KMAN Publication Inc. (KMANPUB) Journal of Assessment and Research in Applied Counseling (JARAC) 8 Serial Number 30 2026 07 01 Explainable Boosting Machine (EBM) for Decision-Making Competence: Cognitive Reflection, Risk Perception, Emotional Bias, and Uncertainty Tolerance Explainable Boosting Machine (EBM) for Decision-Making Competence: Cognitive Reflection, Risk Perception, Emotional Bias, and Uncertainty Tolerance 1 11 EN Journal Article 2025 12 21 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> The present study aimed to model and explain decision-making competence using an Explainable Boosting Machine (EBM) by examining the individual and interactive contributions of cognitive reflection, risk perception, emotional bias, and uncertainty tolerance.</p> <p><strong>Methods and Materials: </strong>This study employed a cross-sectional predictive design with a sample of 428 adults recruited from Canada through online research platforms. Participants completed a battery of validated instruments, including the Adult Decision-Making Competence scale (A-DMC), Cognitive Reflection Test (CRT), Domain-Specific Risk-Taking (DOSPERT) scale (risk perception subscale), Emotional Decision-Making Questionnaire (EDMQ), and the Intolerance of Uncertainty Scale (IUS-12). Data were collected via a secure online system and screened for missing values and outliers. The primary analytical approach involved the use of an Explainable Boosting Machine (EBM), implemented in Python, to model both linear and non-linear relationships among variables while preserving interpretability. Model performance was evaluated using R², RMSE, and MAE indices, and k-fold cross-validation (k = 10) was applied to ensure robustness. Feature importance metrics, interaction effects, and partial dependence functions were extracted to interpret the contribution of each predictor.</p> <p><strong>Findings: </strong>The EBM model demonstrated strong predictive performance, explaining 48% of the variance in decision-making competence (R² = 0.48). Cognitive reflection emerged as the strongest positive predictor, followed by uncertainty tolerance and risk perception, while emotional bias showed a significant negative effect. All predictors contributed significantly to the model, with non-linear patterns indicating diminishing returns at higher levels of cognitive reflection. Interaction analyses revealed that cognitive reflection significantly attenuated the negative impact of emotional bias, while uncertainty tolerance enhanced the positive effects of accurate risk perception. These results indicate that both cognitive and affective variables, as well as their interactions, play a critical role in shaping decision-making competence.</p> <p><strong>Conclusion: </strong>The findings highlight the multifactorial and interactive nature of decision-making competence, demonstrating that higher cognitive reflection and uncertainty tolerance enhance decision quality, whereas emotional bias impairs it, with explainable machine learning offering a powerful and interpretable framework for capturing these complex relationships.</p> </td> </tr> </tbody> </table> https://www.journals.kmanpub.com/index.php/jarac/article/download/5304/9662