alt Hacker Newsskos:Concept RDFS Class: https://www.w3.org/TR/skos-reference/#concepts
schema:Thing: https://schema.org/Thing
atomspace:ConceptNode: https://wiki.opencog.org/w/Atom_types .. https://github.com/opencog/atomspace#examples-documentation-...
SKOS Simple Knowledge Organization System > Concepts, ConceptScheme: https://en.wikipedia.org/wiki/Simple_Knowledge_Organization_...
But temporal instability observed in repeat functional imaging studies indicates that functional localization constant: the regions of the brain that activate for a given cue vary over time.
From https://news.ycombinator.com/item?id=42091934 :
> "Representational drift: Emerging theories for continual learning and experimental future directions" (2022) https://www.sciencedirect.com/science/article/pii/S095943882... :
>> Future work should characterize drift across brain regions, cell types, and learning.
The important part about the statements in the drift paper are the qualifiers:
> Cells whose activity was previously correlated with environmental and behavioral variables are most frequently no longer active in response to the same variables weeks later. At the same time, a mostly new pool of neurons develops activity patterns correlated with these variables.
“Most frequently” and “mostly new” —- this means that some neurons still fire across the weeks-long periods for the same activities, leaving plenty of potential space for concept cells.
This doesn’t necessarily mean concept cells exist, but it does allow for the possibility of their existence.
I also didn’t check which regions of the brain were evaluated in each concept, as it is likely they have some different characteristics at the neuron level.