Journal of Personality and Psychosomatic Research (JPPR)

A Network‑Informed Machine Learning Model of How Trait Negative Affectivity and Pain Catastrophizing Predict Somatic Symptom Intensity

Maia Bolashvili iD , Nino Beridze ^* iD

The objective of this study was to estimate a psychological network of trait negative affectivity and pain catastrophizing to identify critical bridge nodes, and subsequently utilize these specific structural bridges as features in a machine learning model to predict the continuous intensity of somatic symptoms. This cross-sectional study recruited a community sample of adults from Georgia who completed standardized online self-report questionnaires, including the Positive and Negative Affect Schedule (PANAS-NA), the Pain Catastrophizing Scale (PCS), and the Patient Health Questionnaire (PHQ-15). Data analysis utilized a two-phase methodology: first, a regularized partial correlation network analysis was conducted to map item-level interactions and compute bridge expected influence; second, machine learning algorithms (Random Forest, Support Vector Regression, Elastic Net) were trained using the identified bridge nodes as input features to predict somatic symptom intensity. Macroscopic analyses demonstrated robust positive correlations between negative affectivity, catastrophizing, and somatic burden. Network analysis revealed that the emotional state of feeling “afraid” and the cognitive appraisal of “helplessness” served as the paramount structural bridge nodes connecting the two psychological constructs. The network-informed Random Forest model demonstrated strong predictive performance on unseen test data ( ), with feature importance metrics confirming that “helplessness” was the most critical individual predictor, independently accounting for of the model’s out-of-sample predictive accuracy. Targeted clinical interventions that specifically dismantle the catastrophic cognition of helplessness may prove more efficacious in reducing somatic symptom severity than broad treatments aimed at generalized negative affect.

Modeling the Influence of Fear‑Avoidance Beliefs and Cognitive Biases on Psychosomatic Symptom Maintenance Using Machine Learning

Sophie Nakamanya iD , Andrei Dumitrescu ^* iD , Pascal Reinert iD

This study aimed to model the influence of fear‑avoidance beliefs and cognitive biases on psychosomatic symptom maintenance using machine learning techniques. A cross‑sectional study was conducted among 512 adults recruited from the general population in Romania. Psychosomatic symptom severity was assessed using the Patient Health Questionnaire‑15 (PHQ‑15), fear‑avoidance beliefs were measured with the Fear‑Avoidance Beliefs Questionnaire (FABQ), and health anxiety and cognitive bias related to bodily sensations were assessed using the Health Anxiety Inventory and a cognitive bias measure. Data were analyzed using several machine learning algorithms, including Random Forest, Gradient Boosting, Support Vector Regression, and Linear Regression. Model performance was evaluated using mean squared error (MSE), root mean squared error (RMSE), and the coefficient of determination (R²), and feature importance analysis was conducted to identify the most influential predictors of psychosomatic symptom severity. After data screening, the final sample consisted of 498 participants (mean age = 36.8 ± 11.4 years). The mean psychosomatic symptom score measured by the PHQ‑15 was 10.9 ± 5.7. Significant positive correlations were observed between psychosomatic symptoms and health anxiety (r = 0.61), cognitive bias (r = 0.57), and fear‑avoidance beliefs (r = 0.52). Among the machine learning models, the Random Forest algorithm demonstrated the best predictive performance, explaining 68% of the variance in psychosomatic symptom severity (R² = 0.68). Feature importance analysis indicated that health anxiety, cognitive bias, and fear‑avoidance beliefs were the most influential predictors of psychosomatic symptoms. Fear‑avoidance beliefs and cognitive biases play a significant role in the maintenance of psychosomatic symptoms.

Predicting Somatic Symptom Disorder Using Facets of Neuroticism, Somatosensory Amplification, and Interoceptive Sensitivity: An Explainable Machine Learning Analysis

