Panpsychism: A Return to Ancient Intuitions or a Dead End? Toward a Graduated Information-Integration Theory of Consciousness

Panpsychism, the view that consciousness is a fundamental feature of all physical systems, has gained recent philosophical attention but faces two major hurdles: the combination problem (how micro-consciousnesses combine into unified experiences) and a lack of empirical testability. This thesis argues that traditional panpsychism cannot overcome these obstacles in its current form. It proposes a new framework, Graduated Panpsychism with Information-Integration Spectrum (GP-IIS), which merges Integrated Information Theory, evolutionary biology, and computational research.

Study at a glance

Abstract

This thesis provides a comprehensive evaluation of panpsychism as a viable theory of consciousness, examining its explanatory power across multiple domains and its compatibility with contemporary empirical constraints from neuroscience, physics, and artificial intelligence research. While panpsychism has experienced a renaissance in recent decades through the work of philosophers like David Chalmers, Philip Goff, and Galen Strawson, fundamental challenges remain, particularly the combination problem and the lack of empirical testability. This work argues that traditional panpsychism, while philosophically compelling, faces insurmountable difficulties in its current form. However, rather than dismissing panpsychism entirely, this thesis proposes a novel theoretical framework: Graduated Panpsychism with Information-Integration Spectrum (GP-IIS). This framework synthesizes insights from Integrated Information Theory (IIT), evolutionary biology, and computational consciousness research to develop a quantitative, empirically testable version of panpsychism that addresses traditional objections while maintaining core panpsychist intuitions about the fundamental nature of consciousness. The thesis demonstrates that consciousness exists on a continuous spectrum of information integration complexity, where all physical systems possess minimal proto-conscious properties that can be quantified and measured, but only systems above critical integration thresholds exhibit recognizable conscious experience. This approach resolves the combination problem through information integration cascades, provides empirical predictions testable through neuroscience methods, and offers a unified framework for understanding biological and artificial consciousness. The implications extend beyond philosophy of mind to practical questions about AI consciousness, animal welfare, and the future of human-machine interaction.