In partial reference to the Multicellular Parts thread (Perhaps a recent comment in particular), but mainly for our Release Roadmap and the related request in the 1.7 planning thread, I want to try to finalize the design for using the Axon in the Multicellular Stage here.
A bit of science
It’s important to note that the early evolution of muscles and neurons is an active field of research with many unanswered questions. Nonetheless, there is a solid consensus on some things, as well as plenty of hypotheses with some backing. To quickly recap what I have learned:
- The earliest fossils of animals we have, have examples that look like they have muscles, and thus likely nerves to control them. One proposed reason is that animal ancestors that did not yet have muscles and nerves, were microscopic.
- These same cases show that at that point the split between Cnidaria (jellyfish, polyps and corals) and bilateria (from worms to us) was already in place, because we find fossils that have those body shapes. That again means the split happened before we have macroscopic fossils.
- Earlier split-offs are Placozoa (weird grazing flat blobs) and Porifera (sponges, sessile filter feeders), both without neurons. And then there are Ctenophora (comb jellies) possibly even before that, (with neurons). That raises the question of whether the common ancestor of all of these had neurons, with Placozoa and Porifera losing them, or whether Cnidaria+Bilateria and Ctenophora evolved them separately. The important part for us is that either way, the evolution is of muscles and neurons is still before we had fossils of any of these.
- Cnidaria and Bilateria both have very similar muscles and nerve cells within them, and still similar muscles between them. This suggests also that their common ancestor already had these, or close enough.
- Genetic evidence suggests the same as above.
Conclusion: It is considered at least a strong possibility that microscopic animals already evolved muscles and neurons. So it is entirely fair to include them in Thrive’s Multicellular Stage. Similarly, all macroscopic motile animals have neurons, so it would make sense that they’re a semi-requirement there, and you would otherwise be sessile.
The actual benefit neurons provide is coordination. For example, each flagellum cell deciding where to be active based on local conditions/signals ends up in distant ones pushing in different directions. Having a neuron direct them means they all work together.
Across Ctenophora, Cnidaria and simple Bilateria, neurons connect with:
- Sensory cells of all types
- Muscle cells
- Flagella/Cilia cells (both for moving the self, but also for moving other things)
- Gland cells (basically anything that releases compounds)
- Nematocytes (the harpoon things)
Conclusion: There is support for nerve cells on the basic level to affect any movement, but also to have them affect any type of “active ability”.
IRL nerve cells are extremely widespread in motile life, especially anything that hunts around for food. At the same time, they are fairly absent
https://onlinelibrary.wiley.com/doi/10.1111/ede.12472
EVOLUTIONARY ORIGINS OF NEURONS AND NERVOUS SYSTEMS - In the Light of Evolution - NCBI Bookshelf
https://www.cell.com/current-biology/fulltext/S0960-9822(16)30489-4
Evolution of nervous systems - Wikipedia
Objectives
The Axon represents a real neuron/nerve cell, which should translate to having a nervous system from the start in Macroscopic. From existing designs, we want the nervous system to be an expensive invention that is only worth it for motile life forms. Those constraints should therefore be carried over to Axon design for the Multicellular Stage as well.
The Multicellular parts thread does have a concept for a more general “signalling complex”, but this should not be implemented as using the Axon name/model, as it provides more general benefits to any kind of large multicellular creature, rather than just the active motile ones.
Design
In the simplest implementation, the Axon provides a % boost to movement force of the colony. That’s it.
This way it boosts all the other ways your cells provide speed. It would also have a significant ATP cost (either when moving or just always-on). With a fixed cost and a % boost (and balancing the numbers correctly) it will automatically:
- Not be worth it if you have no speed to begin with.
- Not so worth it if you have just a few cells providing speed.
- Gradually become more worth it as you have more cells providing speed.
- Be worth it for very motile species, but not for those that don’t move to begin with.
There are some possible, but very optional complications to the system:
- Unlock condition: 5 cells with Actomyosin.
- Or 5 cells that have flagella, or cilia, etc.
- Have it only affect movement force from organelles, not the “base speed” each cell gets. This further makes it not worth it for species without movement organelles.
- Affect rotation speed from organelles in the same way as movement speed.
- Change the colony movement system to normally have diminishing returns on movement/rotation speed from organelles in many cells. The neuron would then counteract that instead of giving a direct multiplier.
- Only look at whatever is the strongest axon cell, don’t add them up. (to avoid having a lot of neurons in such a small design. But I think this can be achieved through numbers as well)
- Boost other active abilities like toxin shooters, or pulling cilia.
A small visual issue
It’s not a deal breaker by any means, but with the current organelle model and placement method, there are some visual issues I think:
It poking into other cells does not look that great. The membrane stretching making more cohesive makes it a bit worse, and the improved version will likely make it constant. This happens with for example pili too, but at least those are not supposed to be placed pointing towards your other cells.
Perhaps it’s possible to have the model sticking “upwards” so that it’s not sticking into other cells? Of course The Dream™ would be for axons from the cell body to reach and stretch towards cells all over the place, but that’s obviously not happening. So just a model or placement/orientation tweak would be nice.
I would very much like to hear opinions on the above.
If what we want instead is a more generalised signalling system, then the above should be shelved, with the Axon for now reserved to Macroscopic. The “signalling complex” organelle would need a different model and name.