Machine Learning Models for Early Detection of Couples at Risk for Emotional Disengagement
Bibliographic record
Abstract
The objective of this study is to develop and validate machine learning models capable of accurately identifying couples at early risk for emotional disengagement based on psychological, behavioral, and communication-related indicators. This predictive cross-sectional study included 156 couples (N = 312 individuals) from across Canada who completed validated measures assessing emotional disengagement, relationship satisfaction, conflict frequency, attachment dimensions, daily supportive behaviors, and communication responsiveness. Data were collected through secure online surveys, and dyadic datasets were prepared using actor–partner structures to preserve relational interdependence. Five machine learning algorithms—logistic regression, random forest, gradient boosting, support vector classifier, and a neural network model—were trained on cleaned and normalized datasets, with performance evaluated through accuracy, precision, recall, F1 score, and area under the ROC curve. Cross-validation procedures ensured the robustness and generalizability of the models. Inferential analyses demonstrated that gradient boosting achieved the highest predictive performance (AUC = 0.94), followed closely by neural networks (AUC = 0.93) and random forests (AUC = 0.91). All models significantly outperformed logistic regression and support vector classifiers, indicating the superiority of non-linear approaches for capturing subtle relational patterns. Emotional withdrawal was the strongest predictor across all models (p < .001), followed by daily supportive behaviors, communication responsiveness, attachment avoidance, and conflict frequency. Demographic variables showed no significant predictive contribution (all p > .05). Confusion matrix analyses revealed that gradient boosting produced the lowest false-negative rate, confirming its utility for early detection. The study demonstrates that machine learning models, particularly gradient boosting and neural networks, can accurately detect early signs of emotional disengagement in couples by integrating relational, emotional, and behavioral indicators. These findings highlight the potential of computational approaches to support preventive interventions and enhance clinical assessment in couple therapy.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.010 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".