From my experience, reproducibility is the single biggest challenge with any fMRI study. I was a part of a reproduction study done by a group which did the original study in question, and the results were 100% indicative of "random chance" instead of any meaningful pattern in the data. That seems particularly problematic when the same techniques and data set is used by the same group of people.
Of course, this is looking at the brain's power management system and trying to infer from that what processing is going on. If you did that for a CPU, you could see the FPU or the I/O controllers heating up, get a sense of how many cache misses were going on, and see how many CPU cores were doing anything. You could probably tell a game, spreadsheet, or web browser apart. This would not lead to much insight.
Except this is more like studying same where you don't actually know what a cpu is let alone cache miss. It's more like a caveman looking at pictures on the screen and noting that a certain area lights up when there is more red on the screen.
It would probably lead to more insight than not looking at which parts heat up. State of the art before fMRI was looking at people with very specific brain injuries, or using EEGs, which offer even lower fidelity.
And why is that a bad analogy? The fmri results show activity at a resolution that would be equivalent to sayign SoCal has more road activity (and to further your analogy) and economic activity than say Wyoming.
Especially since most fMRI studies of note tend to be ones studying specific inputs. It makes total sense to see what parts of the brain light up with a given stimuli in a given set of conditions. Of course you wouldn't extrapolate that to the entire brain, as brain research is as much about what does light up as it is what doesn't.
I’m mostly using fMRI visual system data and am repeatedly surprised how fine-grained the patterns we can find inside it are, if it‘s recorded well (e.g. lots of data in single subjects, no group averaging, very little motion). My field is well aware of this method‘s shortcomings, many including me would agree with Schmidhuber who said it’s like looking at heat distributions in a processor doing ML in their darkest research moments. Shortly after they discover something interesting in these patterns and another cool preprint is out.. Functional MRI is without doubt the best window into patterns of higher brain functions we currently have.
I wonder what it must have been like for neuroscience and the scientifically interested public when the fMRI method came up in the 90s. For human research they had EEG before (and PET which seemed to be quite cumbersome). Would love to read an anecdote.
PET with Tractographies has been used in a couple of research papers and the real world to map out the extent of brain material related to clinical observations. This is done with TBI's in particular.
This is from 2017 and regarding fMRI research conducted at Stanford. Several of these guys actually worked together 48-years ago and also at Stanford. The person talking is a former CIA officer and medical doctor. And yes he's talking about aliens. This is him - https://www.youtube.com/watch?v=rvovZBT54Oo
------
"Now if there was to be point to point, communication attempt, I can tell you three things about that experiment.
Number 1: About one third of that, I'm the only expert on Earth. Because of the functional MRI, connectivity systems of what parts of the brain are related to other parts of the brain for communication and quantum telepathy.
The second thing is the software associated with the calculations and the manner in which the scan sequences would be done at two different locations. So where the technical people doing the scan sequencing with the radio frequencies used for MRI, in both locations, were coordinated. And using the same (Haley Eye?) coordinates for looking at the brain, simultaneously...
So what the issue is that the sponsor's interested in, is tying that all together. And the sponsor is interested in one dimension of that experiment. And I'm interested in the second. Hal's interested in the third.
And here the dimensions are. I'm interested in the dimension of showing that quantum connectivity between two proven, competent, telepaths can be measured with simultaneity in terms of the portions of the brain that are activated. That that is now possible to do.
I don't do the math. But I know how to determine what portions of the midbrain connected to the caudate-putamen, connected to the prefrontal cortex and connected to the part of the brain that does imaging...is all related to each other. That's something that right now is a very rich area of my world of MRIs.
Hal is the expert in knowing how to design the experiment. And the sponsor is interested in having an experiment design where the thing that you do and the thing that Uri does, happen to be identical to each other.
Both you and Uri have asserted, and in your case, you have, in my judgment, proven - in Uri's case, I can't say - that the connectivities that you get for the acquisition of information, includes The Others."
