[FRIAM] chicken-egg::gumflap-talk

[uǝlƃ ☣]

Excellent! Both these 2 comments are useful. Nick raises composition, high to low dimensional projection, and time slicing. And Steve raises many-to-many maps and mirror neurons [†]. All of these seem relevant to understanding the extent to which we might (not) mimic others and need/want higher bandwidth and/or [a]synchronous media.

As I understand it, the Maegherman paper makes an explicit distinction between simple and complex actions so that even though their study showed no motor activity in relation to the imagery, perhaps more complex actions *would*. So, we can fold in both Nick's and Steve's comments by considering something like 1) the deadlift exercise versus 2) drawing a straight line with a pencil. The deadlift is a dangerous exercise because it's difficult to mimic what you see, whereas it's comparatively easy to *see* all the factors involved with drawing a straight line on a piece of paper (a circle, maybe not so much). Before I started lifting weights as an adult [‡], when I saw someone doing a deadlift, my self-imagery involved hands, knees, feet, and back (particularly the lower back). After lots of advice, analysis, and performance of the exercise, I now image my lats, glutes, and transferse abdominis. That change is radical compared to drawing a line with a pencil. Learning to draft, there was a shift between before and after (e.g. spinning the pencil to get a uniform fatness to the line). But the shift wasn't radical. It's relatively easy to learn to draft by watching a draftsman. It's not very easy to learn the deadlift by watching the deadlift.

I think we could make the same argument using phonemes common to various languages. E.g. is it easier to lip-read German than Spanish? Would a native English speaker be more or less likely to simulate the articulation of a German speaker since English inherits so much from German?

And to touch closer to home, just *how* irritating is it (i.e. how much heat/friction is added to a conversation) when I say Freeam instead of Fryam when rendering FriAM? ... or when Renee' says "pilla", referring to the thing we lay our head on at night? If the signal is difficult for us to simulate, how well can we possibly understand the signal being sent?


[†] I have an as-yet-unread book on my shelves "The Myth of Mirror Neurons" ... that is now rising in my queue of unread books.

[‡] Kids won't understand because they're indestructible and it almost doesn't matter how they do it. There was this kid, maybe 16, at the Y who would do his exercises and *fidget* between exercises. So irritating to those of us who are just trying not to die while we do the exercises our doctors told us we *have* to do so we can die just a tad more gracefully. Pffft. I hate kids.


On 6/9/20 12:31 PM, thompnickson2 at gmail.com wrote:
> Under that notion, consciousness is at least dual: ie, there is a speaker and a listener and the former has influence over the latter and vv.  The writer talks back to the dictator, refusing sometimes to write what is dictated.  And once one admits of two agents of the mind, it becomes really easy to imagine others, so mental life becomes rather like reading last week’s FRIAM posts.    So, under that theory, when I say “I think”, I am like the public relations representative for large complex organization, the Kelley Conway of the mind.  Or like the writer of a Preface for a collection of Friam papers.  
> 
> So, your papers give credence in that view because, while there may be many voices “inside” “the head”, there can only be one that gets control of the focal apparatus, and that must require a lot of inhibition to keep us from blithering, even more than we do. And the battle for who gets the microphone must be constant and vicious.  


On 6/9/20 12:14 PM, Steve Smith wrote:
> I also find this fascinating.   Unfortunately I've not kept up with this
> work... it was very central-to but-unattainable-for the kind of work I
> aspired to do while at LANL with access to the funds and equipment to
> build high-fidelity (by those standards) synthetic sensoria.  We did NOT
> have access to fMRI's and similar... the closest to it I got was with my
> (on my own time mostly) work with the UNM HPC/Arts-Lab and the MInd
> Institute.  Human-Subjects testing is hard enough in academia, but
> adding the (very appropriate) layers that come with being at  Lab that
> works with nuclear materials was overwhelmingly hard.   The theoretical
> physics division had some plays in that area but they were somewhat
> distant from the weapons-physics programs that funded the high-end VR
> systems we were developing.
> 
> Observing the neural processing (in)directly of *lip-reading* would be a
> very specific application of something that was *just* becoming possible
> via fMRI and other things "back in the day".   The role of "Mirror
> Neurons" in non-verbal communication between individuals was
> fascinating, but as-yet understudied.   Your two articles *do* make me
> want to go learn what has happened in those areas... from what I read
> here, there must be a great deal more understood and *measureable* than
> 10-15 years ago.
> 
> I will happily ( I think ) sit back and listen to you and others reveal
> more about this work and it's implications, though I am sure I will be
> tempted to chime in looking for implications and applications to a
> broader experience than lip-reading.   Such as posture/gesture
> recognition human-and-machine. 
> 
> Just as hoity-toity-bait I'll throw in the word "Labanotation" and stand
> back and try to listen.

On 6/9/20 10:59 AM, uǝlƃ ☣ wrote:
> Motor Imagery of Speech: The Involvement of Primary Motor Cortex in Manual and Articulatory Motor Imagery
> https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6579859/
> 
> Observation-execution matching and action inhibition in human primary motor cortex during viewing of speech-related lip movements or listening to speech
> https://www.sciencedirect.com/science/article/abs/pii/S0028393211001801?via%3Dihub

-- 
☣ uǝlƃ