Josh Bongard
Josh Bongard worked on designing living systems with an evolutionary algorithm, in a 2020 pipeline with Kriegman, Blackiston and Levin1. With Levin, Bongard also proposed the term polycomputing for materials that give several computation results in one place and time4. A 2025 paper with Bongard among its authors applied functional connectivity networks to aneural tissue6.
- Field
- researcher
- In the library
- 3 works, named in 6 passages elsewhere
Designing living machines
The 2020 pipeline took a description of biological building blocks and a desired behaviour, and evolved designs in silico that could then be manufactured from cells1. Designs were selected on net displacement over a 10-second period, with randomized contraction cycling at 2 Hz2. The built organisms showed the evolved structure and behaviour, even though the timing of cardiomyocyte contraction was modelled as random noise. Selection for locomotion produced shapes that reduced the randomness of the movement3.
Polycomputing
Bongard and Levin defined polycomputing provisionally as a material giving the results of more than one computation in the same place at the same time4. They required that the computation be evolved or designed and be readable by other parts or devices. They argued that living systems polycompute because each level and component interprets its neighbours in its own way. Life was described as nested, competing and cooperating agents, each predicting and exploiting its microenvironment5.
Use by later authors
Levin cited Bongard and Levin for the machine analogy in biology. In that account the genome sets the hardware, while electrophysiological software shapes the outcome7. The 2025 study with Bongard as co-author used functional connectivity networks. These networks are built from statistical dependencies between activity in different areas of a tissue.
Here, we demonstrate a scalable approach for designing living systems in silico using an evolutionary algorithm, and we show how the evolved designs can be rapidly manufactured using a cell-based construction toolkit.
Kriegman S, Blackiston D, Levin M, Bongard J, 2020 · A scalable pipeline for designing reconfigurable organisms · open at passage 4designs were selected based on net displacement achieved during a 10-s period (with randomized, phase-modulated contraction, cycling at 2 Hz).
Kriegman S, Blackiston D, Levin M, Bongard J, 2020 · A scalable pipeline for designing reconfigurable organisms · open at passage 12As a side effect of selection pressure for locomotion, derandomizing morphologies evolved: evolutionary improvement occurred through changes in overall shape
Kriegman S, Blackiston D, Levin M, Bongard J, 2020 · A scalable pipeline for designing reconfigurable organisms · open at passage 18We provisionally define polycomputing as the ability of a material to provide the results of more than one computation in the same place at the same time.
Bongard J, Levin M, 2023 · There's Plenty of Room Right Here: Biological Systems as Evolved, Overloaded… · open at passage 4Life polycomputes because it is a set of overlapping, competing, cooperating nested dolls, each of which is doing the best it can to predict and exploit its microenvironment
Bongard J, Levin M, 2023 · There's Plenty of Room Right Here: Biological Systems as Evolved, Overloaded… · open at passage 12FC networks are weighted, undirected networks generated based on instantaneous statistical correlations, or statistical dependencies, between activity in different areas of a system.
Blackiston D, Dromiack H, Grasso C, Varley TF, Moore DG…, 2025 · Revealing non-trivial information structures in aneural biological tissues via… · open at passage 4While the genome encodes the hardware (the cellular affordances, such as ion channels), the actual outcome is the result of context- and experience-sensitive electrophysiological software
Levin M, 2023 · Bioelectric networks: the cognitive glue enabling evolutionary scaling from… · open at passage 37
2020
2020 · read
2023
2023 · read
2025
2025 · read