How morphological development can guide evolution
Not on the map. Why?
Abstract (arXiv)
Organisms result from adaptive processes interacting across different time scales. One such interaction is that between development and evolution. Models have shown that development sweeps over several traits in a single agent, sometimes exposing promising static traits. Subsequent evolution can then canalize these rare traits. Thus, development can, under the right conditions, increase evolvability. Here, we report on a previously unknown phenomenon when embodied agents are allowed to develop and evolve: Evolution discovers body plans robust to control changes, these body plans become genetically assimilated, yet controllers for these agents are not assimilated. This allows evolution to continue climbing fitness gradients by tinkering with the developmental programs for controllers within these permissive body plans. This exposes a previously unknown detail about the Baldwin effect: instead of all useful traits becoming genetically assimilated, only traits that render the agent robust to changes in other traits become assimilated. We refer to this as differential canalization. This finding also has implications for the evolutionary design of artificial and embodied agents such as robots: robots robust to internal changes in their controllers may also be robust to external changes in their environment, such as transferal from simulation to reality or deployment in novel environments.
Key facts. Cited by 11 papers in the map (data through 13 Oct 2026). First posted on arXiv 20 Nov 2017.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 2; 2022: 5; 2023: 1; 2024: 2; 2025: 1.
11 papers.
- 2021Embodied Intelligence via Learning and Evolutioncited by 155
- 2021Learning Modular Robot Control Policiescited by 18
- 2022REvolveR: Continuous Evolutionary Models for Robot-to-robot Policy Transfercited by 12
- 2022A Unified Substrate for Body-Brain Co-evolutioncited by 4
- 2024Evolution and learning in differentiable robotscited by 4
- 2025Embodied Co-Design for Rapidly Evolving Agents: Taxonomy, Frontiers, and Challengescited by 4
- 2022Collective control of modular soft robots via embodied Spiking Neural Cellular Automatacited by 3
- 2024Towards Multi-Morphology Controllers with Diversity and Knowledge Distillationcited by 2
- 2022Morphological Wobbling Can Help Robots Learncited by 1
- 2023Lamarck's Revenge: Inheritance of Learned Traits Can Make Robot Evolution Bettercited by 1
- 2022Shape Change and Control of Pressure-based Soft Agentscited by 0
- 2025Embodied Co-Design for Rapidly Evolving Agents: Taxonomy, Frontiers, and Challenges4 Dec 2025
- 2024Evolution and learning in differentiable robots23 May 2024
- 2024Towards Multi-Morphology Controllers with Diversity and Knowledge Distillation22 Apr 2024
- 2023Lamarck's Revenge: Inheritance of Learned Traits Can Make Robot Evolution Better22 Sep 2023
- 2022Morphological Wobbling Can Help Robots Learn5 May 2022
- 2022Shape Change and Control of Pressure-based Soft Agents1 May 2022
- 2022Collective control of modular soft robots via embodied Spiking Neural Cellular Automata5 Apr 2022
- 2022A Unified Substrate for Body-Brain Co-evolution22 Mar 2022
- 2022REvolveR: Continuous Evolutionary Models for Robot-to-robot Policy Transfer10 Feb 2022
- 2021Learning Modular Robot Control Policies20 May 2021
- 2021Embodied Intelligence via Learning and Evolution3 Feb 2021