[Seminar]"Authentic Purpose: developmental neurology for autonomous neurorobots" Dr. Jeffrey White

Authentic Purpose: developmental neurology for autonomous neurorobots

Jeffrey White, Philosopher of AI and Cognitive Science

Visiting researcher, Cognitive Neurorobotics Research Unit

This talk considers how authentic purpose develops during adolescence, orients intelligent behavior across the lifespan, and how associated insights from neuroscience may inform ongoing robotics experiments. The basic idea is that adolescent synaptic pruning irreversibly compresses a rapidly developing high-dimensional neural possibility space into a low-dimensional global attractor. Mechanistically, the basic idea is that insular gating dynamics feed mirrored affective and actional information upward, and that Von Economo neurons permanently fix communication between early-trained and late-developing medial frontal processes as these grow, as modulated through self-justificatory inner dialogue with an imagined universal audience as socially-normative selection pressure, and with increasing specificity with lateral maturation through adult development. Unlike enactivist accounts indexing intelligence to adaptive “grip” on dynamic embedding environments, here adaptation pursues permanently fixed ideal conditions through resolute pursuit analogous to religious commitment. Convergences are drawn with recent work from the Tani group (Matsumoto et al.,and Tinker et al.) which describe intrinsic motivation in robots, and the talk concludes by considering how the present thesis might inform ongoing and future experiments.

----------------------------

Jeffrey White is a philosopher specializing in the philosophy of artificial intelligence and cognitive science. He is a Visiting Researcher in the Cognitive Neurorobotics Research Unit at the Okinawa Institute of Science and Technology Graduate University (OIST), where he collaborates with Prof. Jun Tani.

White earned his PhD in Philosophy from the University of Missouri in 2006. His doctoral work developed an information-processing model of conscience, integrating insights from computational models (Ron Sun was on his doctoral committee) with traditional philosophy (including Heidegger) and cognitive science (working from fundamental imaging studies focusing on insular gating dynamics and what came to be known as the salience network). His research focuses on machine ethics, moral cognition, autonomy, and the societal impacts of AI and associated technologies, including robotics. He also explores topics such as value alignment, trust, safety, and resilience in socio-technical systems, and moral education and purpose development in human beings.

White has taught philosophy and ethics at universities in the United States, South Korea, and the Netherlands. He served as founding editor for Ethics of AI, Moral status of artificial systems, and Machine ethics for PhilPapers (2015-2024), co-chaired the Philosophy of AI track at the 2018 World Congress of Philosophy, and has spoken on different topics with different audiences around the world, including for Embodied Intelligence in 2023. He reviews for numerous journals and book publishers. He lives in South Korea with his wife and two dogs, and lately spends a lot of his time editing the Open Forum for Springer-Nature journal AI & Society (Q1 in HCI, AI, and Philosophy).

White brings a cross-disciplinary perspective to questions at the intersection of computation, cognition, and human meaning, engaging with researchers across philosophy, computer science, and the cognitive sciences. He joins us today at the Okinawa Institute of Science and Technology Graduate School to talk about the timely topic of purpose development in minds and machines.