Home > My Fiction, Science > Memory RNA after all?

Memory RNA after all?

Today I’m experiencing that common occupational hazard for the science fiction writer: Learning that a new scientific discovery has rendered something I wrote obsolete.

I’ll let Tamara Craig, the narrator of my 2010 story “No Dominion” from DayBreak Magazine, explain:

Nearly a century ago, an experiment with flatworms seemed to show that memory was stored in RNA and could be transferred from one organism to another. But the experiment had been an unrepeatable fluke — pardon the pun — and later research showed that memory worked in a completely different way, unfortunately for the science fiction writers who’d embraced memory RNA as a plot device.

(This passage is trimmed down a bit in the version soon to be reprinted in Among the Wild Cybers: Tales Beyond the Superhuman, since that collection’s editor thought the references to SF writers were a bit too meta and distracting.)

What I wrote there was based on memory and was roughly correct. In the late 1950s and early ’60s (“No Dominion” is set in 2059), a researcher named James V. McConnell spent years experimenting with memory in planaria (flatworms), doing things like cutting them up and testing if their regenerated tails retained the memories of their original heads, and — most famously — grinding them up and feeding them to other flatworms. McConnell’s research did seem to show that some learned behavior was passed on by what he proposed to be a form of RNA storing memories created in the flatworm’s brain. It’s true that there was never enough reliable confirmation of his result to establish it as true, and the scientific establishment dismissed McConnell’s findings, although they did inspire a lot of science fiction about RNA memory drips or memory pills as a technique for quick-learning overnight what would normally take months or years. However, it seems that there were some experiments that did appear to replicate the results. There just wasn’t enough consistency to make it definitive.

Apparently, there’s been some renewed experimentation with McConnell’s theory in the past few years, showing promising but uncertain results. What I read about today was a new result, involving snails rather than flatworms:

http://www.sfn.org/Press-Room/News-Release-Archives/2018/Memory-Transferred-Between-Snails

Memories can be transferred between organisms by extracting ribonucleic acid (RNA) from a trained animal and injecting it into an untrained animal, as demonstrated in a study of sea snails published in eNeuro. The research provides new clues in the search for the physical basis of memory.

Long-term memory is thought to be housed within modified connections between brain cells. Recent evidence, however, suggests an alternative explanation: Memory storage may involve changes in gene expression induced by non-coding RNAs.

A more thorough article about the result can be found at the BBC:

‘Memory transplant’ achieved in snails

Now, this doesn’t mean the original memory RNA idea was altogether right. This experiment involved injecting the RNA into the blood of the snails rather than feeding them ground-up snails. And the result probably needs to be repeated more times and studied more fully before it can be definitive. But it does suggest that I was wrong to insist that memory “worked in a completely different way.” It’s possible that memories are stored, not in patterns in the synapses of nerve cells, but in RNA in their nuclei, which has an epigenetic effect on the neurons’ gene expression and therefore their behavior and structure.

Of course, all these results show is that very simple reactions to stimuli can be transferred. There’s no evidence that it would work for something as elaborate as the kind of declarative memory and knowledge that the passage in the story was discussing, or the kind of procedural memory and skills often transferred by memory RNA in fiction (e.g. foreign languages or fighting techniques). Perhaps those kinds of memory are partly synaptic, partly epigenetic. Maybe there’s something else involved. So Tamara’s lines in the story may not be entirely obsolete, just a little inaccurate (forgivable, since she’s a cop, not a scientist).

So I guess it could be worse. It was a minor part of the story anyway. And the actual research itself suggests some interesting possibilities. The articles say that learning more about memory creation and storage — and perhaps memory modification and transfer — could help treat conditions like Alzheimer’s and PTSD. If so, then it’s unfortunate that McConnell’s results weren’t taken more seriously half a century ago.

  1. Mark Thorson
    June 23, 2020 at 10:25 pm

    The story is a bit more complex than that. The Planaria results were never disproven, but they were difficult to reproduce, some people having more success than others. One possible explanation proposed by the connectionists (current dominant memory theory) was that the successful experimenters were cutting up their trained Planaria into larger chunks, and sections of their neural networks were being transplanted into the untrained Planaria. The dismissal of memory RNA theory was largely due to the number of supporters (i.e. cultural) rather than science and the phenomenal success of artificial neural networks based on connectionist models. But having success in an artificial model doesn’t mean it’s true. The connectionist model is based on the McCulloch-Pitts model of the neuron and the Hebb model of learning, both of which are extraordinary simplifications. Natural neurons and neural networks are not completely described by these models by any means. Recent studies of small RNA molecules and peptides (small proteins) present a whole world of possibilities that is as yet poorly explored. There may yet be memory molecules, which could be RNA, peptide, or even an RNA-peptide complex.

    Where connectionist memory theory falls flat on its face is that Paramecia can be trained and exhibit memory. Being unicellular, they can’t possibly implement a connectionist memory model, but they might implement a chemical memory model. Someone should investigate whether memories can be transferred from trained Paramecia to untrained Paramecia.

    With regard to the cultural aspects of science, I am reminded of the polywater affair. The Soviet results originally reported by Fedyakin but then extended by the very respected chemist Deryagin were reproducible, but only in maddingly small quantities, making analysis difficult (to say the least). Numerous hypotheses were advanced to explain the anomalous behavior of the putative polywater. Eventually, science moved on without a definitive explanation, though many people seemed satified with the notion that the polywater samples were in fact saturated solutions of silicic acid dissolved from the quartz capillaries in which the polywater was prepared.

    Today, we know the answer, both through computer simulation studies of liquid water from the late 1970’s/early 1980’s and practical experiments on water in confined spaces (mostly through the work of Robin J. Speedy). The answer? That really was polywater in those tubes. But all liquid water of any reasonable purity is polywater, too. All or nearly all of the water molecules in a glass of water forms a single, interconnected network through hydrogen bonding. Water also has anomalous properties in close proximity to solid surfaces, through hydrogen bonding and other effects such as van der Waals force.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: