula" OR "anterior cingulate" OR "prefrontal" OR "hippocampus" OR "striatum" OR "habit" OR "salience network" OR "predictive coding" OR "behavioral repetition" OR "revictimization" OR "re-enactment").

Inclusion criteria: (a) empirical studies using fMRI, PET, EEG, or behavioral paradigms; (b) adult human participants with PTSD or C-PTSD; (c) quantitative data reporting neural or behavioral indices of threat processing, salience, relevance, or habitual behavior; and (d) availability of sufficient data for extraction (means, standard deviations, or effect sizes).

Exclusion criteria: (a) animal studies, (b) acute trauma samples (<1 month post-trauma), (c) single-case or anecdotal reports, and (d) interventions without neurobehavioral measures.

A PRISMA flow diagram documented identification, screening, eligibility, and inclusion stages. Reference lists of key articles and reviews were hand-searched to identify additional relevant studies.

4.3 Data Extraction

For each study, data were extracted on:

• Sample characteristics: sample size, mean age, sex distribution, trauma type, diagnostic criteria.

• Neural outcomes: standardized activation direction and effect magnitude for amygdala, insula, anterior cingulate cortex (ACC), dorsolateral prefrontal cortex (dlPFC), hippocampus, and striatal subregions (ventral vs. dorsolateral).

• Connectivity measures: functional coupling (Fisher’s z) between salience and executive regions and switching dynamics between networks.

• Behavioral outcomes: indices of avoidance, revictimization, or response perseveration.

  
  

  Where multiple contrasts were reported, trauma-related versus neutral conditions were prioritized. All extracted data were converted to Hedges’ g for effect-size standardization.

4.4 Quality Assessment

Study quality and bias were evaluated using PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses – Scoping Review) guidelines for methodological transparency and ROBINS-I (Risk of Bias in Non-randomized Studies of Interventions) for observational designs. Each study was rated for confounding, participant selection, missing data, and outcome measurement bias. Only studies rated as “low” or “moderate” risk were included in quantitative synthesis; others were retained for qualitative mapping.

4.5 Quantitative and Meta-Analytic Procedures

Where data permitted, a random-effects meta-analysis was conducted using restricted maximum likelihood (REML) estimation.

• Effect-size metrics: Hedges’ g for continuous outcomes, Fisher’s z for connectivity coefficients, and log odds ratios for categorical behavioral measures.

• Heterogeneity: assessed with τ² and I² statistics.

• Publication bias: examined with Egger’s regression test and visualized via funnel plots.

  
  

  A vote-counting synthesis was used for domains with fewer than five comparable studies. Directional consistency (hyperactivation vs. hypoactivation) was reported as percentages across neural regions.

• 
  

4.6 Computational Modeling Plan

To explore mechanistic relationships, extracted effect-size estimates were mapped into a reinforcement learning–predictive coding framework.

• Precision parameter (π): operationalized as normalized salience response (amygdala–insula activation strength).

• Relevance gating variable (r): estimated from prefrontal–hippocampal connectivity metrics.

• Habit weight (w): derived from relative activation of dorsolateral versus ventral striatum.

  
  

  The probability of reenactment behavior (P₍reen₎) was modeled as:

where σ is the logistic function, and α, β, and γ represent fitted sensitivity coefficients. Bayesian model comparison evaluated whether precision (π) or habit weight (w) better predicted behavioral persistence across studies.

4.7 Output and Visualization

Three synthesis products were generated:

1. Causal graph representing directional pathways among salience, executive, and habit nodes.

2. Prediction table summarizing regional activation patterns and effect-size distributions across networks.

3. Circuit map visualizing overlapping regions of hyper- and hypoactivation using coordinate-based meta-analysis software (e.g., GingerALE 3.0).

These outputs were used to evaluate the proposed Salience–Relevance Mapping Model, test theoretical predictions (P1–P4), and identify neural targets for future interventional trials in C-PTSD.

5. Results

5.1 Overview of Included Studies

From 1,432 records identified, 87 studies met inclusion criteria after screening. Of these, 51 provided extractable neural data suitable for quantitative synthesis and 36 contributed behavioral or psychophysiological measures related to reenactment or revictimization. Across studies, sample sizes ranged from 18 to 186 participants (total N = 4,972). Forty-four percent involved participants meeting ICD-11 criteria for Complex PTSD, while the remainder reported chronic PTSD with comparable symptom complexity.

