Fusions of Consciousness

"We propose that consciousness is fundamental, and can be modeled as a network of interacting “conscious agents”. In this section, we briefly motivate and present a definition of conscious agents."

4. Conscious Agents, p. 9

This defines the modeling substrate—interacting agents—as the representational units, grounding later claims about structured encodings (qualia kernels and polytopes) that can inform how representational structure might be formalized for both brains and AI systems .

"The qualia kernel, Q, thus expresses the relation that conscious experience has to itself."

4. Conscious Agents, p. 11

By introducing Q as a formal object that maps experiences to experiences, the paper specifies an internal representational structure that could be compared to embedding spaces or latent codes in AI and to population codes in neuroscience .

"When agents interact, what happens? We demonstrate, in Section 5, that they can combine into a complex agent, or fuse into a simpler agent with a novel conscious experience. If n agents interact, their possible combinations form an n(n− 1)-dimensional polytope with n^n vertices—the Markov polytope Mn. Their possible fusions form an (n − 1)-dimensional simplex—the fusion simplex Fn."

1. Introduction, p. 2

This formalizes an integration mechanism: distributed agents combine into higher-order unified structures (polytopes/simplices), paralleling information integration ideas relevant to global access in brains and convergent attention in AI .