"Being scientific is hard for human brains, but as an adversarial collaboration on a massive scale, science is our only method for collectively separating how we want things to be from how they are."
#MultisensoryPerception changes as we get older, but how? @samuel_a_jones &co use multivariate #fMRI to show that older adults recruit similar brain regions, but increase activations in some of these regions to preserve performance #PLOSBiologyhttps://plos.io/4923V6I
Finally a lab grabbed the bull by the horns and had a go at figuring out the neural basis of the BOLD signal.
And this is quite the assumption indeed in fMRI-based studies, which has been known to be false for a long time: "most work has assumed a disjoint functional network organization (i.e., brain regions belong to one and only one network)."
A significant extension over early work with fMRI on epileptic patients that had chronic electrode implants.
Reading #fMRI methods when it's not your field sounds like a psychedelic voyage:
"Functional scans were collected using a T2*-weighted echo-planar imaging sequence with a multi-band acceleration factor of 4. A field map with dual echo-time images was acquired to correct for geometric distortions due to susceptibility-induced field inhomogeneities. Structural scans were acquired using a T1-weighted MPRAGE sequence with 1x1x1mm voxel resolution. We discarded the first six volumes to allow for scanner equilibration."
Periodic reminder of what fMRI's BOLD signal is measuring, and its temporal dynamics:
"blood oxygenation level–dependent (BOLD) contrast. [...] increased signal in a voxel measured with an EPI [echo planar imaging] sequence indicates recent neuronal activity because of the relative increase in local blood oxygenation that accompanies such activity. The temporal profile of this BOLD response, known as the hemodynamic response function, looks like a bell curve with a long tail, peaking around 4 to 5 seconds after local neural activity and returning to baseline after 12 to 15 seconds."
From "Principles of Neural Science", Kandel et al. 6th edition, page 115.
No matter how fast the EPI imaging is (~100 ms), the BOLD dynamics makes GCaMP look lighting fast. Temporally deconvolving BOLD is possible, to a point, but remember its spatial resolution is measured in millimetres, whereas neuronal somas measure ~0.025 millimetres.
Yes, indeed: the BOLD signal in #fMRI it's an indirect metabolic measurement, and therefore the activity of any cells in the volume will be included, beyond neurons and glia.
... where they report "an additional and equally direct pathway for cortical motor control driven by the primary somatosensory cortex" in mice. This surprising result, breaking with Penfield's century-old division of labour between the motor cortex and the somatosensory cortex, is supported by voltage-sensitive dye (VSD) imaging of the whole mouse brain, showing widespread activity propagation with latencies of tens of milliseconds across most of the whole brain.
Suppose we were to attempt a similar measurement in humans with fMRI. What would we see with the very low pass filter in the temporal and spatial dimensions that we get from fMRI?
Perhaps this is indeed the wrong question, and in humans we ought to be asking different questions. But what questions? Matyas et al. highlights:
(1) how fast and wide signals spread across the whole mammalian brain;
(2) the inaccuracy of old neuroanatomical divisions of the brain;
(3) the fact that a much larger fraction of the brain is involved in any one behaviour than previously thought, larger than would have been ideal to make sense of the brain as a modular system when its activity is monitored with the low pass filter that is fMRI.
My question is: what kind of studies are sensible given fMRI's strengths and limitations? I have one example in the next post.
In 2015, I attended the first talk ever featuring fMRI where I found the application and results convincing. The talk was by Linda Geerligs and she described her, at the time, recent paper:
The work focuses on longitudinal studies on humans, studying the same person over the years with fMRI. In her talk, she pointed out that:
(1) functional networks are not correlated with white matter tracts;
(2) there is a strong decline in functional network modularity with age;
(3) with age; intramodule connectivity decreases and intermodule connectivity increases;
(4) old adults that maintain the highly segregated functional network retain their mental abilities.
Overall I found the technique well suited to the question. In the first place because the study is longitudinal, comparing subjects to themselves over time. In the second place, because it did not attempt to make statements about specific areas of the brain, but rather, on how the relationships between areas, as far as they could be observed, changed over time within an individual.
An interesting study from which what we learn is matched to the strengths and limitations of the technique, and which does not attempt to make sweeping statements about the function of specific areas of the brain. At least that's what Geerligs did in her beautiful talk.