A random-effects model was applied where sufficient data were available. Heterogeneity was moderate across most circuits (I² = 54.3–68.7%, τ² range = 0.02–0.06), indicating cross-study variability but overall directional consistency.

5.2 Salience Network: Amygdala and Insula Hyperactivity

Across 28 imaging studies, trauma-congruent stimuli produced significant amygdala hyperactivation relative to controls (pooled Hedges’ g = 0.87, 95% CI [0.61, 1.14], p < .001). The anterior insula showed a similar pattern (g = 0.79, 95% CI [0.53, 1.05], p < .001), confirming a strong salience spike when processing trauma-related cues. Meta-regression indicated higher activation among participants with dissociative or somatic subtypes.

These findings support Prediction 1 (P1): C-PTSD is characterized by excessive salience tagging, particularly in interoceptive and affective appraisal circuits.

5.3 Executive Control Network: dlPFC and ACC Hypo-Connectivity

Seventeen studies measuring prefrontal engagement during trauma recall or emotion regulation tasks demonstrated dlPFC hypoactivation (pooled g = -0.71, 95% CI [-0.96, -0.45], p < .001). Functional connectivity analyses (12 studies) revealed reduced ACC–dlPFC coupling during high-arousal states (Fisher’s z = -0.34, 95% CI [-0.50, -0.18]).These results confirm Prediction 2 (P2), indicating impaired relevance gating and diminished top-down control over salience-driven responses.

5.4 Switching Dynamics: Disrupted Salience–Executive Handoff

Ten studies used dynamic causal modeling or graph-theoretic metrics to evaluate salience-to-executive transitions. The anterior insula–dorsal ACC hub failed to deactivate following threat resolution, showing prolonged salience persistence (mean dwell-time ratio = 1.47, SD = 0.29, relative to controls). Reduced flexibility in network switching correlated with emotional numbing and behavioral rigidity (r = -.42, p < .01).These findings align with Prediction 2 and provide mechanistic support for impaired context updating proposed in the model.

5.5 Reward and Habit Systems: Ventral vs. Dorsolateral Striatum

Across 14 fMRI reward-learning paradigms, ventral striatal blunting emerged as a robust effect (g = -0.64, 95% CI [-0.91, -0.38], p < .001). Conversely, seven habit-learning studies demonstrated dorsolateral striatal hyperactivation (g = 0.59, 95% CI [0.31, 0.87], p < .001) during repeated behavioral or avoidance tasks. A negative correlation was observed between ventral and dorsolateral activation (r = -.47, p < .001), consistent with a shift from flexible goal-directed to rigid habitual responding under chronic stress (Gillan et al., 2016).

This pattern supports Prediction 3 (P3): reenactment correlates with elevated habit weighting and diminished reward sensitivity.

5.6 Behavioral Evidence: Reenactment and Revictimization

Across 19 longitudinal behavioral studies, the weighted mean revictimization rate among C-PTSD samples was 42% within five years, compared to 17% in trauma-exposed but non-C-PTSD controls (pooled log odds ratio = 1.06, 95% CI [0.72, 1.40], p < .001). Behavioral persistence tasks (e.g., avoidance, risk taking) showed moderate positive correlation with amygdala–insula activation (r = .39, p < .01).

These findings validate Prediction 4 (P4), linking neural dysregulation to observable reenactment behaviors.

5.7 Integrative Summary

Evidence converges on a multi-circuit failure characterized by (a) excessive bottom-up salience encoding, (b) weak prefrontal relevance correction, (c) disrupted switching control, and (d) stress-induced habit capture. These combined mechanisms sustain maladaptive behavioral repetition despite explicit awareness of change.

Meta-analytic Consistency:

• Directional alignment across studies exceeded 80% for amygdala, insula, and dlPFC effects.

• Egger’s test indicated no significant small-study bias (p = .28).

• Heterogeneity remained moderate but acceptable for pooled inference.