Aina Syafiqah Noor ^* iD , Syarifah Maisarah iD , Sanduni Jayawardena iD

https://doi.org/10.61838/kman.jppr.5172

The objective of this study was to utilize an explainable machine learning approach to predict Somatic Symptom Disorder by delineating the complex, non-linear contributions of specific neuroticism facets, somatosensory amplification, and multidimensional interoceptive sensitivity. A quantitative, cross-sectional design was employed with a sample of Malaysian adults. Data were collected using the Patient Health Questionnaire-15 (PHQ-15), the Neuroticism domain of the NEO-PI-R, the Somatosensory Amplification Scale (SSAS), and the Multidimensional Assessment of Interoceptive Awareness (MAIA). Predictive modeling was conducted using advanced ensemble tree-based algorithms (Random Forest, Gradient Boosting, and XGBoost), and model interpretability was achieved using SHapley Additive exPlanations (SHAP) to evaluate the marginal contribution of each psychological feature. The XGBoost algorithm demonstrated superior predictive performance, achieving an overall accuracy of and an Area Under the Curve (AUC) of . SHAP explainability metrics revealed that Somatosensory Amplification was the most profound positive predictor of somatic distress, followed by the specific Anxiety and Vulnerability facets of neuroticism. Conversely, adaptive interoceptive capacities, specifically the “Not-Worrying” and “Body Trusting” dimensions, emerged as the strongest negative (protective) predictors against Somatic Symptom Disorder classification. Explainable machine learning demonstrates that specific sensory appraisal mechanisms and interoceptive deficits are more critical than broad personality domains in driving somatic distress, providing highly specific targets for precision psychological interventions.

Predicting Stress-Induced Somatic Symptoms from Personality and Behavioral Indicators Using Machine Learning

María José Ibañez iD , Becky Lima ^* iD , Daniela Cevallos iD

https://doi.org/10.61838/kman.jppr.5157

This study aimed to predict the severity of stress-induced somatic symptoms using machine learning models applied to personality traits and behavioral data. A cross-sectional study was conducted with 1248 Brazilian adults using digital surveys that included the Patient Health Questionnaire for Somatic Symptoms (PHQ-15), the Big Five Inventory, and a behavioral assessment. Predictive modeling involved training and comparing Multiple Linear Regression, Support Vector Regression (SVR), Random Forest, and eXtreme Gradient Boosting (XGBoost) algorithms on an 80/20 train-test split (test set n=250), evaluating performance via R^2, MAE, and RMSE. The XGBoost model demonstrated superior predictive performance (R^2=0.684, MAE=2.15, RMSE=2.61) for somatic symptom severity (M=9.42, SD=4.65). Feature importance analysis ranked Neuroticism (34.5%) as the strongest predictor, followed by Sleep Duration (18.2%), Conscientiousness (14.6%), and Physical Exercise (11.3%), aligning with significant bivariate correlations for Neuroticism (r=0.58), sleep duration (r=-0.41), and physical activity (r=-0.35). Machine learning algorithms effectively predict somatic symptom severity, highlighting the paramount influence of neuroticism and the protective role of modifiable health behaviors.

Personality‑Based Digital Phenotyping of Psychosomatic Health Using Machine Learning

Lena Hoffmann ^* iD

https://doi.org/10.61838/kman.jppr.5163

This study aimed to investigate whether integrating personality traits with smartphone‑derived digital phenotyping indicators can predict psychosomatic health using machine learning models. A cross‑sectional study was conducted with 524 adult participants recruited from Germany. Personality traits were assessed using the Big Five Inventory (BFI‑44), psychosomatic symptoms were measured using the Patient Health Questionnaire (PHQ‑15), and psychological distress was evaluated using the Depression Anxiety Stress Scales (DASS‑21). Participants also installed a custom smartphone application that passively collected behavioral data for four weeks, including daily phone usage duration, nighttime phone inactivity, communication frequency, response latency, and mobility variability. Data analysis involved descriptive statistics, correlation analyses, and machine learning modeling using Random Forest, Support Vector Machine, Logistic Regression, and Gradient Boosting algorithms implemented in Python with scikit‑learn. Model performance was evaluated using accuracy, precision, recall, F1‑score, and AUC‑ROC metrics. The sample consisted of 524 participants (mean age = 29.4 ± 7.8 years), including 56% females and 44% males. Moderate to severe psychosomatic symptoms were identified in 31.5% of participants based on PHQ‑15 scores. Neuroticism showed significant positive correlations with psychosomatic symptoms (r = 0.46), depression (r = 0.49), anxiety (r = 0.52), and stress (r = 0.47), while conscientiousness and agreeableness showed negative correlations with psychosomatic symptoms (r = −0.28 and r = −0.21, respectively). Participants with higher psychosomatic symptom scores demonstrated longer daily smartphone usage (mean = 5.2 hours vs. 3.7 hours), reduced nighttime inactivity periods (mean = 5.8 hours vs. 7.1 hours), and lower mobility variability. Among the predictive models, Gradient Boosting achieved the best performance with an accuracy of 0.82, F1‑score of 0.80, and AUC‑ROC of 0.87, outperforming Random Forest (accuracy = 0.79), Support Vector Machine (accuracy = 0.76), and Logistic Regression (accuracy = 0.73). The findings suggest that combining personality assessments with smartphone‑based digital phenotyping can effectively predict psychosomatic health and may support early identification of psychosomatic risk through machine learning approaches.

