Collective intelligence of cells

Strasburger's swarm-spores, as Verworn described them in 1899, were flagellated algal cells placed in a drop near a window. In weak diffused light they hurried to the lit edge of the drop and collected there in great crowds.¹ Bose, in 1906, compared the reversals of Desmodium leaflets to Strasburger's Ulothrix spores in continuous light. The spores first retreated from the light, then stayed still, then returned, moving to and fro like a pendulum.² Washburn, in 1908, reported a dispute about bee colonies. Forel was sure insects cannot hear, while von Buttel-Reepen, who knew bees thoroughly, thought hearing plays a considerable part in their life.³

Lyon, in 2015, described the soil bacterium M. xanthus as renowned for collective behaviors, including structured swarming motility and pack-like predation.⁴ Levin, in 2022, argued that remodeling, regulative development and regeneration illustrate the goal-directed nature of cellular collectives. In this account the cells pursue anatomical states much larger than any one cell and solve problems in morphospace in a context-sensitive way.⁵ Fields and Levin, also in 2022, held that all intelligent agents are collective intelligences. Their problem-solving rests on the competencies of their parts and the architecture of the relations between those parts.⁶

Fields and Levin gave a newt example. When the cells are made artificially larger, fewer of them cooperate to build a kidney tubule. When the cells are made very large, a single cell wraps around itself, leaves a lumen, and yields a tubule of the same diameter.⁷ They read this as different molecular mechanisms being called up in service of one large-scale state, namely cell-to-cell communication in one case and cytoskeletal bending in the other. They called the flexible use of diverse parts toward an invariant outcome a hallmark of multi-scale control in life forms.⁷

Shreesha and Levin, in 2023, named the navigation policies of the developmental layer the competency of the cellular collective. They described it as the capacity to sense the environment and each other and to communicate in order to solve problems in morphospace.⁸ They claimed it is essential to morphogenesis, regeneration and cancer suppression, and central to the genome-form-function relationship.⁸ McMillen and Levin, in 2024, described cancer as a kind of dissociative identity disorder of the somatic collective intelligence.⁹ The held passages present these readings as proposals and directions for further work.