Workplace psychological trauma may result in clinically significant trauma-related symptoms, including Post-Traumatic Stress Disorder (PTSD)-like presentations. Survivors, however, often describe these experiences using non-clinical terminology, highlighting the need to bridge survivor-informed language with empirically grounded trauma frameworks to improve clinical recognition without prematurely formalizing non-validated constructs [1]. Exposure to workplace bullying and psychological aggression has been associated with adverse occupational and health outcomes, including post-traumatic stress symptoms, depression, anxiety, burnout, reduced job satisfaction, and increased intentions to leave [2]. Unlike acute traumatic events, workplace psychological abuse often unfolds gradually through repetitive relational stressors, producing cumulative psychological impact rather than a single identifiable incident.

Chronic exposure to unpredictable interpersonal hostility may activate sustained stress responses and contribute to affective dysregulation, cognitive distortions, and disturbances in self-concept [3]. Prolonged relational stress characterized by manipulation, blame-shifting, and emotional invalidation may erode psychological stability through mechanisms such as gaslighting, coercive control, intermittent reinforcement, and social isolation [4]. Several descriptive terms used in the clinical and survivor literature on psychological abuse do not have formal nosological status but are widely used descriptively to characterize recurrent interpersonal dynamics in abusive relationships. For clarity, the key terms used in this manuscript are briefly defined below.

In everyday discourse, survivors often describe such patterns using terms such as “narcissistic abuse,” “gaslighting,” “DARVO,” “flying monkeys,” or “trauma bond” [5,6]. Although these expressions lack formal nosological status, they frequently refer to recognizable psychological processes. ”Narcissistic abuse” refers to patterns of psychological manipulation, emotional devaluation, and coercive interpersonal control associated with individuals displaying pronounced narcissistic traits within interpersonal relationships [7]. Gaslighting refers to a form of psychological manipulation in which a perpetrator systematically undermines the victim’s confidence in their own perceptions, memory, or judgment, often resulting in confusion and self-doubt [8,9]. Coercive control describes patterns of domination aimed at restricting autonomy and eroding agency [4]. DARVO—Deny, Attack, Reverse Victim and Offender—is a manipulation tactic used to deny wrongdoing and reverse blame onto the victim [4]. “Flying monkeys” refers to third-party individuals who knowingly or unknowingly support the perpetrator’s narrative and participate in the social dynamics of manipulation, intimidation, or reputational harm directed toward the target individual. Workplace studies on abusive supervision and narcissistic leadership show how narcissistic traits can mobilize others in harmful relational dynamics [10]. Trauma bonding describes a strong emotional attachment that develops between victim and perpetrator through cycles of abuse interspersed with intermittent reinforcement, which can strengthen dependency and hinder disengagement [11]. Intermittent reinforcement and relational unpredictability may strengthen maladaptive attachment bonds and complicate disengagement, contributing to trauma bonding in contexts of repeated abuse cycles [4,12].

These mechanisms are not confined to intimate relationships but may also emerge in professional hierarchies and organizational contexts. When embedded in workplace dynamics—where reputational risk, power asymmetry, and dependency structures are present—such relational stressors may intensify psychological vulnerability and prolong exposure [2,13,14]. Recent trauma research increasingly recognizes that prolonged interpersonal stress and coercive relational dynamics [15] may produce psychological effects resembling those described in Complex Post-Traumatic Stress Disorder, even outside traditionally studied contexts such as intimate partner violence or childhood abuse [16,17].

While coercive control has been extensively studied in intimate partner violence, similar relational dynamics in workplace environments remain less frequently described in clinical trauma literature [15]. Rather than focusing on personality-based labeling of the alleged perpetrator, the present case conceptualizes the reported experiences within a trauma-informed framework, examining how sustained interpersonal manipulation and psychological invalidation may contribute to trauma-spectrum symptomatology.

Victim/target

Research does not support the existence of a single, fixed “victim profile.” However, certain relational and contextual factors may increase vulnerability to sustained interpersonal exploitation. Individuals high in manipulative and dominance-oriented traits have been shown to test interpersonal boundaries and orient toward relational contexts where resistance and consequence enforcement are less likely [18]. Within everyday discourse, victims of behaviors commonly labeled as narcissistic abuse are often described as highly empathetic, relationally invested, and conflict-avoidant. Rather than reflecting inherent weakness, these characteristics may represent prosocial traits that, under conditions of coercive control, delay boundary enforcement and facilitate prolonged exposure [19].

Empathy accounts for the naturally occurring subjective experience of similarity between the feelings expressed by self and others without losing sight of whose feelings belong to whom. It involves not only affective resonance but also recognition and understanding of another person’s emotional state [20,21].

Importantly, coercive control theory conceptualizes psychological domination as a patterned restriction of autonomy through isolation, intimidation, and manipulation, rather than through episodic violence [15]. In such contexts, relational investment and empathy may unintentionally increase tolerance for destabilizing dynamics, particularly when reality-testing is progressively undermined.

Research further indicates associations between emotional dependency, partner narcissistic traits, and increased risk of psychological abuse within intimate relationships [22]. Prolonged exposure to manipulative or invalidating relational environments may result in trauma-related symptomatology, including PTSD, affective dysregulation, and disturbances in self-concept [23].

