Remember the Tom Cruise movie, Minority Report?
This is just fascinating.
Anyone else attempt the experimental task a la @SkyzykS?
1 Like
I suspect this is more interesting to women, but since we’re talking about things that spark our curiosity. Not my field, but I’m really interested in some of the new studies about brain changes related to pregnancy.
Here’s a journal article about a new fMRI study. If you access it here through this CNN summary, and click on the Nature Neuroscience link, they will let you see the full study without hitting a paywall. Otherwise, you need a subscription. This is a workaround.
Note, gray matter reduction is not necessarily a bad thing at all, and it’s fascinating that we see this very dramatic kind of structural change in adolescence as well. Synaptic pruning can mean a more efficient, specialized system. Certainly we see that in the maturing brain. Adaptation within the lifespan like this is just fascinating to me.
Not a scholarly journal, but for a layman explanation.
Yes. And with the exact same result.
1 Like
So, to summarize the Kelso work: He found that ‘forcing’ (in this case, increasing) one parameter of the alternating movement resulted in the sudden reorganization of the phase relation of the system. It is as if the system has two preferred modalities (or ‘attractors’ in the parlance of chaos theory). Not surprisingly, Kelso found that the frequency at which the phase transition occurred varied widely across participants. However, if he first identified each individual’s ‘preferred frequency’ (determined by telling them to find their ‘I could do this all day’ rate), and he formed a ratio between their preferred frequency and the frequency at which they underwent the phase transition, he discovered that this ratio was constant across participants. Even when he changed their preferred frequencies (by manipulating the resistance against which they had to move their hands), he found that the frequency at which they underwent the phase transition changed as well, and in a manner that maintained the same constant as before. Hmm…Sounds like one of those ‘scaling laws’ we talked about with respect to the speech production research @Aragorn presented.
Weird. But it’s about to get a lot weirder.
Richard Schmidt, a doctoral student working with an extraordinary mentor named Mike Turvey, wondered if the same sort of phenomena would be found during inter-individual coordination of limb movements (as opposed to the intra-individual work Kelso did). So Schmidt got pairs of volunteers, and had them sit side-by-side at an angle allowing them to see each other. They sat on special stools designed such that one of their legs could dangle (and move) freely. A metronome was turned on, and the participants were told to swing their legs either in-phase (ie, up and back at the same time), or out-of-phase (ie, one going up, the other going back) manner. The metronome speed was slowly increased…
And the rest of the story can be found at the link. Again, once you open it, click on the PDF link on the right; this will open a copy of the original published paper. Fair warning: This paper is more jargon-y than the Kelso. If you look at nothing else, look at the two figures at the top of document page 235 (PDF page 9 for me). The top figure shows a pair of participants who were instructed to swing their legs in-phase the entire time (note that the frequency of swinging increases as you go from left to right on the graphs). The bottom figure shows a couple instructed to swing their legs out-of-phase; again, frequency of swinging is increasing as you go from left to right along the graphs. If you scanned the Kelso paper, this bottom graph should look familiar…
1 Like
I haven’t had the time to more analytically look at the Kelso piece other a quick glance. However, the first thought I had, was to wonder if women might perform better due to brain lateralization and efects on motor skills. But more importantly, I was curious as to if my wife’s decades of being heavily kinesthetic oriented - decades of playing Bach on the piano, doing court reporting school, and lifelong pursuit of handicrafts would allow her more skill in the test.
The article seems to refute that notion.
Fascinaring stuff and as Young Frankenstein tells the student The heart and kidneys are tinker toys. I’m talking about the central nervous system!
1 Like
Really, really outstanding breakdown of that paper! I was really hoping for some exchanges like this when I came up with the idea for this thread!
