Behavior biometrics is an innovative identity authentication method that uses unique behavioral traits instead of traditional physical attributes. It focuses on user interactions with digital systems, such as keystrokes, mouse movements, and typing patterns, for continuous and non-intrusive authentication. In the digital age, these personalized behaviors-like typing speed, pause intervals, and mouse trajectories-serve as distinctive authentication markers. Keystroke and mouse dynamics provide consistent patterns over time, making them crucial for security. However, challenges like context variability and privacy remain. This paper examines the role of mouse dynamics in behavior biometrics, exploring its applications, challenges, and potential. Our research also evaluates deep neural network models on collected data and introduces a novel recurrent mesh-based model for mouse dynamics-based user authentication.