The term “Post-Narcissistic Stress Disorder” (PNSD) has emerged in survivor communities as a descriptive label for PTSD-like symptoms associated with prolonged psychological manipulation, reflecting the use of lay terminology to describe complex relational trauma [24]. Although not recognized within formal diagnostic systems, the symptom clusters described under this label overlap substantially with established trauma-spectrum conditions. Framing such presentations within empirically grounded trauma models allows clinical focus to remain on stress mechanisms and recovery processes rather than on personality-based categorization [24]. In emergency and acute care settings, individuals exposed to prolonged interpersonal workplace stress may present with crisis manifestations—such as panic episodes, dissociative symptoms, suicidal ideation, or acute stress reactions—without explicit identification of relational abuse as the precipitating factor. Trauma-spectrum symptomatology frequently remains under-recognized in acute medical environments, despite its significant impact on functional impairment and risk presentation [25]. In this case, trauma-informed assessment proved important in the emergency setting.

AI-facilitated exploratory interventions

The growing demand for psychotherapy and limited availability of mental health professionals have stimulated interest in artificial intelligence (AI) as a potential adjunct in mental health care [26]. Within clinical contexts, generative AI systems may serve as structured reflective tools rather than substitutes for professional treatment. Potential benefits include providing a non-judgmental space for narrative articulation, emotional labeling, and cognitive organization. AI-assisted dialogue may support reflective processing between sessions and facilitate recognition of relational patterns [27]. Emerging evidence suggests that AI-based conversational agents may contribute to short-term reductions in depressive and anxiety symptoms when used adjunctively [28]. However, important limitations must be acknowledged. AI-generated responses lack genuine affective attunement and may provide incomplete or overly generalized feedback [29]. Risks include inaccurate guidance, contextual misinterpretation, privacy concerns, and possible overreliance that could delay appropriate clinical help-seeking [30]. Moreover, empirical validation of many commercially available AI mental health tools remains limited.

Accordingly, AI-assisted reflection should be conceptualized as a supplementary narrative aid embedded within trauma-informed care, not as a replacement for psychotherapy.

Novelty statement

To our knowledge, this is among the first clinical case reports describing the use of a generative AI system as a structured reflective adjunct within clinician-led psychotherapy for chronic workplace interpersonal trauma.

Rather than functioning as a stand-alone digital intervention, the AI system was used as a supervised reflective tool supporting narrative processing between psychotherapy sessions.

This study is a single-case clinical report with an embedded mixed-method assessment framework. The aim was exploratory and translational: to examine trauma-related psychological sequelae following workplace psychological abuse and to conceptually map survivor-informed terminology onto exploratory clinical constructs. The design does not seek to establish a new diagnostic category but to provide structured clinical documentation of symptom patterns.

Given the exploratory nature of this single-case report, no a priori primary outcome variable was predefined; instruments were selected to capture multidimensional trauma-related, occupational, and interpersonal domains.

Participant

The participant was an adult female professional who reported prolonged exposure to psychological manipulation and workplace abuse by a colleague. At the time of assessment, she presented with significant emotional distress and occupational impairment. Written informed consent was obtained for anonymized publication. All identifying details have been removed.

Procedure

Data were collected through:

A semi-structured clinical interview and self-report method [31–34]

The semi-structured interview explored:

• Chronology of workplace events
• Subjective emotional responses
• Survivor-informed terminology (e.g., “gaslighting,” “discard,” “no contact”, “hoovering”, “love bombing”). Several terms used in the clinical and survivor literature on psychological abuse do not have formal nosological status but are widely used descriptively to characterize recurrent interpersonal dynamics in abusive relationships. For clarity, the key terms used in this manuscript are briefly defined below.

“Gaslighting” involves systematic reality distortion and attributional reversal, leading to self-doubt and destabilization of autobiographical coherence [8,9].

“Discard” refers to the abrupt withdrawal or termination of a relationship by the abusive individual, often following a period of devaluation, and may involve emotional abandonment, rejection, or reputational harm [35].

“No contact” is a protective strategy in which the affected individual intentionally terminates all communication and interaction with the abusive person in order to prevent further psychological harm and facilitate recovery [11].

“Hoovering” describes attempts by the abusive individual to re-establish contact after separation through apologies, promises of change, or renewed affection intended to re-engage the relationship [36].

“Love bombing” refers to intense early-stage displays of affection, attention, or admiration used to rapidly establish emotional dependency and attachment in the target individual before later cycles of devaluation or control [36].

• For clarity, the key terms used in this manuscript are briefly defined below.
• Functional impairment
• Coping strategies

The semi-structured interview was conducted by a clinician who was not the victim/target’s treating therapist. To avoid role conflict, participation was voluntary and had no impact on the victim/target’s ongoing treatment. Semistructured in-depth interviews are commonly used in qualitative research and are the most frequent qualitative data source in health services research. This method typically consists of a dialogue between researcher and participant, guided by a flexible interview protocol and supplemented by follow-up questions, probes and comments. The method allows the researcher to collect open-ended data, to explore participant thoughts, feelings and beliefs about a particular topic and to delve deeply into personal and sometimes sensitive issues [31,32].

