Do you want to avoid the hassle of traveling with your printed poster? IAPHS2026 is pleased to make poster printing available to you through our supplier PosterSessionOnline. Your poster will be professionally reviewed, printed and shipped directly to Portland and you will be able to pick it up from the Poster desk. Click here to learn more.
Primary Submission Category: Mental health/function
Examining Factors Influencing Positive Mental Health: A Neural Network Approach
Authors: Memuna Aslam,
Presenting Author: Memuna Aslam*
Mental health challenges among college students have risen significantly, highlighting the need to better understand the key determinants influencing psychological well-being. Using data from the Healthy Minds Study (HMS) 2022–2023, covering students across 530 colleges in the United States, this study examines the relative importance of factors such as overall health, sleep patterns, current financial situation, substance use, and obesity in predicting positive mental health outcomes.
Traditional statistical approaches often overlook complex, non-linear relationships and may oversimplify how multiple risk factors interact. To address these limitations, this study applies an Artificial Neural Network (ANN), a computational model inspired by human brain functioning, to improve prediction accuracy and capture multidimensional interactions among variables. After standard preprocessing procedures including data cleaning, scaling, encoding, and dataset splitting, a neural network with two hidden layers was trained using robust optimization techniques.
Results indicate that overall good physical health has the strongest positive importance score in predicting positive mental health, while obesity shows the highest negative importance score. Substance use is also associated with a substantial negative influence on positive mental health outcomes. In contrast, sleep on weeknights demonstrates a relatively negligible importance score in the model. Additionally, current financial situation shows a meaningful positive importance score, suggesting that better financial conditions are associated with improved mental well-being among students.
These findings demonstrate that ANNs can effectively capture complex and non-linear relationships that may be missed by traditional regression models. However, limitations include reduced interpretability due to the “black box” nature of neural networks and potential risks of overgeneralization. Future research should incorporate additional psychosocial and environmental variables and explore more advanced neural network architectures to enhance predictive performance and policy relevance.