Modeling Personality–Emotion–Somatic Symptom Pathways with Explainable Machine Learning

Mariana Torres ^* iD , Fabricio Peixoto iD , Jaime Martínez-Pacheco Martínez-Pacheco iD

https://doi.org/10.61838/kman.jppr.5164

This study aimed to examine the pathways linking personality traits, emotional distress, and somatic symptoms using explainable machine learning techniques. This cross‑sectional study included 1,024 adults from Mexico (54.3% women; mean age = 34.7 years, SD = 11.2). Participants completed validated psychological instruments, including the Big Five Inventory (BFI‑44) to assess personality traits, the Depression Anxiety Stress Scales (DASS‑21) to measure emotional distress, and the Patient Health Questionnaire Somatic Symptom Scale (PHQ‑15) to assess somatic symptom severity. Descriptive statistics and Pearson correlations were first conducted to examine relationships among variables. Subsequently, machine learning models—Random Forest, Gradient Boosting, and XGBoost—were developed to predict somatic symptoms. Model performance was evaluated using the coefficient of determination (R²), and SHapley Additive exPlanations (SHAP) were used to identify the relative importance of predictors and interpret the models. Correlation analyses indicated significant positive associations between neuroticism, depression, anxiety, stress, and somatic symptoms (r range = .29–.48, p < .001). Neuroticism showed the strongest correlations with anxiety (r = .48) and somatic symptoms (r = .41). Among the machine learning models, XGBoost demonstrated the best predictive performance (R² = 0.42), followed by Gradient Boosting (R² = 0.39) and Random Forest (R² = 0.36). SHAP analyses revealed that anxiety and neuroticism were the most influential predictors of somatic symptoms, followed by stress and depression. Interaction analysis suggested that individuals with high neuroticism combined with high anxiety exhibited nearly twice the predicted level of somatic symptom severity compared to individuals with low scores on these variables. The findings highlight the central role of emotional distress and neuroticism in predicting somatic symptoms and demonstrate the value of explainable machine learning for identifying complex psychosomatic pathways.

Modeling the Contribution of Cognitive Fusion, Rumination, and Worry to Psychosomatic Distress: An Explainable Machine Learning Study

Ardit Hoxha iD , Yasmine Ben Salah ^* iD , Nouhayla Hafidi iD

https://doi.org/10.61838/kman.jppr.5165

The present study aimed to model the unique and interactive contributions of cognitive fusion, rumination, and worry to psychosomatic distress within a Tunisian adult sample by applying and interpreting an explainable machine learning framework. A cross-sectional design was utilized to evaluate 1254 adult participants (M_age=32.4, SD=8.7) from Tunisia, recruited via snowball sampling. Self-report data were collected using the Cognitive Fusion Questionnaire (CFQ), Ruminative Responses Scale (RRS), Penn State Worry Questionnaire (PSWQ), and Patient Health Questionnaire-15 (PHQ-15). The data analysis employed an advanced machine learning pipeline, comparing Random Forest, Support Vector Regression, and eXtreme Gradient Boosting (XGBoost) algorithms. The optimal model was subsequently interpreted using the SHapley Additive exPlanations (SHAP) framework to extract global feature importance, non-linear thresholds, and interaction effects. Bivariate correlations demonstrated significant positive associations between all cognitive variables and psychosomatic distress (r=0.51 to 0.62). The XGBoost algorithm achieved the highest predictive performance (R^2=0.463, RMSE=2.41). SHAP global feature importance analysis ranked Worry as the top predictor of psychosomatic distress (Mean ∣SHAP∣=1.45), followed sequentially by Rumination and Cognitive Fusion. Furthermore, SHAP dependence plots revealed critical non-linear dynamics, identifying a specific threshold where worry scores exceeding 55 exponentially accelerated physical distress, as well as a significant interaction effect indicating that high cognitive fusion severely amplifies the detrimental somatic impact of rumination. Treating severe psychosomatic distress requires targeted interventions that rapidly reduce worry below critical severity thresholds and utilize cognitive defusion strategies to break the compounding cycle of rumination and psychological entanglement.