Psychological assessment measures

To ensure comprehensive and multidimensional assessment, validated instruments were grouped into four conceptual domains:

Trauma-related symptomatology

• International Trauma Questionnaire (ITQ) – assesses PTSD and Complex PTSD symptoms according to ICD-11, with 18 items measuring core PTSD clusters and Disturbances in Self-Organization (DSO) clusters. Each item is rated on a 5-point Likert scale [17,37].
• Impact of Event Scale – Revised (IES-R) – intrusion, avoidance, hyperarousal, a 22-item self-report measure assessing subjective distress from traumatic events across three dimensions: Intrusion, Avoidance, and Hyperarousal. Items are rated on a 5-point Likert scale [38,39].
• Trauma Recovery Measure (TRM) – trauma recovery processes, a 15-item self-report questionnaire measuring psychological recovery after trauma, with subscales assessing Validation, Liberation, and Positive Self. Scores track recovery stages from Early to Late [40].

These instruments were selected to evaluate whether symptom presentation aligned with exploratory trauma frameworks.

Workplace psychological abuse and manipulation

• Negative Acts Questionnaire – Revised (NAQ-R) – exposure to workplace bullying behaviors, a 22-item self-report measure assessing exposure to workplace bullying across three dimensions: work-related, person-related, and physically intimidating bullying. Items are rated by frequency, with higher scores indicating greater exposure [41,42].
• Gaslighting at Work Scale (GWS) – measures perceived gaslighting behaviors enacted by supervisors or colleagues, focusing on manipulation, distortion of reality, and psychological undermining within workplace relationships. It captures subtle and overt tactics that destabilize an employee’s sense of competence, memory, and judgment. Eleven items [1–5]. Higher scores reflect greater perceived gaslighting exposure, self-report measure [9].

Emotional distress and psychological functioning

• Depression Anxiety Stress Scales – 21 (DASS-21) – a self-report measure with three subscales: Depression, Anxiety, and Stress. Each subscale includes seven items, and final scores are categorized into five severity levels [43,44].
• CORE-10 – global psychological distress, a 10-item self-report instrument assessing general psychological distress. Items are rated on a 5-point Likert scale, with higher scores indicating greater distress [45,46].
• Perceived Stress Scale (PSS-10) – a 10-item self-report measure assessing perceived stress over the past month. Items are rated on a 5-point Likert scale, with higher scores indicating greater perceived stress. Four items are reverse-scored, and subscale scores capture Perceived Helplessness and Perceived Self-Efficacy [44,47].
• Professional Quality of Life Scale (ProQOL) – a 30-item self-report measure evaluating compassion satisfaction, burnout, and secondary traumatic stress in helping professionals [48, 49].

These instruments assessed current emotional burden, stress reactivity, and occupational impact.

Personality and interpersonal moderators

• Psychological Vulnerability Scale (PVS) – consists of six items assessing cognitive patterns that make individuals more susceptible to stress, particularly dependence on achievement or external validation for self-worth [50,51]. Responses are rated on a five-point Likert scale (1 = “Does not describe me at all” to 5 = “Describes me very well”), with higher scores indicating greater psychological vulnerability.
• Conflict Management Style Questionnaire – was administered to assess the participant’s preferred approach to managing interpersonal conflict. The questionnaire identifies dominant conflict-handling styles and examines how these may vary across relational contexts. Scores ranging from 21 to 25 indicate a very strong preference for a particular conflict style. The self-report format allowed reflection on personal tendencies and behavioral patterns in conflict situations, consistent with exploratory approaches to measuring interpersonal dynamics [33,52].
• Empathy Quotient (EQ) – a 60-item self-report questionnaire measuring both cognitive and affective components of empathy [53], which has demonstrated acceptable reliability and validity across different populations in more recent validation studies [53,54].
• Brief Emotional Intelligence Scale (BEIS-10) – a 10-item self-report instrument developed to assess emotional intelligence (EI) in adults. It is grounded in Salovey and Mayer’s theoretical model of emotional intelligence and measures five dimensions: Appraisal of Own Emotions, Appraisal of Others’ Emotions, Regulation of Own Emotions, Regulation of Others’ Emotions, and Utilization of Emotions. Each subscale consists of two items rated on a 5-point Likert scale ranging from 1 (“strongly disagree”) to 5 (“strongly agree”), with higher scores indicating greater perceived emotional intelligence [55,56].

These measures were included to explore potential moderating factors influencing interpersonal dynamics and stress processing.

The administered instruments (e.g., IES-R, ITQ) are self-report measures reflecting symptom-level equivalents of diagnostic criteria; no structured clinician-administered interview (e.g., CAPS-5) was conducted to formally establish PTSD or Complex PTSD diagnosis.

Data integration strategy

No formal qualitative coding framework was applied; interview data were used for structured clinical contextualization rather than thematic generalization [57,58].

Quantitative scores were interpreted according to exploratory clinical cut-offs where available and integrated with qualitative interview themes to document trauma-related psychological sequelae, differentiate symptom patterns from formal PTSD and Complex PTSD diagnoses, and examine the descriptive construct referred to in lay discourse as “post-narcissistic stress” [59].

The term was used conceptually and descriptively, not diagnostically.

Use of ChatGPT as a self-reflection tool

The patient additionally engaged in non-standardized AI-assisted reflective writing; this was not part of the formal methodology and no data from these interactions were used in the analysis.

The AI-assisted reflection was not treated as a standardized intervention and no data derived from these interactions were used for analysis or outcome evaluation [60].

Qualitative findings

Semi-structured interview

The qualitative analysis identified five interrelated experiential domains.

Early attachment activation and accelerated trust formation: The participant described an initial phase of intensified interpersonal engagement characterized by heightened attention, affirmation, and rapid emotional closeness. She retrospectively used the term “love bombing” to describe this period. She reported a strong perceived alignment of values and professional goals, rapid trust development, and a sense of exceptional emotional understanding.

