Introducing the Action-Mode Network – AMN. Better late than never! 17 years ago, we clumsily called this the Cingulo-opercular Network (CON) based on its anatomy (we weren’t exactly sure what it does). Now we’ve finally got the evidence to give it the functional name it deserves:
Psilocybin's powerful short- and long-term effects on your brain are revealed by this intrepid precision imaging drug trial published in
@Nature
. As one of the scientists I was also a study volunteer. Spending hours in an MRI scanner, while tripping on 25 mg was the experience
After a high dose of psilocybin, the brain desynchronizes at a massive scale, causing loss of our sense of self, time, and space. This may drive the burst of plasticity caused by psychedelics. The next day, brain activity has largely returned to normal, but an echo remains – a
This paper is not about fMRI. It's about brain-wide association studies, BWAS = studies of the associations between common inter-individual variability in human brain structure/function and cognition or psychiatric symptomatology’. fMRI ≠ BWAS.
Ever notice how mind and body states interact. How thinking about a future action, can make your heart race, even though you’re just sitting in a chair. How trying to solve a problem can make you pace back and forth. There's a new
#brain
network for that:
We’ve identified the somato-cognitive action network (SCAN), a newly recognized network within human primary motor cortex that disrupts the famous—but incorrect—motor homunculus, and has strong connections to high-level control networks. Now out in
@Nature
'Precision Neuroimaging for Localization-Related Psychiatry' JAMA Psychiatry opinion w/ Tim Laumann & Chuck Zorumski: " ... mapping the brain’s true functional organization, not the fictitious group-averaged brain, is just beginning."
Meeting PS1 in 2012 inspired individual-specific precision
#fMRI
mapping. Article out today in
@lancetneurology
w/ Tim Laumann
#neuroplasticity
Brain network reorganisation in an adolescent after bilateral perinatal strokes
"Parallel hippocampal-parietal circuits for self- and goal-oriented processing" aka most of your hippocampus is busy with the simulation of 'you' ... and the 'get stuff done' parts of your brain are left with the tail (15%):
By the brilliant
@itsyungzheng
The brainstem’s red nucleus was evolutionarily upgraded to support goal-directed action () ... Originally for quadrupedal locomotion, it is for action in humans. Functionally connected to the action (CON, SCAN) and salience networks, including ant.
We show that the midbrian red nucleus is functionally connected with the salience network (involved in motivated behavior/reward) and cingulo-opercular network (CON; involved in action control). See preprint here:
@DrDamienFair
@ndosenbach
@gordonneuro
Did a thing I always dreamt of: wrote a piece for
@sciam
. It's about uncovering the somato-cognitive action network (SCAN); just hiding in motor cortex for 90 yrs (). Thx for the great edits
@gstix1
and the encouragement
@meharpist
. Lessons learned 👇👇👇
21 MacArthur ‘Genius grant’ winners — including a neuroscientist, econometrician and historian — receive $625K each
@DrDamienFair
wins the genius award! Way to go. Most deserved. & you're buying bro ... for the rest of the century.
A Precision Imaging Drug Trial (PIDT) on the cover of Nature - big score for Precision Functional Mapping (PFM)
Psilocybin is the hallucinogenic compound found in magic mushrooms. As well as its psychedelic effects, it also shows promise as part of a treatment regime for
has lots of parts to it. I like
#neuroscience
Twitter for getting a quick overview of what's happening.
@smarek0502
&
@tervoclemmensb
wrote detailed 🧵s with figs. I'll just give bullet pts of what the study is & isn't saying 👇
This
@Nature
precision functional mapping in depression article by Chuck Lynch
@cl9681
and Conor Liston is unbelievably good. It's got the best and the most data, all the analyses, ... thorough, thoughtful. It establishes important new facts about depression ... not ideas or
Happy to share that our work applying precision mapping to individuals with depression sampled longitudinally over an extended period has been published this week in
@nature
. Brief recap of the main findings and their potential implications below.
Tried out the new
@caseforge
custom mold & combined it with
@firmmsoftware
. So we can get the lowest movement
#BOLD
#fMRI
data ever collected. Also ... you can look cool ... like a Spartan from 300 ... or Magneto
Precision Neuroimaging Opens a New Chapter of Neuroplasticity Experimentation
@DrDamienFair
and
@bttyeo
putting network 'disuse/plasticity pulses' in perspective and calling for help figuring out what they're all about.
ICYMI:
Structural and functional MRI are extremely useful. The point we’re putting forward here is that if you’re doing cross-sectional correlations against behavioral traits you need large samples. These results don’t speak to within subject research.
Here’s the summary. The active mode is the yin to the default mode’s yang ☯️ … and just as the default-mode network (DMN) has a special role in establishing the (brain-wide) default mode of brain function, the active-mode network (AMN) sets up the action mode. In both cases, an
The hippocampal-parietal circuitry for navigating spacetime seems to come in two versions. A small one for for the world (hippocampal tail - parietal memory network [PMN]) and the lion's share for the self (anterior 4/5th hippocampus - default mode network [DMN]). Go Annie!