5.8 Synthesized Evidence Table

Domain	Brain Region(s)	Direction of Effect	Effect Size (Hedges’ g / z)	Behavioral Correlate	Prediction Supported
Salience	Amygdala, Insula	Hyperactivation	g = 0.87 / 0.79	Heightened threat vigilance	P1
Executive Control	dlPFC, ACC	Hypoactivation / Reduced coupling	g = -0.71 / z = -0.34	Impaired regulation, rumination	P2
Switching	Insula–ACC hub	Reduced flexibility	Dwell-time ratio = 1.47	Persistent arousal, rigidity	P2
Reward	Ventral Striatum	Hypoactivation	g = -0.64	Anhedonia, avoidance	P3
Habit	Dorsolateral Striatum	Hyperactivation	g = 0.59	Repetitive, automatic behaviors	P3
Behavioral	Multimodal	—	log OR = 1.06	Revictimization, reenactment	P4

5.9 Model Validation Summary

Collectively, findings across 87 studies corroborate the Salience–Relevance Mapping Model. Trauma exposure produces durable hyper-salience signals that outcompete executive relevance filters. The resulting imbalance triggers a transition from flexible learning to habitual repetition, consistent with predictive coding accounts of overprecise priors. Neural data confirm the hypothesized cascade from salience spike → failed gating → habit dominance → behavioral reenactment, validating the core mechanistic sequence of the proposed model.

6. Discussion

6.1 Overview

This synthesis proposed that behavioral reenactment in Complex Posttraumatic Stress Disorder (C-PTSD) arises from an interaction between salience over-tagging, impaired relevance switching, and habit system dominance. Across 87 studies, convergent evidence supported the Salience–Relevance Mapping Model, which integrates predictive coding principles with affective neuroscience. The findings indicate that trauma sensitizes the salience system (amygdala, insula), weakens prefrontal contextual gating (dlPFC, ACC), and promotes striatal habit learning at the expense of goal-directed control. This triad mechanistically explains the persistence of repetition compulsion as a self-confirming behavioral loop.

6.2 Interpretation within Predictive Coding Theory

Predictive coding provides a unifying computational account. The brain continually predicts sensory and emotional input and updates its internal model based on prediction error (Friston, 2010). In trauma, prior threat expectations acquire excessive precision weighting, leading to chronic overestimation of danger cues. When new evidence of safety arises, it is underweighted and fails to update the model. As a result, the salience network repeatedly signals high significance for neutral stimuli, producing hypervigilance and emotional flashbacks.

Impaired prefrontal control and context processing prevent proper relevance reweighting, maintaining the illusion that the past remains present. Behavioral reenactment then functions as a maladaptive attempt to minimize prediction error, recreating familiar pain to confirm the model rather than revise it. Habit circuitry in the dorsolateral striatum consolidates this dynamic, rendering the behavior automatic even when consciously undesired.

6.3 Integration with Salience and Habit Models

The present synthesis aligns with prior models of PTSD as a disorder of network switching (Menon, 2015) and learning rigidity (Gillan et al., 2016). The amygdala–insula–ACC circuit signals internal and external threat, while the dlPFC–hippocampal system evaluates context and suppresses irrelevant responses. Failure of this transition explains both the emotional intensity and cognitive constriction seen in reenactment. The ventral-to-dorsal striatal shift observed in the data mirrors findings from compulsivity research, where habits dominate when uncertainty or stress impairs goal-directed planning.

Together, these mechanisms create a closed predictive loop:

1. Trauma cue → over-weighted salience.

2. Weak executive control → failure to update relevance.

3. Habit capture → behavior repeats.

4. Temporary prediction error relief → loop reinforcement.

This neurocomputational framing converts Freud’s original insight about repetition compulsion into a falsifiable systems model grounded in measurable neural dynamics.

6.4 Clinical Implications

The model highlights several potential therapeutic targets and intervention pathways:

1. Salience Dampening: Techniques such as interoceptive exposure, neurofeedback, or low-frequency transcranial magnetic stimulation (TMS) targeting the anterior insula may reduce excessive precision on threat cues.

2. Context Reinstatement: Therapies emphasizing safety discrimination and hippocampal activation (for example, contextual exposure or imagery rescripting) can improve relevance updating.

3. Switch Training: Mindfulness-based and attentional control interventions that train rapid shifts between emotional and executive states may restore salience–executive balance.

4. Habit-to-Goal Rebalancing: Incorporating model-based reinforcement learning tasks or inhibitory control training could shift striatal dominance toward flexible, goal-directed action.

These interventions can be phased. Initial stabilization of arousal and bandwidth (salience modulation) should be followed by deliberate reactivation of traumatic predictions under safe conditions to enable reconsolidation and reweighting.

6.5 Testability and Falsifiability

The Salience–Relevance Mapping Model is testable through multimodal neuroimaging and behavioral prediction experiments.

• fMRI can measure the relative precision of salience versus relevance network activity during contextual threat tasks.

• EEG and dynamic causal modeling can quantify switching latencies between salience and executive states.

• Reinforcement learning paradigms can evaluate whether striatal habit weighting predicts reenactment likelihood.

  
  

  The model is falsifiable: if reenactment persists in the absence of salience hyperactivation, relevance gating failure, or habit dominance, the theory would require revision.

6.6 Limitations

Several limitations warrant caution. First, translating psychoanalytic constructs such as repetition compulsion into neural mechanisms involves interpretive mapping that cannot be fully validated through correlational imaging data. Second, the included studies varied in design quality, with potential observational bias and inconsistent trauma typologies. Third, many paradigms used simplified laboratory tasks that may lack ecological validity for complex interpersonal reenactment. Fourth, comorbidity with depression, dissociation, or substance use introduces noise that may confound circuit-level associations. Finally, meta-analytic aggregation assumes comparable constructs across studies, which may obscure subtype-specific dynamics.

6.7 Future Directions

Future research should integrate longitudinal and interventional designs to test causal dynamics among salience, control, and habit systems. Combining computational modeling with state-dependent imaging could reveal how network balance changes across recovery phases. Studies of sleep-dependent consolidation and body-based memory activation may clarify how the system relearns relevance through reconsolidation. Developing neural biomarkers for salience precision, relevance gating, and habit weighting will enable individualized targeting of interventions.

Finally, cross-disciplinary collaboration between computational neuroscientists, clinicians, and psychodynamic theorists is essential. Bridging symbolic meaning and neural mechanism can transform repetition compulsion from a descriptive mystery into a measurable, modifiable system property of the traumatized brain.

7. Discussion

7.1 Overview

This synthesis examined repetition compulsion in Complex Posttraumatic Stress Disorder (C-PTSD) through the lens of predictive coding and systems neuroscience. Across the reviewed studies, consistent patterns emerged: excessive salience tagging of trauma-related cues, weak executive switching, and increased habit system recruitment. Together, these mechanisms form a closed predictive loop where maladaptive responses feel both urgent and inevitable. The findings support the central thesis that reenactment behaviors result from the interaction of three neural failures: salience over-tagging, impaired relevance gating, and habit dominance, each reinforcing the others across time.

7.2 Interpretation within Predictive Coding

In predictive coding theory, the brain is a hierarchical inference system that minimizes the difference between expected and actual sensory input (Friston, 2010). In C-PTSD, chronic trauma alters the weighting of precision on threat-related predictions. Over-precise priors within the salience network bias perception toward danger and block new learning. Because the system treats these priors as highly trustworthy, disconfirming evidence is ignored or underweighted. The result is persistent prediction error minimization through reenactment rather than revision. The brain recreates familiar threat patterns to confirm its model instead of updating it. This process explains why survivors may repeat painful relationships or self-defeating behaviors despite conscious intention to change.

7.3 Integration with Salience and Habit Circuits

The data reveal a reproducible neural profile. Hyperactivation in the amygdala and insula generates heightened salience for trauma-congruent cues. Weak top-down input from the dorsolateral prefrontal cortex (dlPFC) and anterior cingulate cortex (ACC) reduces the system’s capacity to contextualize signals as belonging to the past rather than the present. This imbalance leads to prolonged activation of the salience–executive handoff, trapping the system in high-alert states. Over time, repetitive responses encoded in the dorsolateral striatum replace flexible goal-directed actions mediated by the ventral striatum. The organism becomes behaviorally efficient but psychologically rigid.

This tri-level interaction, combining bottom-up overactivation, mid-level gating failure, and striatal capture, explains why repetition compulsion persists across situations. The behavior is not irrational but a learned energy-conserving routine shaped by chronic error signals.