Using Machine Learning to Identify Latent Profiles of Somatosensory Amplification and Catastrophic Misinterpretation of Bodily Cues in Psychosomatic Patients

Iheanyi Iwueze iD , Zanele Mthembu ^* iD

https://doi.org/10.61838/kman.jppr.5171

The study aimed to identify distinct latent profiles of psychosomatic patients based on their levels of somatosensory amplification and catastrophic misinterpretation of bodily cues using a person-centered machine learning approach. This cross-sectional study involved a sample of psychosomatic patients recruited from clinical settings in South Africa. Participants were assessed using the Somatosensory Amplification Scale (SSAS) and the Body Sensations Questionnaire (BSQ). Statistical analysis was performed using Gaussian Mixture Modeling (Latent Profile Analysis) to identify unobserved subpopulations. Model selection was determined by fit indices, including the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and the Lo-Mendell-Rubin Likelihood Ratio Test (LMR-LRT). A strong significant positive correlation was observed between somatosensory amplification and the catastrophic misinterpretation of bodily cues ( , ). Latent Profile Analysis identified an optimal three-profile solution: “Low Symptom Burden” ( ; ), “Moderate Symptom Burden” ( ; ), and “Severe Symptom Burden” ( ; ). Comparative analyses revealed that the “Severe Symptom Burden” profile was characterized by a significantly higher proportion of female participants, higher unemployment rates, and longer illness durations ( ), whereas age and educational background did not significantly differentiate the groups. The identification of three distinct latent profiles underscores the clinical heterogeneity of psychosomatic populations and highlights the necessity for tailored, precision-based psychological interventions that address specific levels of cognitive-perceptual distortion.

Journal of Personality and Psychosomatic Research (JPPR)

About the Journal

Current Issue

Articles

Predicting Somatic Complaints from Personality Traits and Smartphone-Derived Behavioral Data Using Machine Learning

Personality-Driven Adaptive Psychosomatic Treatment Planning via Artificial Intelligence

Trait Emotional Stability as a Protective Factor Against Psychosomatic Symptom Amplification: A Multilevel Mediation Analysis

Alexithymia, Trait Avoidance, and Somatic Distress: A Transdiagnostic Model of Emotion–Body Disconnection

Predicting Stress-Induced Somatic Symptoms from Personality and Behavioral Indicators Using Machine Learning

Modeling Personality–Emotion–Somatic Symptom Pathways with Explainable Machine Learning

Personality‑Based Digital Phenotyping of Psychosomatic Health Using Machine Learning

Modeling the Contribution of Cognitive Fusion, Rumination, and Worry to Psychosomatic Distress: An Explainable Machine Learning Study

Using Machine Learning to Identify Latent Profiles of Somatosensory Amplification and Catastrophic Misinterpretation of Bodily Cues in Psychosomatic Patients

Predicting Somatic Symptom Disorder Using Facets of Neuroticism, Somatosensory Amplification, and Interoceptive Sensitivity: An Explainable Machine Learning Analysis

Language

Keywords

Latest publications

Developed By

Information

A Network‑Informed Machine Learning Model of How Trait Negative Affectivity and Pain Catastrophizing Predict Somatic Symptom Intensity

Modeling the Influence of Fear‑Avoidance Beliefs and Cognitive Biases on Psychosomatic Symptom Maintenance Using Machine Learning

Predicting Somatic Symptom Disorder Using Facets of Neuroticism, Somatosensory Amplification, and Interoceptive Sensitivity: An Explainable Machine Learning Analysis

Predicting Stress-Induced Somatic Symptoms from Personality and Behavioral Indicators Using Machine Learning

Personality‑Based Digital Phenotyping of Psychosomatic Health Using Machine Learning

Modeling Personality–Emotion–Somatic Symptom Pathways with Explainable Machine Learning

Modeling the Contribution of Cognitive Fusion, Rumination, and Worry to Psychosomatic Distress: An Explainable Machine Learning Study

Using Machine Learning to Identify Latent Profiles of Somatosensory Amplification and Catastrophic Misinterpretation of Bodily Cues in Psychosomatic Patients