Behavioral shift and intermittent relational withdrawal: The participant reported a subsequent marked change in the colleague’s behavior, characterized by emotional distancing, reduced warmth, and episodes of being ignored or dismissed, which she described as “silent treatment.” She experienced this shift as abrupt and psychologically destabilizing.

She further described alternating periods of renewed closeness, which she labeled “hoovering,” creating cycles of affection and withdrawal. She reported developing what she termed an “intense, unhealthy emotional attachment” and later used the expression “trauma bonding” to describe difficulty disengaging from the relationship.

Perceptual uncertainty and erosion of self-confidence: During later interactions, the participant described increasing self-doubt and confusion. When attempting to clarify prior events, she reported experiences she referred to as “gaslighting,” including being told she was overly sensitive or had misinterpreted situations.

She began questioning her perceptions and emotional responses. In group settings, she reported feeling undermined and devalued. Over time, she described increased rumination, diminished confidence, and persistent psychological tension within the workplace.

Workplace Stress Escalation and Social Isolation: The participant retrospectively questioned the professional dynamics of the collaboration, reporting that she may have compensated for perceived deficiencies while interpreting her actions as supportive. This realization was associated with shame and self-blame.

She described perceived reputational harm and social exclusion, including colleagues acting against her and spreading false information. These experiences contributed to isolation, emotional distress, and reduced occupational functioning.

Trauma-related psychological sequelae: Following the deterioration of the relationship, the participant reported intrusive thoughts, recurrent distressing memories, nightmares, avoidance behaviors, heightened anxiety, and physiological hyperarousal when reminded of the events.

She described persistent sadness, diminished trust in others, self-blame, and disruption of her sense of identity. Due to cumulative distress and functional impairment, she resigned from her position and relocated. She adopted complete disengagement (“no contact”) as a coping strategy.

She further reported that after implementing “no contact,” the colleague reinitiated communication with renewed validation, which she experienced as emotionally confusing and distressing.

Quantitative findings

Baseline assessment (At relationship termination)

At the time of relationship termination, the participant demonstrated clinically significant distress across multiple domains.

Trauma-related symptomatology

• IES-R: 80 — extremely elevated trauma-related distress
• ITQ: PTSD symptom cluster score consistent with severe PTSD symptomatology, with marked functional impairment CORE-10: 32–35 severe global psychological distress

The trauma symptom pattern included high intrusion, avoidance, hyperarousal, and pronounced sense of threat.

Emotional distress

• DASS-21: Extremely severe overall distress 
• Anxiety: Extremely severe
• Stress: Severe
• Depression: Moderate
• PSS-10: 28 — high perceived stress
• Perceived helplessness was markedly elevated

These findings in this case indicate acute psychological destabilization.

Workplace psychological abuse

• NAQ-R: 92 — high risk / probable exposure to workplace bullying
• Gaslighting at Work Scale GWS: 4 High level of workplace gaslighting- Loss of self-trust 4, Abuse of power 4, 6.

These scores indicate substantial exposure to workplace psychological aggression and destabilization.

Occupational functioning

• ProQOL: Moderate burnout and secondary traumatic stress
• Compassion satisfaction remained moderate despite high distress

Follow-up assessment (6 months post-breakup; 4 months post no-contact)

Marked symptom reduction was observed across domains.

Trauma-related symptoms

• IES-R: Reduced to 30 (clinically significant but markedly decreased)
• ITQ: PTSD symptom severity reduced to moderate range; DSO symptoms minimal
• CORE-10: 3 — within healthy range
• TRM: 73 — advanced stage of recovery

Emotional distress

• DASS-21: Returned to normal range
• PSS-10: Reduced to 8 — low perceived stress
• Helplessness and self-efficacy scores normalized

Workplace abuse indicators

• NAQ-R: Reduced to 24 no significant bullying exposure
• Gaslighting at Work scale GWS: Loss of self-trust 2, Abuse of power 1,1.

Summary of quantitative findings

These findings reflect self-reported symptom severity consistent with trauma-spectrum presentations; however, they should not be interpreted as a formally established psychiatric diagnosis in the absence of a structured clinician-administered interview.

The initial assessment demonstrated severe trauma-related distress, high perceived stress, and strong indicators of workplace psychological abuse. Follow-up assessment demonstrated substantial symptom reduction and functional recovery following cessation of contact and implementation of therapeutic interventions.

Therapeutic interventions and clinical course

The participant engaged in an integrative, trauma-informed psychotherapeutic approach. Interventions included cognitive-behavioral techniques targeting maladaptive beliefs (e.g., excessive self-blame and perceived inadequacy), trauma-focused cognitive restructuring, psychoeducation regarding interpersonal manipulation dynamics, boundary-setting training, and assertiveness development [61–65].

Specific attention was directed toward addressing shame and guilt responses associated with prolonged interpersonal psychological stress. Mindfulness-based techniques and self-compassion practices were incorporated to reduce emotional hyperarousal and support stabilization [66].

Given the occupational context, coping strategies relevant to workplace psychological stress were introduced, including emotional distancing techniques and structured communication strategies [23].

The participant also engaged in peer-support communities during the early phase of recovery, which facilitated normalization of her experiences [67]. A strategy of strict no contact was implemented to reduce ongoing exposure to the identified stressor and to interrupt maladaptive relational cycles [11].