There may be caveats and limitations to the study, but as a non-fMRI specialist, I am unable to see them. Reads humble and as precise as it can be, not more and not less.
And to note that Geerlig's point (1) above on the mismatch between white matter tracts and functional networks should be pondered by those using tractography (another coarse spatial resolution technique) to make models of brain function. There are many such papers.
@NicoleCRust@albertcardona@tdverstynen@JosetAEtzel To understand, predict, control the phenomenon, one has to specify the phenomenon (“x” above). The timescale of the phenomenon can then help guide one’s search for a method with sufficient temporal (and other) resolution to match (or surpass) the phenomenon. For instance, would you like to predict value-based choice? #FMRI can help, because both occur on a second-to-second basis.
“In neuroscience, researchers have mostly focused on how the brain builds models of the world using a single example, such as how you build a model of the room you are in. This ignores the fact that your model of the room is built from building blocks that can be reused to make any other room or scene,” Tim Behrens, SWC Group Leader
In a new paper published in Cell, Behrens and colleagues at University of Oxford, UCL, Beijing Normal University and DeepMind, explored how the brain reuses such building blocks. They found that generative replay in the hippocampus and medial prefrontal cortex underpins such compositional computations: https://www.sainsburywellcome.org/web/blog/solving-problems-through-replay-brain
Practices such as preventing inflated error rates in statistical tests, replicating studies, and preregistering predictions are intuitive enough for a 1st year student to understand, and follow logically from the dominant philosophy of science (methodological falsificationism).
Even then, their adoption is slow, demonstrating how difficult it is to improve science. Researchers can always hide behind the handful of critics of these practices or arguing that reward structures need to change first.
@lakens Another fundamental of science is honesty and transparency. I think emphasizing (and requiring) that data and analysis code be shared as part of the publication is also a logical practice, and often easier to implement than, e.g., a full preregistration of an #fMRI study.
>Lonely people see the world differently, according to their brains
Brain activity differs among people who feel out of touch with their peers.
>To investigate what goes on in the brains of lonely people, a team of researchers at the University of California, Los Angeles, conducted noninvasive brain scans on subjects and found something surprising. The scans revealed that non-lonely individuals were all found to have a similar way of processing the world around them. Lonely people not only interpret things differently from their non-lonely peers, but they even see them differently from each other.
>“Our results suggest that lonely people process the world idiosyncratically, which may contribute to the reduced sense of being understood that often accompanies loneliness,” the research team, led by psychologist Elisa Baek, said in a study recently published in Psychological Science.
>Baek want to see if there was something to an idea known as the “Anna Karenina principle.” Leo Tolstoy’s iconic novel Anna Karenina opens with the line, “Happy families are all alike; every unhappy family is unhappy in its own way.”
>In this context, Tolstoy turned out to be right. The fMRI scans showed that the reactions of non-lonely individuals to the videos they watched were extremely similar. Lonely individuals had brain activity that was not only significantly different from that of non-lonely individuals but was even more dissimilar from each other, meaning that each lonely person in this study perceived the world in a distinct way.
>Baek suggests that having a point of view different from others makes the lonely even lonelier, as they’re less likely to feel understood (though she does mention that it’s not clear whether this is a cause or effect of loneliness—or both).
>Anyone who is lonely can now be assured there is probably someone out there who feels just as isolated—just in a completely different way.
Hey Mastodon users! This is my first post as I just opened my account migrating from @doctorpoe on twitter. I hope to find my old friends and make new ones among the mastodon conoscenti.
@ginapoe Welcome! Note that here hashtags can be followed, for example #neuroscience, #fMRI or #connectomics. And since search is mostly limited to usernames or hashtags, it's customary to add hashtags to posts so as to make them discoverable.
More on the story of that retracted paper on “predicting” suicidal ideation that @kordinglab and I identified serious errors in.
It turns out that the review process actually worked as it was supposed to. The paper was originally rejected, with two reviewers raising similar concerns as the post-publication public critiques (including ours) raised. Yet, despite the critical concerns and initial rejection, somehow Nature Human Behavior accepted the article anyway, without checking back with the original reviewers to see if the concerns (which ended up leading to the later retraction) were addressed.
This sort of editorial behavior only serves to accelerate the loss of faith in the scientific publishing.
Starting a list of all the amazing early career researchers that have made the wise decision to join us on this worthy platform (but might also take an exposure hit for that). Let's all boost and follow!