I really wanted your input on that paper because I found it fascinating and it is completely out of my zone. To your general point, I do not believe there is any chance that the ML-dervied models are going to hit the big time soon, so I do not want to give that impression. They are, however, laying the ground work for it and this is a point in time where we finally are starting to have access to the kind of computing power needed to crunch these numbers quickly and cheaply so I don’t doubt it’s coming. That, to me, is actually one of the major concerns I have–I have an inbred doubt about any “black box” system that cannot be easily peered into mechanistically.
Moreover, the bigger problem I have is the envisioned replacement of a doctor’s judgement. I see a LOT of highly educated people talk about “evidence based medicine” and practices, and this is a good thing of course, but the flip side of this is that many of these same people are under the impression that “evidence based” is a point-and-click interface. What if evidence is spotty, or equivocal, or if the evidence comes from poorly designed studies that says something that’s wrong (cough* steroids don’t work for athletes cough*)? Correct me if I am wrong AG, but your post seems to indicate you lean towards that skepticism to some degree.
It’s a funny position for me to hold because I am loving all the things big data mining can do, but any attempt to override professional judgment should be made very carefully and cautiously. As ED said, it’s already happening. This is an inevitable side effect of data in healthcare policy I think.
Ok so…back to the paper: I have zero doubt that EHR datasets are hugely problematic. I don’t there is any possibility that they could validate that many records with staff, and if they did I missed their description of it. The one thing I would have really liked to see is another group tested against a more complicated risk calculator–certainly not going to be popular with family practice docs in real life but for research I would have really liked to see whether the ML-derived work would have bested a more comprehensive calculator, say such as a specialist might use rather than a family practitioner.
You’re completely right about the risk calculator–it HAS to be simple to be readily applied by a non-specialist. And the small incremental improvement over a simple assessment model is a nice proof of concept, but most certainly nothing mindblowing.
I like brains, especially people’s brains. There are some things that need explanation (ML algorithms need it in my humble opinion as I already stated), and there are some things that machines cannot do as well as a human, and that is intuition and relationship.
2 Likes
That is a sexy, sexy car. I want it. I want it to have my babies haha.
Yes I did try the hand tap experiment you suggested earlier, I just didn’t have time to write back immediately. Scaling law—that’s exactly what comes to mind! Also I had exactly zero idea this inter-individual research existed til you posted it. Talk about cool.
1 Like
That is a great question. In fact, Kelso alludes to this issue (ie, performance among highly-trained individuals such as pianists) on page 90. Check out the paragraph that starts “Others as well as ourselves…” near the top of the page.
If you’re so inclined, do the experiment yourself, and have your wife do it, and let us know if there’s an obvious difference between the two of y’all.
Same result for me. I can’t contribute to the discussion as this is all new to me, but enjoying the thread.
I did read that about that about the pianist observation - counterintuitive to my pea brain.
I tried the tapping in 2 syncopated styles - right, left, right (like the sound of a horse in trot) and right, right, left also. The first is sustainable (for me) longer than the second, but both eventually cycle together.
One final observation - I wonder if the famed drumroll has traditionally culminated in the crashing of the cymbal, through the intuition of the drummer, knowing he can only hold the rhythm so long before synchronization or worse - spazzing out.
PS Maybe a drummer will comment, as l play guitar. Ha
Intriguing speculation regarding the development of a cultural tradition.
Science being harnessed for the good of mankind.
Not a drummer, but this is one of my favorites-
I don’t know how to analyze that, but it sure is neat.
Would have been funny if Kelso had inadvertently signed up Peart as a participant. Probably would have crushed Kelso’s entire thesis.
Lots of head scratching… “Why isn’t this working?”…
Exactly…While Peart is tapping out a syncopated rhythm in 6/4 time with one hand, a bossa nova with the other, and doing long division with a pencil held in his teeth…
Black Swan disproves theory?
Meh, that Black Swan stuff is way overrated. 
Here’s a fun one- Holding hands out, wrists straight, index fingers pointing toward each other- rotate one finger clockwise while rotating the other in the opposite direction simultaneously.
We used to try that when we were kids, but there was one guy that could do it.