Over the course of treatment, the participant reported gradual reduction in intrusive symptoms, improved emotional regulation, increased clarity in interpersonal boundaries, and partial restoration of occupational self-confidence.

Adjunctive use of generative AI (ChatGPT 5) for structured self-reflection

During the recovery process, the participant utilized a publicly available generative AI platform ChatGPT 5 as a structured self-reflective tool. The AI system was not used to provide therapeutic guidance or clinical advice, nor was it a substitute for psychotherapy. All interactions were non-identifying, and safety monitoring occurred within the clinical context.

The authors report no financial or professional affiliation with the AI platform provider.

This case combines the patient’s longitudinal symptom measurements with qualitative clinical observations to illustrate trauma-related psychological sequelae following prolonged workplace psychological distress [68]. The qualitative findings describe a progression from intense perceived emotional attunement and accelerated interpersonal bonding to relational destabilization, cognitive self-doubt, occupational disruption, and eventual psychological decompensation. These experiential domains closely paralleled the quantitative findings obtained at baseline assessment.

At the time of relationship termination, standardized measures demonstrated severe trauma-related symptomatology.

It is important to note that all quantitative findings were derived from retrospective self-report measures administered after termination of the interpersonal relationship. No structured clinician-administered diagnostic interview (e.g., CAPS-5) was conducted to independently verify PTSD diagnostic criteria [58]. Consequently, symptom severity should be interpreted as self-reported distress consistent with trauma-spectrum presentations rather than as a formally established psychiatric diagnosis.

Importantly, the symptom pattern did not emerge in the context of a single discrete traumatic event but rather followed prolonged interpersonal psychological stress within a professional setting. This aligns with literature describing the cumulative impact of chronic relational stressors and workplace bullying on trauma-related symptom expression [69].

Bridging survivor-informed terminology with empirically grounded trauma frameworks in occupational contexts

Quantitative assessment revealed extreme traumatic stress (IES-R=80), severe overall distress (DASS-21), and PTSD-consistent symptomatology (ITQ). These converging measures indicate a trauma-related profile marked by intrusive recollections, hyperarousal, avoidance coping, emotional dysregulation, and functional impairment, highlighting the clinical necessity for trauma-informed intervention.

The participant used survivor-generated terminology (e.g., “PNSD,” “gaslighting,” “love bombing,” “discard,” “hoovering,” “trauma bond,” “flying monkeys,” “no contact”) to describe her experiences. These terms are not recognized diagnostic categories and are treated here as descriptive labels rather than clinical constructs [4]. The reported symptoms—intrusions, avoidance, hyperarousal, and negative self-concept—overlap with established trauma-spectrum conditions, supporting interpretation within PTSD/Complex PTSD frameworks rather than introducing a novel diagnosis. PNSD is not recognized as a diagnostic category in either DSM-5-TR or ICD-11 [70,71].

Given the reliance on self-report data and the absence of a structured clinical interview, findings should be interpreted descriptively. Accordingly, experiential terminology is translated into empirically grounded constructs to preserve conceptual rigor without assigning diagnostic validity to non-clinical terms.

For example, the phase described as “love bombing” corresponds clinically to accelerated intimacy formation, exaggerated idealization, and rapid attachment activation. The reported experience of “gaslighting” aligns with repeated reality distortion, emotional invalidation, and attributional reversal, processes associated with erosion of self-trust and cognitive destabilization. Alternating warmth and withdrawal reflect intermittent reinforcement dynamics, which are known to intensify attachment bonds under unpredictable conditions. The so-called “discard phase” can be conceptualized as relational rupture accompanied by workplace exclusion and reputational undermining, phenomena overlapping with documented workplace bullying and mobbing dynamics [2,13].

The participant’s symptom constellation—including hyperarousal, intrusive recollections, avoidance behaviors, persistent shame, and diminished self-worth—is consistent with trauma-spectrum responses observed in contexts of prolonged interpersonal psychological abuse. Prior research has demonstrated associations between exposure to narcissistic relational traits and elevated anxiety, depression, and dependency-related distress [7], as well as links between emotional dependence and psychological abuse [72].

When the victim protects her emotional well-being by distancing herself, the person with narcissists' behavior may initiate a renewed cycle of idealization and devaluation, beginning with “hoovering” acting with exaggerated kindness, offering compliments, and creating a false sense of value, often accompanied by guilt-tripping and destabilizing tactics that impair rational decision-making [4].

This study does not seek to legitimize non-clinical terminology as diagnostic categories. Rather, it suggests that survivor-generated language may correspond to identifiable psychological mechanisms measurable within established trauma frameworks. Translating experiential terminology into empirically supported constructs allows clinicians to preserve conceptual rigor while validating lived experience and strengthening therapeutic alliance.

Nevertheless, caution is warranted. The findings derive from a single retrospective self-report account and cannot establish causality, diagnostic certainty, or temporal sequence with precision. The absence of a structured clinical interview limits diagnostic verification, and reliance on self-report introduces potential recall bias, expectancy effects, and post hoc meaning reconstruction. Additionally, no independent corroborative workplace data were available. These methodological constraints restrict generalizability and require conservative interpretation of findings. The findings derive from a single self-report account and cannot establish intent, personality structure, or psychiatric diagnosis of the colleague involved. Future research should examine whether patterns described through survivor terminology systematically map onto validated constructs of workplace relational trauma.

