Differences between expert and novice brains in mice: study
cshl.edu
cshl.edu
Summary:
Main question: how can one develop an effective intuition?
The development of an effective intuition relies on: a high validity environment, a high amount of practice and a high interoceptive awareness. The first two elements account for intuitive answers being right. The third element accounts for the gut feeling being felt by the person. As a result the brain changes, and integrates certain brain areas more into the default mode network. The idea that if one has high activation in a brain area a lot, then the structure will be changed.
High validity environment: rules are easily understandable and deterministic (think chess), or you have a lot of practice moments (think poker). Understanding psychological patients would be an example of a low validity environment.
In your review you cite cross-sectional research, not a longitudinal, within-subject effect of change with study/practice/learning. Therefore your conclusions should not be about change, but on group differences. Those difference might have arisen from the training, or pre-existed in those individuals that were more prone to study/practice/learn/meditate.
Post-Note: Sorry. As a developmental neuroscientist, you hit on a pet-peeve. Cheers.
Personally, I do think the brain is plastic in this sense though. I haven't looked deep enough into it to be a 100% certain. Nevertheless, from all the research summaries I read, I gather that whenever a skill is practiced in a focused manner for 1 to 2 hours, neural plasticity occurs.
I wish there was a literature review around this idea. But I got the feeling that 4 years ago the literature was mostly on clinical patients. Normal people and neural plasticity seems to be researched less well.
I always figured people were speaking mostly metaphorically about "gut feelings." Certainly, some kinds of intuitions (e.g. ones involving disgust or fear) are held onto as "physiological state" (i.e. written through motor-cortex commands; read back through pattern-detection in embodied cognition.) But I would expect most intuitions to present as e.g. sudden emotional states, or other weird purely-mental qualia.
Intuitive feelings that I acted on were all body-related in my experience. In most cases, these intuitive feelings always had to do with assessing who another person is. My intuition about that seems to be pretty accurate.
I think there's a lot more happening in your body than you might realize. I've noticed that myself by practicing mindfulness. However, I don't think it rules out your idea that some intuitions are purely mental.
By the way the word emotion is a tough one, as I believe there is always a bodily component part to an emotion. I subscribe to the James-Lange theory [1] and the appraisal theory. In part, because I experience emotions in those ways before I even knew about the names and knew that these theories existed.
In short how I view emotions is process-based:
- An event occurs (thought, feeling, someone talks to you)
- This automatically gets processed by you whether you want it or not (though when it's automatic, obviously you're not using your full cognitive resources on it)
- This automatic processing concludes into two things:
1. You get thoughts about it
2. You get feelings about it (feelings in your body)
Note: thoughts and feelings are immediately re-triggering this process. So it gets meta pretty quick as you can have feelings about feelings about thoughts about feelings about <mix_and_match_as_long_as_you'd_like>
During my meditation retreat I realized that feelings come slightly earlier than thoughts. At least that's how I experienced it and I happened to be laser focused on observing my feelings and thoughts.
Coincidentally, that's also why I believe I have no free will as my feelings give rise to my thoughts, even when I simply observe my feelings. I don't choose that these feelings give rise to thoughts (and there were times where I really didn't want them to give rise to thoughts), therefore there's no free will. But that's another story :D
[1] There's probably a more advanced theory of this one as it is pretty old. For practical purposes, this theory was good enough for my understanding on what emotions are.
Conversely, this lack of experience and humility and the bubbling exuberance and over abundance of confidence that comes with inexperience tempts people to walk a path that someone with more experience dare never venture. It's these paths that help move us forward into the unknown.
Our confidence comes from different places. An expert's from the confidence that they know their place in the world. They've learned their reality, they've learned to build tools to deal with and capitalize on that reality. Their experience has taught them how to survive and, if they're lucky, prosper. While a novice's confidence is often borne out of idealism, hope, arrogance, ego and sheer audacity.
There is not only room for this in the world but, I believe, a need for it. That we need each other for humanity to move forward.
And in AI it’s basically the same. The fewer number of hidden nodes in your neural network the better, as long as there are enough nodes to accomplish the task.
I just didn't think it was relevant to this discussion to go deeply into the details of neurogenesis.
In ML, think of evolution as hyper-parameter tuning. Most ML models have a defined structure but are tuned via learning. Pruning is probably such a tuning strategy in the brain.
But the first thing I noticed is that here's a guy writing a paper about skill progression and he hasn't encapsulated the act of writing a paper about skill progression as a progression of its own.
I suspect it's more like Novice, Advanced Beginner, Competent, Proficient, Expert, Artist, Philosopher.
1. Unconscious incompetence The individual does not understand or know how to do something and does not necessarily recognize the deficit. They may deny the usefulness of the skill. The individual must recognize their own incompetence, and the value of the new skill, before moving on to the next stage. The length of time an individual spends in this stage depends on the strength of the stimulus to learn.[5]
2. Conscious incompetence Though the individual does not understand or know how to do something, they recognize the deficit, as well as the value of a new skill in addressing the deficit. The making of mistakes can be integral to the learning process at this stage.
3. Conscious competence The individual understands or knows how to do something. However, demonstrating the skill or knowledge requires concentration. It may be broken down into steps, and there is heavy conscious involvement in executing the new skill.[5]
4. Unconscious competence The individual has had so much practice with a skill that it has become "second nature" and can be performed easily. As a result, the skill can be performed while executing another task. The individual may be able to teach it to others, depending upon how and when it was learned.
