Sude Ozkaya

I am an undergraduate student at Carnegie Mellon University pursuing a double major in Cognitive Science and Human-Computer Interaction, with a minor in Creative Writing. My research interests lie at the intersection of human-centered AI and cognitive science, with a focus on designing AI systems informed by cognitive theory that can reason, communicate, and adapt in ways that better reflect human cognition, emotion, and perception. My previous work includes modeling social worlds from life stories at the CMU Robotics Institute, building an emotion-aware dialogue system for the Furhat robot at Koç University, and extending AI audio generation and automating a preprocessing pipeline at Stanford University.

At the Visual Inference Lab, I am working with Dr. Akihito Maruya on computational modeling of motion textures, studying dynamic, naturalistic non-rigid motions such as flickering fire, lapping ocean waves, drifting smoke, rustling leaves, and swaying people. Our work asks how the visual system summarizes these complex motions into recognizable percepts, and which motion statistics carry perceptual weight.

Outside of research, I enjoy reading, making ceramics, playing the piano, and playing basketball.