<p>In high-pressure environments such as education, healthcare, and high-stakes workplaces, cognitive load, multitasking, and time constraints create significant challenges. Traditional methods often overlook psychological factors like emotional resilience and reasoning capacity, while failing to capture nonlinear interactions. This study proposes a two-stage hybrid model combining fuzzy logic and machine learning to analyze and predict cognitive load and working memory performance. In the first stage, fuzzy logic is used to categorize cognitive load based on electroencephalography-derived metrics of emotional resilience and thinking capacity through membership functions and rule-based inference. In the second stage, the predicted cognitive load is integrated with the performance of working memory to assess its impact on learning outcomes and mental health. Analysis shows that when both emotional resilience and thinking ability are high, the cognitive load is also high, aligning with theoretical expectations. Machine learning models, including Support Vector Machine, Naive Bayes, Random Forest, and Decision Trees, validate the framework, demonstrating superior classification of cognitive load and enhanced prediction of working memory performance compared to traditional linear approaches. By bridging the gap between cognitive load dynamics and mental health, this framework provides actionable insights to optimize performance and well-being in demanding settings. The findings underscore the importance of integrating fuzzy logic and machine learning to address complex cognitive challenges in high-demand sectors.</p>