An Emotion And Cognitive Based Analysis Of Mental Health Disorders Fro

Background: The relationship between emotional symptoms and cognitive impairments in major depressive disorder (MDD) is key to understanding cognitive dysfunction and optimizing recovery strategies. This study investigates the relationship between subjective and objective cognitive functions and emotional symptoms in MDD and evaluates their contributions to social functioning recovery. Methods: The Prospective Cohort Study of Depression in China (PROUD) involved 1,376 MDD patients, who underwent 8 weeks of antidepressant monotherapy with assessments at baseline, week 8, and week 52. Measures included the Hamilton Depression Rating Scale (HAMD-17), Quick Inventory of Depressive Symptomatology-Self Report (QIDS-SR16), Chinese Brief Cognitive Test (C-BCT), Perceived Deficits Questionnaire for Depression-5 (PDQ-D5), and Sheehan Disability Scale (SDS). Cross-lagged panel modeling (CLPM) was used to analyze temporal relationships. Results: Depressive symptoms and cognitive measures demonstrated significant improvement over 8 weeks (p < 0.001).

Baseline subjective cognitive dysfunction predicted depressive symptoms at week 8 (HAMD-17: β = 0.190, 95% CI: 0.108-0.271; QIDS-SR16: β = 0.217, 95% CI: 0.126-0.308). Meanwhile, baseline depressive symptoms (QIDS-SR16) also predicted subsequent subjective cognitive dysfunction (β = 0.090, 95% CI: 0.003-0.177). Recovery of social functioning was driven by improvements in depressive symptoms (β = 0.384, p < 0.0001) and subjective cognition (β = 0.551, p < 0.0001), with subjective cognition contributing more substantially (R2 =... 0.075). Conclusions: Subjective cognitive dysfunction is more strongly associated with depressive symptoms and plays a significant role in social functioning recovery, highlighting the need for targeted interventions addressing subjective cognitive deficits in MDD. Keywords: Chinese Brief Cognitive Test (C-BCT); Hamilton Depression Rating Scale (HAMD-17); Perceived Deficits Questionnaire for Depression-5 (PDQ-D5); Quick Inventory of Depressive Symptomatology-Self Report (QIDS-SR16); Sheehan Disability Scale (SDS); cognitive impairment; cohort study; depressive symptoms;...

Background: Emotional disorders (EDs) are the most prevalent worldwide. Despite psychotherapies are their treatment of choice, there are difficulties to apply them properly in mental health services. Since literature shows that cognitive processes are associated with anxiety and depressive symptoms, more information is needed in order to improve psychological treatments. Aims: To determine the relation between cognitive factors with specific and non-specific ED symptoms in order to promote the development of accurate psychological treatments. Methods: We analyzed the relation between rumination, worry, and metacognition with generalized anxiety, panic, and depression disorder symptoms from a clinical sample of 116 individuals through correlation and linear regression analyses. Results: Although each specific disorder had a closer link with a particular cognitive process, all general ED symptoms were associated with the three cognitive factors studied.

Conclusions: For "pure" disorders, targeting a concrete cognitive process might be an optimal therapeutic option. However, due to the high comorbidity among EDs, we support the dissemination of the transdiagnostic treatment approach in which all cognitive factors are taken into account. npj Mental Health Research volume 4, Article number: 63 (2025) Cite this article Accurate psychiatric diagnosis and assessment are crucial for effective treatment. However, current diagnostic approaches heavily rely on subjective observations constrained by time and clinical resources. This study investigates the potential of using Large Language Models (LLMs) to identify the symptoms in psychiatrist-patient dialogues and use them as intermediate features to predict the diagnostic labels.

We collected audio recordings of 1160 outpatients with depressive disorder and anxiety disorder. LLMs were trained and utilized to identify clinical symptoms, rate assessment scales, and an ensemble learning pipeline was designed to classify diagnostic results and symptoms with 10-fold cross-validation. The system achieved 86.9% accuracy for identifying the appearance of clinical annotations and 74.7% (77.2%) accuracy for identifying symptoms of anxiety (depression). In addition, analysis of LLM-generated features shows that depression cases exhibited prominent markers of anhedonia and decreased volition, whereas anxiety disorders were characterized by tension and an inability to relax. Depression and anxiety disorders represent two of the most prevalent mental health conditions globally. Globally, it is estimated that over 300 million people suffer from major depressive disorders, which is equivalent to 4.4% of the world’s population.

A similar number of people suffer from anxiety disorders, often with co-occurring depression1. The emerging field of digital phenotyping, which involves the nuanced quantification of human phenotypic expression at the individual level through digital device data, offers a quantitative approach to longitudinal observation2. The emerging field of digital phenotyping, characterized by continuous and nuanced quantification of human phenotypic expression at the individual level by leveraging digital device data, provides a quantitative approach for longitudinal observation2. Researchers have demonstrated that social signals (e.g., linguistics, speech, etc.) play a crucial role in the diagnosis and assessment of patients with depression and anxiety3,4. In particular, the content of a patient’s speech provides rich information about their mental state, cognitive patterns, and emotional experiences5,6. The linguistic features, topic choices, and narrative structures employed by individuals can offer valuable insights into their psychological well-being6.

Recent advances in NLP, particularly in LLMs such as GPT7, Gemini8, and Qwen9, demonstrate diverse capabilities in clinical reasoning, social media analysis, and psychiatric education10, which could potentially provide objective, data-driven insights in psychiatry. Moreover, LLMs are able to process, generate, and respond to natural language inputs, which fit naturally into the NIMH’s Research Domain Criteria (RDoC) framework, which suggests new ways of classifying mental disorders based on... In recent psychiatric studies, these LLMs excel at understanding and generating complex linguistic patterns with human-like performance, making them widely explored for social media content analysis12,13, treatment performance enhancement14,15,16, chat counselor17,18, and supporting clinical... Although LLMs demonstrate linguistic understanding and generation, they remain relatively scarce in producing objective digital biomarkers in psychiatry21. Studies have shown that the speech of patients with depression and anxiety contains distinctive quantitative verbal and nonverbal digital markers compared to healthy controls4,6, but these characteristics often remain too subtle for humans to... LLM is able to generate diagnostic results and provide reasoning steps, benefiting from a large amount of pre-training data.

However, the interpretation and alignment of answers or decisions generated by LLM remain challenging23. Moreover, most studies on depression and anxiety rely primarily on two data sources: social media and structured clinical reports, and are often constrained by limited data availability3. Distinguishing between depression and anxiety in clinical settings remains challenging due to the overlap of symptoms and the high comorbidity rate, with limited research on the discovery of objective biomarkers for both conditions21. In addition, during clinical interviews, psychiatrists translate patients’ informal symptom descriptions into professional diagnostic terminology; however, there remains a lack of approaches to automatically and effectively bridge this “semantic gap" between patients and clinicians.

People Also Search

• An emotion and cognitive based analysis of mental health disorders from ...
• A longitudinal analysis of the relationship between emotional symptoms ...
• (PDF) Cognitive processes associated with emotional disorders ...
• PDF Cognitive emotional processing across mood disorders
• PDF Associations Between Cognitive Appraisals and Emotions: A Meta-Analytic ...
• Cognitive processes associated with emotional disorders: implications ...
• Identifying psychiatric manifestations in outpatients with depression ...