This can end up being an uncomfortable position when people ask you to explain exactly why you made a decision, and even though you probably could explain it with some effort, it's difficult because you made your choice based on experience-derived intuition. Being at the conscious competence stage may actually require less work in general, unless one gets to work with other unconscious competent people.
The code structure just needs to feel right. It's hard to throw principles at it to show which structure is better. To me it always feels like there are so many aspects to take into account, that cannot just be expressed by priciples.
Some code feels right, and some doesn't. Sometimes extending code "falls into place" and confirms that your previous reasoning was the right one. And other times it just remains a complex mess ;).
When you're a native English speaker, you're in the unconscious mastery phase for the most part. Things feel right or wrong.
The 2 levels above are being able to explain why. That's when you're at the "able to teach English" level.
I think the same applies to code. You can be very very good at something, but true mastery is being able to impart that knowledge on others as well.
Is that true though? How do we know that the ability to teach is not a unique skill in and of itself rather than just an extension of one's "mastery"?
Anecdotally, I've engaged with "masters" of various fields as different as blacksmithing to theoretical physics. Of these "masters", I've encountered good teachers and bad. But I've not personally felt any correlation between their mastery of the subject (i.e. their ability to produce results in the area) vs. their ability to teach the area.
I've had various trainers in boxing. Most of them were damn good boxers. One of the best was a real talent, been boxing since he was 10. He never went very far with competing, but ho boy did he look great at the gym.
When explaining how to do something, he really struggled. Most of the time he could show us what he does, but not explain how he does it, why he does it, or sometimes not even realize what exactly he does until he tries it. Then he's like "Oh yeah I guess I do take a little step there, you're right"
Then there's a trainer who is a little less talented himself, competed at a pretty decent level as an adult, but decided to pursue training instead. He can tell us what to do and why we should do it. Also how to think about strategies and whatnot. He's great but we need a lot of repetition and trial and error.
Then there was the lady who is an 8 time professional world champion on Muy Thai. She came to our gym to become a better boxer and to get free membership she offered to teach.
Dude, that was on another level. She could not only explain why we should do something, but also how. She'd look at your punch and be like "No no, move your wrist 5 degrees this way" and damn did it immediately make an impact.
3 months of working with her created more boxing improvement than 3+ years of other trainers has. It was phenomenal.
The difference was that she wasn't just talented, she also had a deep understanding of exactly what you have to do, why you have to do it, and how you have to do it. Her advice was measured in half-inch adjustments to your technique.
Teaching might be a separate skill, but true mastery is being able to break down exactly how you do what you do and why.
Oh and better yet: she explained how it should feel when we get it right. So not only would we know what to do, but also how to tell if we did it right.
I also have taught English to novice speakers and sometimes am struck by such questions they receive, sometimes being all of: obtuse, irrelevant and pedantic at the same time. But that is their primary objective: to sell tests and teaching materials to their tests, not direct competency.
But teaching isn't simply being able to explain why something is the way it is. It's more than that.
I could explain to you why planets move the way they do by using the language of tensor fields built up out of tensor products mapping vectors to their duals. It would be a thoroughly accurate definition of why we see the movement we see but it would not necessarily be a good way to teach. Particularly if the recipient of the teaching wasn't familiar with the formalism of general relativity.
It's this aspect, that teaching is more than simply understanding that makes me think it potentially could be orthogonal to mastery of a subject.
The author had almost no mountain biking skills and a friend who was one of the best in the world. The book was built up during the author’s advancement through the stages by observing the expert, who was largely incapable of communicating exactly how he was doing what he was doing.
The line "The individual may be able to teach it to others" addresses precisely this point. Having unconscious competence is orthogonal to being able to explain what you are doing.
ps: and ultimately, when I compare my younger self to me right now (in music) .. the only difference, is a subtle blend of patience and sensitivity. When you don't know you try small things without waiting too long or digging too deep. A newcomer may often go paralysed or stuck on the wrong intuition.
One amusing state you can add when you get old is something like this:
5) Semi-conscious semi-competence
The individual used to be an actual expert in something, but so many years have passed that they are only semi-competent. Yet they retain the manners and affectations of someone who is a true expert. This state tends to confuse and alarm others who are themselves competent and can’t figure out exactly what you are.
Freshmen know not, and know not that they know not.
Sophomores know not, and know that they know not.
Juniors know, and know not that they know.
Seniors know, and know that they know.
...you realize that you still "know not that you know not, know that you know not, know not that you know, and know that you know" about most everything.
Thus becoming humble and wise...?
He knows not and knows not that he knows not is fool. Shun him.
He who knows not and knows that he knows not is a student. Teach him.
He who knows and knows not that he knows is asleep. Wake him.
He who knows and knows that he knows is wise. Follow him.
I'd prefer to say wisdom is knowing what you know and what you don't, and it's much harder and more valuable than you might think.
* https://terrytao.wordpress.com/career-advice/theres-more-to-...
― Niels Bohr
But more seriously from Suzuki:
the beginner's mind has many possibilities, the expert's mind has few.
From the wonderful book "Zen mind, beginner's mind"
How about an MRI of a newbie in flight school and an MRI of a seasoned pilot. What does that show in terms of neural development and associated pathways? What does the newbie's brain look like after extensive training? Can you measure neural pathways over many years and see how they grow or change?
A novice will see so many pieces. So many possibilities. They have no idea how to choose between them. And there are so many, they'll fail to see an excellent move, or that one of their pieces is about to be killed.
An expert sees far fewer moves, and so is able to look much further ahead.