Slide Deck of My Talk on Jeff Hawkins' On Intelligence

[James Heaps-Nelson]

Given that Jeff Hawkins wrote his new book A Thousand Brains in 2021, I thought I'd give the book a mention here. I gave a talk on the first book, On Intelligence, to Arizona Objectivists in 2007. Hawkins gives this updated view on a theoretical framework of the human neocortex.

On Intelligence talk.pdf

[Michael Stuart Kelly]

James,

Long time, no see. 

It's good to see you again.

 

Here is the video in the PDF slides. (btw - Do you have a video of your talk?)

 

And here is a link (referral) to A Thousand Brains: A New Theory of Intelligence, the more recent book by Jeff Hawkins.

There are several videos of Hawkins on YouTube that discuss this new book and I have not yet seen them. So I cherry-picked a couple that seems to be good for a 2-3 hour crash course each.  I will watch them later and probably comment.

 

and

 

I read all of the slides in the PDF and I cannot discuss this well yet since I am not familiar with Hawkins's work. I like the way you tried to tie all of this to Objectivist epistemology. (btw - I never read the Efron stuff essays. I need to correct that.)

 

I have always been frustrated that Objectivism does not have a good theory of memory. That is one of the things that led me to study neuroscience, granted from a layman's perspective. And that led me to several people I want to mention as a complement to any discussion this thread prompts.

This first deals with narrative and its role in memory, mainly through the default mode network, but also through the hippocampus and through the blend of motor neurons and sensory neurons. The more I study that, the more I get into schema instead of narrative per se. (In my understanding, and to use a simplification from Aristotle, a narrative has a virtual beginning, middle and end. A narrative schema would just have the beginning and middle, but the end is open-ended and depends on what appears or what happens. Also, the beginning and middle can be changed at times, but all that's for another discussion. Just one thought. A narrative deals with the past. Schema, although containing part of the past, exist to deal with the unknown future.)

 

Speaking of the default mode network, there is fascinating work being done at Princeton by a guy named Uri Hasson. He doesn't just talk about motor neurons, he shows how the entire default mode network lights up on fMRI scans in the same manner at the same time in the brains of the storyteller and the listener. This has some deep implications for memory if the story is about something the listener does not yet know, thus has no memory of it. How's them apples, hmmm?

There are finally a lot of Uri Hasson videos available on YouTube. I looked for them for years, and now they are coming out. But Uri-baby looks like a German Nazi and has an almost unintelligible accent.  Since transcriptions are available, you have to watch a video and read the transcript at the same time. Despite my quip (I can't help myself  ), his stuff is great.

 

About a year ago, Stephen Boydstun posted a thread about an Irish neuroscientist, Veronica O'Keane, who specialized in memory. She wrote a marvelous book, A Sense of Self: Memory, the Brain, and Who We Are. Another title that got published for this same book is The Rag and Bone Shop: How We Make Memories and Memories Make Us.

Here is the thread and there is quite a lot of information I added to it at the time: Veronica O'Keane. I remember Stephen posting his own comments about O'Keane's book, but he comes and goes on OL and often deletes his stuff when he goes. So there is only a link in his opening post right now. I think he gets pissed because I like Trump and Trump just won't go away, nor will I change my mind.

But I like Stephen. Oh well.  

Since we no longer have the benefit of his comments about O'Keane without trying to find it somewhere out there on the Internet, let me give an update to my own comments on that thread. I got this book and audiobook and I went through a few chapters, but stopped. I am trying to figure out Anki so I can do flashcards from her book. (sigh... Life is short and all this neuroscience mumbo-jumbo is long. And so is learning new programs.  )Then I will start it again and learn all the big words along with the ideas. This is an incredible book from a neuroscience and memory perspective. I believe this is relevant in the max for this thread.

My only caveat is that O'Keane mars her book with a constant bug up her ass about Sigmund Freud--from a feminist view at that.  If you can ignore the excess of feminist anti-Freud ranting, the rest of her book is pure gold.

Apropos, this kind of reminds me of another book I read a while back on neuroscience and emotions: How Emotions are Made by Lisa Feldman Barrett. This lady presents a lot of great information, really, top-quality stuff, but she mars it with a constant bug up her own ass about her assertion that free will does not exist and that emotions do not really exist because you cannot see one. (You can't see a concept, either, but I don't remember her going into that. She may have, though. It's been a while since I read her book.  ) 

What is it about big brains with pet peeves?  

 

I am going through quite a lot of other books and videos, too. The Master and His Emissary by Iain McGilchrist for knock-your-socks off left-brain, right brain stuff. I already went through some Damasio a few other select neuroscientists, but I feel like I only scratched the surface. I intend to read a lot more from them. As an aside, I also did Gazzaniga with snow shovels for chickenshit in split brain patients with a severed corpus callosum, but I have only read one book of his so far.  

