Polymathic Pizza: Trojan Family Weekend - Figures of Speech

Khalil Iskarous is a Professor of Linguistics at USC. He’s interested in animal motor control, from the undulation of worms to the smooth maneuverings of octopus arms, and the agile motions of the human tongue, which are responsible for many hundreds of sound distinctions in the world’s languages. His research interests include Laboratory Phonology, Motor Control, Hydrostatic Skeletons, Experimental Methods in Linguistics, Endangered Languages.

Iskarous is intrigued by the computational nature of two animal skills: cognition and motor control. Iskarous uses two tools to try and understand the computational structure underlying the two skills: dynamical systems analysis and several interrelated methods from modern AI: Neural Turing Machines, Modern Hopfield Neural Networks, and Attention-based Neural Networks. 

Iskarous has worked on different aspects of phonological cognition and the integrated phonetic skills of speech production and perception, and the application of this knowledge to language change, various forms of atypical speech, and understanding the inner workings of modern AI systems. In terms of motor control, Iskarous has worked on human tongue dynamics using MRI with Dani Byrd (USC Dornsife), Krishna Nayak (USC Viterbi), and Shri Narayanan (USC Viterbi and Dornsife), octopus arms with Jennifer Mather (University of Lethbridge), and C. elegans, a beautiful worm, with Andrew Gracey (USC Dornsife). Iskarous pursues the hypothesis that the computational similarities of cognition and motor control can allow us to understand each through its similarities and differences with the other. 

 Three NSF grants have supported this research program: INSPIRE: Dynamical Principles of Animal Movement (NSF 1246750, Iskarous (PI) and Gracey (co-PI), 2012-2017), MRI: Development of a High Performance Low-Field MRI for Dynamic Imaging (NSF-2240349, Nayak (PI), Byrd, Haldar, Iskarous, Naranan (co-PIs), 2018-2024), CompCog: Deep Causal Inference grounds the Perception of Cognitive Objects in Speech (NSF-2240349, Iskarous (PI), Byrd and Narayanan (co-PIs), 2023-2026).