Origins and
Transitions in the
Evolution of
Learning

An Enduring Challenge

The last decades have witnessed an explosive development in artificial intelligence, built upon over a century of scientific advances and millennia of human development. One can only imagine that the small brachiopod Lingula experienced similarly transformative conditions some 500 million years ago during the Cambrian explosion. Both moments involved innovations in learning, shaped by sudden transitions and slow antecedents, giving rise to new modes of life and restructuring the world.

The shared connection is a puzzle that Bedau and colleagues (2000) include among their open questions in artificial life: to "determine minimal conditions for evolutionary transitions from specific to generic response systems." This workshop invites experimentalists and theorists to revisit this enduring challenge for learning specifically.

Three Guiding Questions

Our inquiry concerns three questions. First, how should learning be identified in the context of this transition? Second, what mechanisms, molecular, neural, behavioural, and otherwise, are necessary and sufficient for its development? Third, what ecological and environmental conditions are necessary and sufficient for its emergence?

We welcome contributions from the full range of modelling traditions in artificial life (whether completed, ongoing, or proposed) as well as more theoretical reflections on prior work and future directions of inquiry.

Towards a Common Understanding

This workshop intends to bring together fragmented paradigms, analytic frameworks, and established results across disciplines, backgrounds, and perspectives. The aim is to familiarise ourselves with the breadth of current knowledge (both empirical and methodological) and work towards a shared understanding of the most promising ways to address this challenge.

When & Where

What to Expect

The workshop will bring together researchers with different backgrounds, expertise, and views of learning. Our goal is to start conversations about how practitioners in and around artificial life understand learning, and how artificial life can contribute to research on its origins and evolutionary transitions.

The session will include invited talks from researchers within and outside artificial life, followed by a panel discussion and time for broader exchange among participants.

• Szathmáry, E., & Maynard Smith, J. (1995). The major evolutionary transitions. Nature, 374, 227–232. https://doi.org/10.1038/374227a0
• Ginsburg, S., & Jablonka, E. (2021). Evolutionary transitions in learning and cognition. Philosophical Transactions of the Royal Society B, 376, 20190766. https://doi.org/10.1098/rstb.2019.0766
• Barron, A. B., Hebets, E. A., Cleland, T. A., Fitzpatrick, C. L., Hauber, M. E., & Stevens, J. R. (2015). Embracing multiple definitions of learning. Trends in Neurosciences, 38(7), 405–407. https://doi.org/10.1016/j.tins.2015.04.008
• McGregor, S., Vasas, V., Husbands, P., & Fernando, C. (2012). Evolution of associative learning in chemical networks. PLoS Computational Biology, 8(11), e1002739. https://doi.org/10.1371/journal.pcbi.1002739
• Ferguson, A. J., & Ofria, C. (2023). Potentiating mutations facilitate the evolution of associative learning in digital organisms. In Proceedings of the Artificial Life Conference 2023 (ALIFE 2023). MIT Press. https://doi.org/10.1162/isal_a_00684
• Pontes, A. C., Mobley, R. B., Ofria, C., Adami, C., & Dyer, F. C. (2020). The evolutionary origin of associative learning. The American Naturalist, 195(1), E1–E19. https://doi.org/10.1086/706252