In this paper, we study human locomotor patterns using the Virtual Carpet Paradigm™, a neuropsychological assessment in virtual reality. This setup captures participants’ movements as raw spatial data, which we analyze using kinematic methods. By applying the Ant Colony Optimization algorithm, we identify 3 distinct clusters of navigation behaviors. The algorithm distinguishes between individuals with cognitive spatial impairments and healthy participants. These findings highlight key behavioral differences and support earlier research. Our study demonstrates the potential of this method for understanding human movement patterns, offering valuable applications for research and clinical practice.