Interpretation of qualitative findings

The qualitative findings suggest a trajectory of progressive relational destabilization culminating in clinically significant trauma-related symptomatology and functional impairment. The early phase of accelerated emotional closeness may be understood as robust attachment system activation, increasing relational salience and psychological investment. When such activation is followed by abrupt withdrawal and unpredictability, the resulting incongruence may intensify affective arousal and attachment-related preoccupation [73].

The alternating pattern of relational reward and withdrawal described by the participant is consistent with intermittent reinforcement processes [24]. Experimental and clinical literature indicates that unpredictable reinforcement schedules can strengthen behavioral persistence and emotional dependency, particularly under conditions of relational ambiguity. In interpersonal contexts, such unpredictability may amplify vigilance, cognitive rumination, and efforts to restore perceived relational stability [24].

Repeated experiences of perceived invalidation and perceptual contradiction may contribute to erosion of self-confidence and increased reliance on external validation. Chronic cognitive dissonance in relational contexts can disrupt coherent self-representation and increase stress reactivity. Over time, sustained ambiguity and perceived social threat may consolidate maladaptive cognitive schemas centered on self-doubt, guilt, and hyper-responsibility [73].

From a neurobiological perspective, prolonged exposure to unpredictable interpersonal stress is associated with dysregulation of stress-response systems. Activation of the amygdala in response to perceived social threat may heighten threat appraisal and emotional reactivity. Recurrent activation of the hypothalamic–pituitary–adrenal (HPA) axis can contribute to altered cortisol regulation and impaired stress recovery [74]. Chronic stress exposure may further increase cumulative allostatic load, reflecting the physiological burden of repeated adaptation to relational threat [3].

Workplace bullying and techniques for coping: detached empathy, reverse DARVO, and the BIFF technique

Recent research has increasingly recognized workplace interpersonal aggression and bullying as significant occupational stressors associated with anxiety, depression, and trauma-related symptoms. Longitudinal studies demonstrate that chronic workplace harassment may produce sustained psychological distress and functional impairment [2], and emerging evidence suggests that exposure to workplace bullying may also be associated with post-traumatic stress symptomatology in affected individuals [75].

To assess the psychological impact of workplace bullying, validated instruments were used. The Professional Quality of Life Scale (ProQOL) [48,49] revealed moderate compassion satisfaction (percentile 11.4), moderate burnout (percentile 62.2), and high secondary traumatic stress (percentile 98.2). The Negative Acts Questionnaire–Revised (NAQ-R) score of 92 confirmed frequent indirect and manipulative bullying [41,42].

Practical strategies for coping with workplace bullying include detached empathy, Reverse DARVO [76]. Detached empathy involves maintaining emotional distance from the perpetrator while acknowledging the situation, which helps the victim/target stay grounded. Reverse DARVO—Detach, Assert, Validate, Observe supports recognition and counteraction of manipulative tactics. Because persons with narcissistic behavior use DARVO (Deny, Attack, Reverse Victim and Offender roles), a defensive pattern that shifts responsibility from the perpetrator to the victim, generating confusion, self-blame, and psychological destabilization [77,78].

Clinical results indicate severe self-reported psychological distress consistent with trauma-spectrum responses in the context of perceived coercive relational dynamics. A gaslighting screening measure indicated a high likelihood of pervasive manipulative interpersonal behavior. Given the reliance on self-report and the exploratory nature of certain measures, interpretations should remain cautious and descriptive rather than confirmatory. Such dynamics may be particularly reinforced in hierarchical workplace environments, where organizational structures can enable bullying and toxic interpersonal conduct [79–81].

No-contact strategy

Although based on a single case, the longitudinal documentation of symptom trajectories provides a structured observation of psychological changes over time [17].

A marked reduction in self-reported symptom severity was observed at six-month follow-up, four months after implementation of a strict no-contact approach. Trauma-related measures (IES-R, ITQ) and global distress indices demonstrated substantial improvement compared to baseline.

However, this association should be interpreted as temporal rather than causal. Symptom reduction coincided with the implementation of strict no-contact boundaries. Symptom reduction occurred in parallel with ongoing trauma-informed psychotherapy, occupational disengagement, relocation, environmental change, and the natural course of recovery processes. In a single-case design without controlled comparison, the independent contribution of relational disengagement cannot be determined [57,58].

It is also important to note that all outcomes were derived from self-report instruments reflecting symptom-level equivalents of diagnostic criteria rather than clinician-administered diagnostic confirmation [58]. Consequently, improvement should be understood as reduction in perceived trauma-related distress rather than remission of a formally established psychiatric diagnosis.

From a descriptive standpoint, removal from a chronic interpersonal stressor and establishment of firm relational boundaries coincided with progressive psychological stabilization, consistent with general trauma-recovery principles emphasizing environmental safety [17].

Whether boundary-based disengagement independently facilitates recovery in similar contexts remains a question for future controlled longitudinal research.

Adjunctive use of ChatGPT as a self-reflection tool

The patient additionally engaged in non-standardized AI-assisted reflective writing; this was not part of the formal methodology and no data from these interactions were used in the analysis.

The AI-assisted reflection was not treated as a standardized intervention and no data derived from these interactions were used for analysis or outcome evaluation.

