Enhanced Object-Oriented Simulation and Visualization of Filter-Based Attention Models: A Unified Framework

This paper presents an enhanced object-oriented framework for modeling, simulating, and visualizing filter-based attention systems, building upon foundational models by Broadbent and Treisman. Using Unified Modeling Language (UML) and Python-based simulations, we propose a unified architecture that integrates early and attenuated filtering mechanisms within a cognitive attention system. The framework is validated through simulations of dichotic listening tasks, demonstrating the dynamic switching and attenuation of attention in response to routine and suspicious sensory events. The results highlight the potential of object-oriented modeling in cognitive systems and its applicability in artificial intelligence and human-computer interaction.