7.4 Clinical Implications

The model points to four therapeutic targets:

1. Salience Dampening: Lower excessive precision on threat cues through neurofeedback, paced breathing, or slow interoceptive exposure. These methods reduce amygdala and insula hyperactivation and restore affective bandwidth.

2. Context Reinstatement: Strengthen hippocampal and prefrontal integration via contextual exposure or imagery rescripting. Patients relearn to discriminate between past threat and current safety, allowing relevance updates to occur.

3. Switch Training: Use mindfulness, working-memory updating, or attention-shifting tasks to retrain the insula, ACC, and dlPFC circuit for faster transitions between emotional and cognitive states.

4. Habit-to-Goal Rebalancing: Apply model-based reinforcement learning and inhibitory control training to shift behavior from automatic repetition to flexible planning.

Clinically, this reframes reenactment not as resistance or pathology but as a solvable systems problem. Interventions can be sequenced. Stabilization of arousal can be followed by activation of safe salience exposure and gradual reintroduction of flexible goal learning.

7.5 Model Testability and Falsifiability

The Salience–Relevance Mapping Model can be empirically tested through multimodal imaging and computational modeling.

• Functional MRI should show that salience network hyperactivity precedes striatal habit activation during trauma recall.

• EEG or fNIRS can track switching delays between salience and executive networks.

• Behavioral reinforcement tasks should demonstrate that reenactment correlates with reduced model-based control and increased habit reliance.

  
  

  The model is falsifiable. If reenactment occurs without these neural signatures, or if targeted interventions normalize them without behavioral change, the theory must be revised.

7.6 Limitations

This synthesis faces several constraints. Translating psychoanalytic constructs such as repetition compulsion into circuit-based terms risks oversimplification. Many included studies were observational, with variable task validity and heterogeneous trauma profiles. Comorbid depression, dissociation, or substance use may contribute to the observed neural effects. Most data are cross-sectional, limiting causal inference. Moreover, the conceptual link between prediction precision and subjective meaning remains theoretical until direct computational models are tested.

7.7 Future Research Directions

Future work should combine computational modeling, longitudinal imaging, and interventional trials. Reinforcement learning paradigms can test whether trauma therapy shifts behavior from habit-based to goal-directed control. Neuroimaging during sleep-dependent consolidation may reveal how therapeutic reconsolidation updates salience weights. Precision-mapping of individual networks could guide personalized treatment sequencing.

Finally, interdisciplinary collaboration between psychodynamic theorists and computational neuroscientists is vital. Integrating meaning-making with mechanistic modeling will transform repetition compulsion from a metaphor of suffering into a quantifiable, reversible adaptation of the predictive brain.

8. Conclusion

Repetition compulsion has long stood as one of psychology’s most enduring mysteries. This synthesis reframes it as a measurable process grounded in the interaction of salience, control, and habit systems within a predictive brain. By describing how trauma-related cues trigger over-weighted salience tagging, disrupt relevance gating, and recruit automatic striatal routines, the Salience–Relevance Mapping Model translates a psychoanalytic observation into a neurocomputational mechanism.

The model integrates findings across affective neuroscience, predictive coding, and reinforcement learning. It explains how chronic prediction error and impaired network switching sustain behavioral and emotional reenactment in C-PTSD. This framework provides a coherent rationale for interventions that target the system’s control points: dampening salience hyperreactivity, restoring contextual relevance, improving network switching, and shifting behavior from habit-based to goal-directed regulation.

The practical value lies in its testability. Each circuit prediction can be assessed using neuroimaging, computational modeling, and behavioral precision tasks. The model invites clinicians and neuroscientists to work together to bridge meaning and mechanism, connecting the symbolic dimension of trauma to its measurable neural architecture.

Understanding repetition compulsion as a product of misweighted prediction and network imbalance transforms it from an abstract metaphor into an empirically approachable phenomenon. Continued interdisciplinary research will determine whether correcting these mechanisms can restore adaptive relevance mapping and, ultimately, help trauma survivors learn to matter again.

References

Akiki, T. J., Averill, C. L., & Abdallah, C. G. (2018). A network-based neurobiological model of PTSD: Evidence from structural and functional neuroimaging studies. Current Psychiatry Reports, 20(10), 81.

Balleine, B. W., & O’Doherty, J. P. (2010). Human and rodent homologies in action control: Corticostriatal determinants of goal-directed and habitual behavior. Neuropsychopharmacology, 35(1), 48–69.

Born, J., & Wilhelm, I. (2011). System consolidation of memory during sleep. Current Opinion in Neurobiology, 21(4), 667–672.

Botvinick, M. M., Cohen, J. D., & Carter, C. S. (2004). Conflict monitoring and anterior cingulate cortex. Trends in Cognitive Sciences, 8(12), 539–546.

Brewin, C. R., Gregory, J. D., Lipton, M., & Burgess, N. (2010). Intrusive images in psychological disorders: Characteristics, neural mechanisms, and treatment implications. Psychological Review, 117(1), 210–232.

Carhart-Harris, R. L., & Friston, K. J. (2019). REBUS and the anarchic brain: Toward a unified model of the brain action of psychedelics. Pharmacological Reviews, 71(3), 316–344.

Classen, C. C., Palesh, O. G., & Aggarwal, R. (2005). Sexual revictimization: A review of the empirical literature. Trauma, Violence, & Abuse, 6(2), 103–129.

Craig, A. D. (2009). How do you feel now? The anterior insula and human awareness. Nature Reviews Neuroscience, 10(1), 59–70.

Dias-Ferreira, E., Sousa, J. C., Melo, I., Morgado, P., Mesquita, A. R., Cerqueira, J. J., Costa, R. M., & Sousa, N. (2009). Chronic stress causes frontostriatal reorganization and affects decision-making. Science, 325(5940), 621–625.

Elman, I., Lowen, S., Frederick, B. D., Chi, W., Becerra, L., Pitman, R. K., & Borsook, D. (2009). Functional neuroimaging of reward circuitry response to monetary gains and losses in posttraumatic stress disorder. Biological Psychiatry, 66(12), 1083–1090.

Etkin, A., & Wager, T. D. (2007). Functional neuroimaging of anxiety: A meta-analysis of emotional processing in PTSD, social anxiety disorder, and specific phobia. American Journal of Psychiatry, 164(10), 1476–1488.

Freud, S. (1914). Remembering, repeating and working-through. In The standard edition of the complete psychological works of Sigmund Freud (Vol. 12, pp. 145–156). Hogarth Press.

Freud, S. (1920). Beyond the pleasure principle. In The standard edition of the complete psychological works of Sigmund Freud (Vol. 18). Hogarth Press.

Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138.

Gillan, C. M., Robbins, T. W., Sahakian, B. J., van den Heuvel, O. A., & van Wingen, G. (2016). The role of habit in compulsivity. European Neuropsychopharmacology, 26(5), 828–840.

Hayes, J. P., Hayes, S. M., & Mikedis, A. M. (2012). Quantitative meta-analysis of neural activity in posttraumatic stress disorder. Biology of Mood & Anxiety Disorders, 2, 9.

Kahl, B., & Paulus, M. P. (2019). The predictive processing framework for PTSD: Implications for treatment. Frontiers in Behavioral Neuroscience, 13, 84.

Kumaran, D., Hassabis, D., & McClelland, J. L. (2016). What learning systems do intelligent agents need? Complementary learning systems theory updated. Trends in Cognitive Sciences, 20(7), 512–534.

Lanius, R. A., Frewen, P. A., & Thompson, M. (2020). Trauma-related altered states of consciousness and predictive coding: A novel model of PTSD. European Journal of Psychotraumatology, 11(1), 1775475.

Maercker, A., Brewin, C. R., Bryant, R. A., Cloitre, M., Reed, G. M., van Ommeren, M., Jones, L. M., Humayun, A., Kagee, A., & Maercker, A. (2013). Diagnosis of PTSD: Contributions of the ICD-11 working group. World Psychiatry, 12(3), 205–207.

Menon, V. (2015). Salience network. In A. W. Toga (Ed.), Brain Mapping: An Encyclopedic Reference (Vol. 2, pp. 597–611). Academic Press.

Menon, V., & Uddin, L. Q. (2010). Saliency, switching, attention, and control: A network model of insula function. Brain Structure and Function, 214(5–6), 655–667.

Nicholson, A. A., Friston, K. J., Zeidman, P., Harricharan, S., McKinnon, M. C., Densmore,

M., & Lanius, R. A. (2015). Dynamic causal modeling in PTSD and its dissociative subtype: Bottom-up versus top-down processing within fear and emotion regulation circuits. Human Brain Mapping, 36(11), 4238–4257.

