More biological learning systems?

November 28, 2017 6:04 AM

I've been building a list of ways that biological systems accumulate knowledge. Which ones am I missing?

What I've got so far:

DNA changes followed by selection:
- random mutation
- sexual reproduction
- plasmid exchange/horizontal gene transfer
- V(D)J recombination + clonal selection
- CRISPR systems
- random uptake of DNA from the environment

Neural and sensory changes:
- sensitization and habituation
- synaptic plasticity and pruning
- neural apoptosis (probably)

What else? Are there hormonal systems that accumulate knowledge? What about plants?

Thanks!

posted by clawsoon to Science & Nature (18 answers total)

Best answer: Acquired immunity, e.g. targeted antigens, most notably in the case of vaccines.
posted by supercres at 6:10 AM on November 28, 2017 [2 favorites]

Wait, sorry, shows I I should have kept up with my cell biology better; that’s V(D)J recombination
posted by supercres at 6:12 AM on November 28, 2017

Response by poster: Oh, I forgot about positive feedback in gene regulatory networks.

If you're wondering what definition of "learning" I'm using, it's obviously a very broad one: Pretty much any change in state or organization which was caused by a past event that allows the organism to respond to a similar future event in a better/faster/more effective way. Short term and long term learning are both of interest.
posted by clawsoon at 6:14 AM on November 28, 2017

Best answer: Post translational protein modification (methylation and phosphorylation)?

For example, in E. coli chemotaxis the methylation state of receptor proteins could be considered a kind of memory of what concentrations of chemoattractants/repellents have been recently encountered.

This also relates to Long Term Potentiation, since one of the mechanisms by which this occurs is phosphorylation of AMPA receptors.
posted by James Scott-Brown at 6:36 AM on November 28, 2017

Best answer: Also RNA interference as an immune response, particularly in plants, where it can spread though a plant and be transferred by grafting.
posted by James Scott-Brown at 6:48 AM on November 28, 2017

maternal transfer of mitochondrial DNA?
posted by peacheater at 6:56 AM on November 28, 2017

Best answer: Mating-type switching by recombination in Saccharomyces cerevisiae might also count as an memory system (where the system 'remembers' its current mating type).
posted by James Scott-Brown at 6:57 AM on November 28, 2017

You might find some things of interest in the field of biomimicry
posted by yohko at 7:01 AM on November 28, 2017

Best answer: Broad categories: You should think about phsyiological changes things like building (or losing) muscle mass due to exercise, calluses, or plants growing towards the sun. And biochemically don't limit yourself to DNA; the feedback systems that ramp up protein levels that are on activated pathways contain information in those levels as well.

A specific thing to look at is melanin and the cycle that gives us the circadian rhythm.
posted by mark k at 7:08 AM on November 28, 2017

Best answer: Here, I'll make up for the first one: motor learning requires active central myelination, i.e., changes to white matter tracts in the brain.

Neuro is more my thing anyway
posted by supercres at 7:10 AM on November 28, 2017

Response by poster: These are all exactly what I'm looking for. :-)
posted by clawsoon at 7:29 AM on November 28, 2017

Best answer: For plants: pretty much any plant that spreads through rhizomes will "forage" for resources. In this manner, a patch of strawberries will "learn" where the good nutrients are, where the light is, where the water is. See p. 16 in this book.

Lots of plants have "memories" related to herbivory, regulated via hormones. Here's some recent work on the mechanisms of plant memory mediated through jasmonic acid.

There's a whole journal of plant signalling and behavior, see here for a nice overview of plant "neurobiology".

At broader scale plants have what is termed "ecological memory", including the informational legacies left by past processes. See recent research here.

NB, some of those links are paywalled. If anyone wants a copy and cannot get access, drop me a line and I will (eventually) send you a copy.
posted by SaltySalticid at 7:36 AM on November 28, 2017 [1 favorite]

Best answer: Epigenetic changes for sure.
posted by tchemgrrl at 7:39 AM on November 28, 2017 [2 favorites]

Best answer: yes, epigenetic can include methylation of DNA, which is transmitted. some research on how this can transmit the effects of, e.g., stress to offspring.
posted by acm at 3:06 PM on November 28, 2017

Response by poster: I guess we have no idea what biochemical mechanism underlies learning and teaching in slime molds.
posted by clawsoon at 4:30 PM on November 28, 2017 [1 favorite]

Best answer: opioid (for example) receptor down-regulation leading to drug tolerance?
posted by gaspode at 8:29 PM on November 28, 2017

Best answer: Symbiosis. I'm sure it happens with our gut bacteria but an easier to summarize one is zooanthellae in corals--if the environment changes the corals can host ones that do better in the new environment.
posted by mark k at 9:15 AM on December 2, 2017

Best answer: Genetic changes (followed by selection) through acquisition of viral DNA, sometimes in quite exotic ways, through all sorts of inner-genomic dynamic (Jim Shapiro: "natural genetic engineering"), possibly through reverse transcription of RNA into germ cell DNA (what differs from other epigenetic processes). Though I myself would hesitate to call this (like most of the processes above) "knowledge acquisition", I think it fits into the list.

Interesting enough, the most prominent knowledge-acquisition process, neural memory in the brains, seems to have originated itself by ... acquisition of viral genes (plus selection -> new knowledge in biological system).
posted by megob at 12:14 PM on May 16, 2018 [1 favorite]