[frankinstien]

Inheritance of associative memories and acquired cellular changes in C. elegans Now if we could just learn to do this what a time saver it would be for future generations. 

[ivan.moony]

But what if the entire memory has an upper bound? Wouldn't it then take space for other interesting stuff?

[DaltonG]

Inheritance of associative memories and acquired cellular changes in C. elegans Now if we could just learn to do this what a time saver it would be for future generations. 

The was an excellent interview done with David Glanzman on the BrainInspired AI podcast hosted by Paul Middlebrook on the topic of RNA transmissible memories. As I told him in a response to the podcast, I don't buy what he's selling. The following is my brief comments outlining why:

 There's a big problem with regard to the potential of RNA being encoded as a memory. RNA codons encode for amino acids, the constituent building blocks of proteins. Now proteins come in 2 major classes; structural proteins and enzymes (enzymes do work). Additionally, the strings of amino acids require other enzymatic proteins (members of the proteom) to fold them as they come off the ribosomes to establish a its phenotype. It's inconceivable to think that structural proteins could be the source of memory. Neurotransmitters are fundamentally proteins, but the role of neurotransmitters, be they full fledged enzymes or just subset peptides, is to activate ion exchange between intracellular and extracellular ions. RNA will take part in a communications system, but not be the source of a memory. As for the effect of RNA transplantation, it may be that for a short time (lifespan of the RNA), it served to heighten synaptic sensitivity and there by lower the trigger threshold for the neuron in question through increased production of vesicles or neurotransmitters localized to the subcellular synaptic compartment(s).

All in all, with regard to RNA transfer and its effect, I would say that his interpretation is a bit short sighted. I would agree that the pyramidal neuron is the memory and the synaptic pattern of activated synapses are selectors for a cell in a competitive population.

On of the things with regard to the RNA transfer that was not made clear was where the RNA sample was introduced or injected? The soma, the dendrite, the synapse, or the interstitial space between neurons?