The patient expressed an interest in using ChatGPT as an adjunctive self-reflection tool and therefore, prior to implementation, an explicit agreement was reached with the patient, that ChatGPT’s function is limited to facilitating reflective processes (e.g., clarification, summarization, and neutral prompting). The system does not provide therapeutic guidance and is not conceptualized as a substitute for psychotherapy or any form of professional mental health treatment together with the patient that safety considerations would be monitored throughout the process. The use of ChatGPT follows predefined ethical boundaries, and patient safety is monitored during its use.

All interactions should be non-identifying, and no personally identifiable information is intentionally entered into the system.

The publicly available version of ChatGPT accessible through the online ChatGPT interface at the time of use is employed. As model versions within the interface are managed and updated by OpenAI, users do not have the ability to manually select, control, or modify the underlying model architecture.

Model parameters such as temperature and top-p, which influence response variability, are internally configured by OpenAI and are not adjustable within the standard ChatGPT interface. The API version is not used.

The AI system was used solely as a structured reflective writing aid and was not intended to provide diagnosis, psychotherapy, or clinical decision-making.

The AI-assisted reflection was not treated as a standardized intervention and no data derived from these interactions were used for analysis or outcome evaluation.

The authors and the patient declare no financial relationship with OpenAI and receive no compensation or commission related to the use of publicly available generative AI systems.

Comparison with existing literature

Recent research has increasingly recognized workplace interpersonal aggression and bullying as significant occupational stressors associated with anxiety, depression, and trauma-related symptoms. Longitudinal studies demonstrate that chronic workplace harassment may produce sustained psychological distress and functional impairment comparable to other major stress exposures [75,82,83].

Within organizational psychology, workplace bullying is characterized by repeated exposure to hostile interpersonal behaviors such as humiliation, social exclusion, persistent criticism, or reputational undermining within professional relationships [2]. Prolonged exposure to such dynamics has been associated with increased psychological distress, emotional exhaustion, and a range of adverse mental health outcomes in affected employees [2]. Such chronic workplace interpersonal aggression may also function as a form of relational trauma, particularly when exposure is frequent and involves ongoing interpersonal threat, contributing to symptoms consistent with trauma-related presentations, including hyperarousal, intrusive recollections, and affective dysregulation [1].

From the perspective of emotional resource dynamics, workplace bullying can also be understood as a process that progressively depletes employees’ emotional capacity, destabilizing what has been described in discourse as the “emotional economy” of work and undermining the psychological safety necessary for healthy organizational functioning [84,85]. In the present case, prolonged exposure to destabilizing interpersonal dynamics may have contributed to a gradual erosion of emotional resources and increased vulnerability to trauma-related symptoms.

Contemporary organizational psychology recognizes that narcissism is a multidimensional construct encompassing different expressions, most notably grandiose and vulnerable forms [86]. While some expressions of narcissism may be associated with ambition and leadership drive, darker forms—characterized by entitlement, low empathy, rivalry, and arrogance—can contribute to interpersonal hostility and toxic workplace dynamics [86]. Such individuals often maintain a carefully constructed self-image of superiority and competence, relying on external validation while simultaneously devaluing others, which reinforces hierarchical and psychologically unsafe workplace environments [86].

Such traits may facilitate manipulative relational patterns, including gaslighting and reputational attacks, which are frequently described in workplace bullying dynamics [79] and are conceptually related to mechanisms such as denial, attack, and victim reversal described in the literature [78].

At the same time, emerging literature has begun to explore the role of artificial intelligence and conversational agents in mental health support. Systematic reviews suggest that AI-based interventions, including chatbot-supported programs and AI-enhanced cognitive-behavioral approaches, may improve symptoms of stress, anxiety, and burnout in some populations, although evidence remains preliminary and heterogeneous [87,88].

Early empirical studies further suggest that conversational AI systems may facilitate emotional expression, psychoeducation, and reflective processing, potentially supporting narrative exploration and cognitive restructuring in individuals experiencing psychological distress [89,90]. A multidisciplinary alliance of clinicians, ethicists, data scientists, and policy experts has emphasized the need to develop AI applications in mental health care that prioritize ethical safeguards, patient-centered design, and trauma-informed principles [91–93].

The present case report contributes to the literature by describing a different model: the use of generative AI as a structured reflective adjunct within clinician-led psychotherapy in the context of chronic workplace interpersonal trauma.

By integrating trauma-informed clinical frameworks with AI-assisted reflective dialogue, this study illustrates how conversational AI may function as a supportive tool for narrative processing between psychotherapy sessions rather than as a replacement for clinical care. These observations suggest a potential model for integrating generative AI as a reflective support instrument within trauma-informed psychotherapy, warranting further systematic investigation.

First, this is a single-case report, and the findings cannot be generalized [57,58]. The purpose of this study is illustrative and exploratory rather than confirmatory. No control condition was included, and no comparative case was available; therefore, the relative contribution of specific therapeutic components cannot be determined.

Second, all quantitative measures were based on retrospective self-report completed after the termination of the interpersonal relationship. Although validated instruments were used, recall bias, subjective interpretation, expectancy effects [94,95], and post hoc meaning-making cannot be excluded [95]. During periods of active psychological manipulation, individuals often lack sufficient psychological distance or insight to complete structured assessments, as experiences may feel confusing or destabilizing. Consequently, the temporal gap between exposure and assessment may affect the precision of symptom reconstruction.