I could go on, but there are some other thinkers and works I have delved into that bring me greater pleasure. They go deep into neuroscience and story. That's what rings my ding-a-ling. One is Angus Fletcher. This guy teaches military dudes at on DARPA how to weaponize story.  There is a fascinating screenwriting book, The Science of Screenwriting: The Neuroscience Behind Storytelling Strategies by Paul Joseph Gulino and Connie Shears. Gulino is a screenwriting professor and Shears is a neuroscientist.

I am only going to mention one other right now, but there is a world of knowledge about this stuff out there and I am like a kid in a candy-shop.

 

There is a book by Bill Birchard called Writing for Impact. This one is a bit different. Bill is a marketing copywriter and ghost writer. One day he got it into his head to look at the published scientific literature about neuroscience and the different parts of writing. He noticed that a lot of writing bores readers while other writing keeps them interested and turning pages.

So he went through a truckload of that stuff to see if science could give some answers as to why. He has an interesting starting premise, too. If a certain writing technique, when read, caused fMRI and EEG scans to light up, the reader would be a lot more invested than when it did not cause any activity.

Through his study, he found 8 areas where different parts of the brain light up and the writing techniques that cause this in experiments documented by peer reviewed articles. Briefly, he came up with 8 writing techniques (or focuses would be a better term) that improve reader engagement with repeatable results under controlled environment experiments.

These are:

1. Writing that is simple
2. Writing that is specific (as opposed to general),
3. Writing that has surprises,
4. Includes emotions (even just mentioning an emotion works),
5. Prompts anticipation and curiosity,
6. Contains insights and epiphanies,
7. Writing that is what he calls "social," which means letting the reader inside the heads of characters and the writer to see them thinking and feeling as humans, and
8. Story writing--where he goes to town on the default mode network, including its impact on memory when lit up with story.

Apropos, I want to revise this post using this as a checklist to show how powerful this stuff is, but I just don't have the time. It's like that old saying that I am sorry this is so long. I didn't have time enough to make it short.  

 

I want to share your enthusiasm for relating this stuff to Rand's epistemological theories, but there are a few areas where I find her conclusions problematic. I never want to say she is wrong because of the constant controversy people get into when discussing her. But I do like to separate the good from the bad. And I like to add to what she did (and giving her credit for it) while discarding the stuff that just doesn't make the cut.

Here's an example: her idea of measurement omission in creating concepts. This works for a lot of cases, but when I look at schema and how that works, I figure there is a lot left to discuss. For example, you can force the measurement omission idea onto a concept like birthday party, but schema does it so much better. What do you think about when you think birthday party? I think about friends, cake, candles, presents, festive environment and so on. (Granted, all with measurements omitted, but with a hellish differentia to find to tie it all together.  ) Schema is made by memory, not abstraction per se. Schema makes it easy to focus on what needs focusing on when violations of the schema appear. For example, if a cadaver falls through the ceiling during a birthday party, that would violate the schema and everybody would look. Or ditto if a homeless dude came to the door smelling awful.   

When we look at motor memory and compare it to sensory memory (each with different kinds of neurons that process time differently), that starts to explain where Rand's ideas and these others can be combined to make good epistemology, meaning correct and useful at the same time.

 I want to go on longer, but, hell I don't even know if you are going to read this sucker.

 

Anyway, it's good to see you, James.

Michael

[James Heaps-Nelson]

Hi Michael,

Unfortunately I do not have a video of my talk, but it followed the slides closely enough that I don't think there is anything significant left out. I'm glad that I did a careful review of the first Hawkins book because the second has a quite revised human intelligence theory involving a multitude of reference frames sort of voting on an answer as Hawkins puts it and he deemphasizes the role of hierarchy in cognition. 

I'm glad you took the time to suggest a multitude of avenues of new study in neuroscience. I need to go wider. My main influences are Damasio, Hawkins, Kandel, Joseph Ledoux, Nancy Andreason, Howard Gardner, Daniel Levitin, Gerald Edelman and Daniel Amen. I have a book by Gazzaniga whom you mentioned, so I'll probably read that one and I will follow up on the videos and most of the references you mention. I especially like Joseph LeDoux's book, The Synaptic Self because it sort of goes through the development of neuroscience in the '70's with a detailed history of the study and understanding of long term potentiation in the hippocampus and a detailed description of multiple neurotransmitter pathways. It's interesting to compare Nancy Andreason's book The Broken Brain with LeDoux's The Synaptic Self because Andreason's book was written in  1985 and Ledoux's book was written in 1996 and there was this huge explosion of growth in the understanding of neurotransmitter pathways in the intervening years and the number of neurotransmitter pathways indentified, characterized and understood grew dramatically. 

Anyway, thanks for your thoughtful response!

Jim