
Artificial neural networks (ANNs) are increasingly deployed in psychological assessment and diagnosis, yet systematic appraisals of their methodological rigor remain scarce. This systematic review synthesizes 37 studies published between 2018 and 2024 that employ ANNs to predict or evaluate psychological constructs. Applications span diagnostic enhancement, test automation, and outcome forecasting, with feed-forward architectures predominating. Performance metrics such as accuracy, F1-score, and ROC AUC are commonly reported. However, substantial heterogeneity in study design, network architecture, and reporting standards complicates cross-study comparison. Findings underscore the need for standardized protocols to ensure replicability and clinical interpretability in this sensitive domain.
This cover image was created using Al-generated content from "Chatgpt". The authors confirm that no human likenesses, copyrighted elements, or misleading representations are included in the image.
View this paper