Third, no independent structured clinical interview (e.g., Clinician-Administered PTSD Scale; CAPS-5) was conducted to corroborate self-reported symptomatology [58]. The absence of a structured clinician-administered diagnostic interview limits the ability to confirm whether self-reported symptom clusters met full diagnostic thresholds for PTSD or Complex PTSD according to DSM-5-TR or ICD-11 criteria.

Fourth, to study psychological processes directly, we should observe them where they occur: within the individual and overtime. In this case study no baseline psychological data prior to the interpersonal stressor were available, and no repeated pre-exposure measurements were conducted. As a result, pre-existing vulnerability factors cannot be definitively disentangled from stress-induced psychological changes, and within-person symptom trajectories across multiple time points cannot be established [96].

Fifth, no independent or objective workplace data were available (e.g., organizational reports, HR documentation, third-party observations, or formal complaint records). All descriptions of interpersonal dynamics are based solely on the participant’s account and should not be interpreted as diagnostic characterizations, statements of intent, or assertions regarding stable personality traits of the colleague involved. Therefor common method bias may occur when the data for the predictor and criterion variables come from the same person using the same response method. Uncontrolled method variance may produce biased estimates of the reliability and validity of the examined constructs and erroneous parametric estimates of the relationships between two constructs [97].

Sixth, the observed temporal association between the implementation of “no contact” and symptom reduction does not establish causality. Symptom improvement may reflect multiple interacting factors, including psychotherapy, environmental changes, relocation, social support, expectancy effects [94,95], and the natural course of recovery.

Seventh, the study highlights a conceptual and methodological gap between survivor-informed terminology used in everyday discourse and formal psychological nomenclature. While these expressions capture lived experiences central to survivor narratives, their construct boundaries and empirical status remain underdeveloped [98]. Language choice and self-identity can be shaped by both perceptions of victim-survivors, as well as language usage by professionals that victim-survivors may encounter and broader society. Victim-survivor language choices cannot be viewed solely in isolation as an individual language choice but are impacted by discursive contexts, perceptions and underlying associations [98].

Eighth, the exploratory use of a generative AI platform as a reflective adjunct represents an emerging methodological area without standardized protocols, predefined interaction models, or reproducibility frameworks. No structured prompt design or fidelity monitoring was implemented, limiting replicability. Observed effects may be influenced by digital literacy and individual differences in narrative engagement where AI solutions can also exploit cognitive limitations, making one vulnerable to illusions of understanding. Such illusions obscure the scientific community's ability to see the formation of scientific monocultures, in which some types of methods, questions and viewpoints come to dominate alternative approaches, making science less innovative and more vulnerable to errors [99].

Ninth, the contextual specificity of workplace psychological abuse within a particular organizational environment limits transferability to other relational contexts (e.g., intimate partnerships or family systems). Organizational hierarchy, professional identity, and reputational concerns may uniquely shape trauma responses in occupational settings [2].

Suggested future research directions

Future research should incorporate larger samples, structured clinical interviews, prospective longitudinal designs with multiple assessment points, validated measures of interpersonal manipulation, and integration of objective organizational data where feasible.

This case demonstrates that prolonged workplace interpersonal psychological distress may be associated with clinically significant trauma-related symptoms and functional impairment. Symptom patterns were consistent with trauma-spectrum presentations and showed substantial reduction over time following disengagement from the stressor and implementation of trauma-informed therapeutic interventions.

However, given the single-case design and concurrent interventions, no causal relationship can be established. Findings are based on self-report measures and should be interpreted as indicative of changes in perceived distress rather than formal diagnostic outcomes, supports the clinical relevance of interpreting survivor-described experiences within trauma-informed frameworks.

Overall, the case supports the clinical relevance of recognizing trauma-related processes in the context of prolonged workplace interpersonal stress.

Future research employing structured clinical interviews, longitudinal designs, and validated measures of interpersonal psychological manipulation is required to further clarify mechanisms and generalizability.

This case report describes standard clinical assessment and treatment procedures and does not constitute biomedical research involving human participants as defined under Danish law. Therefore, formal approval from a Regional Research Ethics Committee was not required.

Written informed consent for publication was obtained from the patient. All identifying information has been removed or modified to ensure anonymity in accordance with the General Data Protection Regulation (GDPR).

The use of AI-assisted self-reflection occurred as part of routine clinical practice. No experimental intervention was conducted, and patient safety was continuously monitored throughout the treatment process.

The patient expressed interest in using ChatGPT-5 as an adjunctive tool for self-reflection. An explicit agreement was established that ChatGPT-5 would be used solely to facilitate reflective processes (e.g., clarification, summarization, and neutral prompting) and not to provide therapeutic guidance or replacement of psychotherapy or professional mental health treatment. Patient safety and ethical boundaries were monitored throughout its use.

All interactions were conducted without entering personally identifiable information. The publicly available ChatGPT-5 interface was used. Model versions and parameters (e.g., temperature, top-p) are managed by OpenAI and cannot be modified by users; the API version was not used.

The authors and the patient declare no financial relationship with OpenAI and received no compensation related to the use of publicly available generative AI systems.

The authors and the patient have no financial connection with the company OpenAI and do not make any commission if a reader chooses to start using Public Generative AI solutions.

No conflicts of interest.

The authors have no disclosures to make.

Ema Nicea Gruber: Methodology, Writing – original draft, review & editing, Project Administration, Methodology, Validation, Conceptualization.

Sanja Martic Biocina: Writing –original draft, review & editing, Formal analysis, Conceptualization, Project administration, Methodology.