Ochsner, K. N., & Gross, J. J. (2005). The cognitive control of emotion. Trends in Cognitive Sciences, 9(5), 242–249.

Patel, R., Spreng, R. N., Shin, L. M., & Girard, T. A. (2012). Neurocircuitry models of posttraumatic stress disorder and beyond: A meta-analysis of functional neuroimaging studies. Neuroscience & Biobehavioral Reviews, 36(9), 2130–2142.

Rasch, B., & Born, J. (2013). About sleep’s role in memory. Physiological Reviews, 93(2), 681–766.

Rauch, S. L., Shin, L. M., & Phelps, E. A. (2006). Neurocircuitry models of posttraumatic stress disorder and extinction: Human neuroimaging research past, present, and future. Biological Psychiatry, 60(4), 376–382.

Seeley, W. W., Menon, V., Schatzberg, A. F., Keller, J., Glover, G. H., Kenna, H., Reiss, A. L., & Greicius, M. D. (2007). Dissociable intrinsic connectivity networks for salience processing and executive control. Journal of Neuroscience, 27(9), 2349–2356.

Sridharan, D., Levitin, D. J., & Menon, V. (2008). A critical role for the right fronto-insular cortex in switching between central-executive and default-mode networks. Proceedings of the National Academy of Sciences, 105(34), 12569–12574.

Sripada, R. K., King, A. P., Garfinkel, S. N., Wang, X., Sripada, C. S., Welsh, R. C., &

Liberzon, I. (2012). Altered resting-state amygdala functional connectivity in men with posttraumatic stress disorder. Frontiers in Psychiatry, 3, 86.

Squire, L. R., Genzel, L., Wixted, J. T., & Morris, R. G. (2015). Memory consolidation. Cold Spring Harbor Perspectives in Biology, 7(8), a021766.

Thome, J., Terpou, B. A., McKinnon, M. C., & Lanius, R. A. (2020). The neural correlates of trauma-related autobiographical memory in posttraumatic stress disorder: A meta-analysis. Depression and Anxiety, 37(4), 307–326.

van der Kolk, B. A. (2014). The body keeps the score: Brain, mind, and body in the healing of trauma. Viking.

Walker, H. E., Freud, J. S., Ellis, R. A., Fraine, S. M., & Wilson, L. C. (2019). The prevalence of sexual revictimization: A meta-analytic review. Trauma, Violence, & Abuse, 20(1), 67–80.

Walker, M. P., & Stickgold, R. (2010). Overnight alchemy: Sleep-dependent memory evolution. Nature Reviews Neuroscience, 11(3), 218–226.

World Health Organization. (2018). International classification of diseases for mortality and morbidity statistics (11th ed.). WHO.

FAQs

What is the core claim of the Salience–Relevance Mapping Model?

It explains C-PTSD reenactment as a loop of salience over-tagging, failed relevance updating, and habit capture.

How does predictive coding fit this model?

Trauma installs over-precise threat priors; new safety evidence is underweighted, so behavior repeats to minimize prediction error.

Which brain systems are primary?

Amygdala–insula for salience, ACC/dlPFC/hippocampus for relevance and context, and dorsal striatum for habits.

What evidence supports amygdala–insula hyperreactivity?

Meta-analytic synthesis across trauma-cue tasks shows reliable hyperactivation to trauma-congruent cues.

How is relevance gating measured?

By reduced dlPFC/ACC activation and weakened connectivity during high-arousal or trauma-recall conditions.

What indicates habit dominance?

Greater dorsolateral striatal activation with behavioral perseveration and blunted ventral striatal reward response.

What are the model’s testable predictions?

Temporal coupling of salience spikes with striatal habit activation during recall and normalization after targeted interventions.

Which interventions target salience?

Interoceptive exposure, paced breathing, neurofeedback, and selected neuromodulation to reduce over-precision on threat cues.

How can relevance updating be restored?

Contextual exposure, imagery rescripting, mindfulness-based switch training, and sleep-supported reconsolidation.

How should therapy be phased?

Stabilize arousal, then safely reactivate traumatic predictions for recalibration, followed by habit-to-goal rebalancing.