Episode Summary
Executive Summary: The episode explores complexity science through the lens of cognition, asking whether there are universal principles governing intelligent systems across biology and AI. Ricard Solé argues that cognition emerges from networked, threshold-based information processing, spanning solid brains, liquid brains (colonies, immune systems), and unusual forms like slime molds, while suggesting that future AI and synthetic biology may uncover or engineer unexplored regions of the “space of cognitions.”
Main Topics: The Space of Cognitions (Priority: 5/5): The conversation centers on mapping the full range of systems that can process information, learn, and exhibit cognition, from neurons to colonies and AI. Complexity, Emergence, and General Laws (Priority: 5/5): Sean Carroll and Ricard Solé discuss whether complexity science can yield universal laws or must remain a collection of context-specific rules; Solé ties complexity to emergence. Evolution of Cognition and Neurons (Priority: 5/5): Solé frames cognition as an evolutionary advantage tied to prediction, sensing, movement, and the rise of neurons and interneurons in multicellular organisms. Solid Brains vs Liquid Brains (Priority: 4/5): The episode contrasts fixed neural architectures with mobile, colony-based or immune-system-like networks, arguing that different substrates can still support cognition. Criticality and Network Structure in Brains (Priority: 4/5): They discuss the idea that healthy brains operate near criticality, balancing order and disorder to support flexible, rapid responses and information processing. AI, Language, and Human Uniqueness (Priority: 4/5): The discussion compares large language models and neural networks with human cognition, emphasizing gaps in theory of mind, mental time travel, embodiment, and genuine understanding. Synthetic Biology and Unexplored Cognitions (Priority: 4/5): Solé describes using synthetic biology to probe whether evolution missed possible cognitive architectures and whether humans can engineer new kinds of cognition.
Key Arguments: Complex systems may share universal principles, but the field is still developing and lacks a fully fleshed out theory of complexity. Cognition is central to biological complexity because predicting the future and responding adaptively confers major evolutionary advantages. Neurons were a major evolutionary transition because they enabled dedicated information propagation rather than only sensing or executing tasks. Interneurons are crucial because they mediate information processing, not just input/output, making genuine cognition possible. Brain dynamics appear to be healthy when near criticality, balancing order and disorder for responsiveness and flexibility. Liquid brains, such as ant colonies, termites, the immune system, and even some slime molds, can process information and store memory despite lacking fixed neural wiring. Artificial neural networks and modern AI may converge on some brain-like design principles, but they still lack key human capacities such as memory-rich development, theory of mind, and mental time travel. Embodiment and interaction with a world may be necessary for stronger forms of artificial cognition. There may be empty regions in the space of cognitions—architectures that evolution did not find but that synthetic biology or AI could potentially engineer. Human social intelligence depends on culture, language, imitation, and theory of mind; isolated human brains are far less capable than socially embedded ones.
Data Points: Cambrian explosion timing: ~550 million years ago - Used as a reference point for the emergence of major animal nervous-system innovations and predation-driven evolution. Multiplicity of termites vs individual: ~1,000 times larger - Termite nests are cited as hyperstructures vastly larger than individual termites, illustrating emergence. Evolutionary comparison: Before the Cambrian - Solé says organisms with neuron nets existed before the Cambrian, such as hydra-like or jellyfish-like systems. Scale-free/scaling law reference: Renz rule - Mentioned in the discussion of microprocessor networks showing statistical properties similar to parts of the brain cortex.
Pivotal Quotes: "What is the space of all possible cognitions?" — Sean Carroll: Introduces the episode’s central framing: cognition as a broad design space spanning biology and AI. "For me, the definition comes from emergence, essentially." — Ricard Solé: Solé defines complexity in terms of emergent collective behavior, using termite nests as his example. "We are mental time travelers." — Ricard Solé: Used to distinguish human cognition from current AI by emphasizing memory, imagination, and future-oriented thought.
Implications: The episode suggests cognition may be a general, engineerable property of complex networks, not just brains. That opens research paths in AI, neuroscience, and synthetic biology while raising questions about which forms of intelligence evolution missed.
About Sean Carroll MindScape
Ever wanted to know how music affects your brain, what quantum mechanics really is, or how black holes work? Do you wonder why you get emotional each time you see a certain movie, or how on earth video games are designed? Then you’ve come to the right place. Each week, Sean Carroll will host conversations with some of the most interesting thinkers in the world. From neuroscientists and engineers to authors and television producers, Sean and his guests talk about the biggest ideas in science, ...