After 3yrs of gestating as a PhD student, I’m introducing “Parallel Hippocampal-Parietal Circuits for Self- and Goal-oriented Processing”. My 1st paper out of my PhD is the result of 3yrs of failing better. Here to normalize that sometimes it takes a while.
As you might expect, the action-mode network (AMN) and default-mode network (DMN) are anti-correlated. The AMN consists of a subset of
@foxmdphd
’s and Marc Raichle’s original task-positive region. That is: the AMN is task-positive, but not all task-positive regions are part of
Great article by
@avashkam
capturing the nuances of
#BWAS
() with quotes from the great
@russpoldrack
and Sébastien Jacquemont. You can study brain-behavior links in individuals (H.M.; not BWAS), but cross-sectional correlations require big samples
#fMRI
#BWAS
variables are correlated with head motion & even after rigorous denoising there's trait-specific motion distortion.
@ScienceBenKay
found solutions. Good news: The effects can be quantified & frame censoring works! ... but beware of BMI
The good news is that frame censoring does a fantastic job of reducing false positive motion impact score. That means if you motion censor your data after standard processing then you are much less likely to get spurious results due to motion. Highly recommend trying it!
Awesome
#Parkinsons
work by
@DavidRen555
and
@hesheng3
. Must read preprint. Peter Strick was the first to suggest to us that the SCAN (somato-cognitive action network: ) is the PD circuit. I was convinced, but we didn't have evidence. ... it just dropped
🧠🎯Check out our new work on the somato-cogntive action network (SCAN) in
#Parkinsonsdisease
and its relationship with diverse neuromodulatory targets for PD.
A few key findings below!🧵
Academia fun. Who wouldn't want to review () a review of their article (). There's a great thread with details. Summary: We're doubling down on everything.
#NOmunculus
Muret and colleagues recently posted a public review of our
@Nature
paper describing a novel somato-cognitive action network (SCAN) within human primary motor cortex. ().
Bill Seeley’s aptly named Salience network has been most commonly confused with the AMN (previously CON), but they are different networks (Bill agrees: ). They’re both cingulo-opercular, with representation in the cingulate and frontal operculum (another
I've been dreaming that the Ooi-Yeo calculator will one glorious day replace the grant "power analyses" everyone enjoys so much. ... & remember ... if you don't enjoy logistics and don't have much $ funding. It doesn't have to be BWAS. There's also PFM ... 3 volunteers might be
1/11 The poll has ended: 53% (scan time) vs 47% (sample size)!
Here's our in-depth take on brain-wide association studies (with a 134-page supplement!):
Led by
@Leon_Oo1
@csabaorban
, there are a few twists, so do read till the end! Our previous study ...
I met Daniel & Kellie 10 years ago in
#pedsneuro
at
#WashU
. Their inspiring story is the cover of
@washingtonpost
Health today: . It's been driving precision functional brain mapping with Tim Laumann & co, ever since . Thanks Carr fam
Precision Functional Mapping of Corticostriatal and Corticothalamic Circuits: Parallel Processing Reconsidered . No need to read our full article (). The brilliant Charles Lynch and Conor Liston explain it best in the preview.
What are these pulses ... that show up in temporarily under-used circuits ... for? Are they good, bad, neutral? Are they epiphenomenal or are they helping with brain allostasis? Are they similar to sleep spindles or prolonged periods of beta in Parkinson's? Something else? Not
How to find massive motor plasticity effects in the subcortex? Find out in our preprint. Turns out these new findings change our understanding of
#Parkinson
disease and some
#sleep
processes. Let me unfold it
With
@ndosenbach
@gordonneuro
@dillannewbold
Many small neuroimaging studies have probably reported false positive findings.
Especially for traits related to cognition & mental health, thousands of subjects are required 🧠
New paper from our talented PhD student Shu Liu in
@NatureHumBehav
:
Solving the BWAS reproducibility crisis requires large samples:
Sticking with the status quo won't be enough. De facto, smaller data sets often ended up overfitted. Great showcase here:
Note!:
#BWAS
≠
#fMRI
Cross-validation is critically important, but insufficient by itself.
Published out-of-sample effect size estimates (g) look very similar to in-sample overfitted effect sizes (f, light blue).
If you like the NOmunculus, marionette model of M1, you might like working with
@JarodLRoland
and me on precision functional mapping in pediatric neurosurgery patients. We're looking for a Clinical Research Study Assistant:
We’ve identified the Mind-Body Interface, a novel distributed network within human primary motor cortex that disrupts the famous—but incorrect—motor homunculus, and that exhibits strong connections to high-level control networks. Preprint here:
Thanks for open reviewing our somato-cognitive action network (SCAN) article! ... which reminded me that the
@Nature
reviews are available:
& a comment by David Leopold
A response to Muret et al is coming
🧠⚖️ Weighing the evidence in the recent
@gordonneuro
paper published in
@Nature
on M1 functional organization. Our open review with
@diedrichsenlab
and
@plasticity_lab
questions some of the claims & calls for more thoughtful interpretation.
Read it here:
Washington University is hiring! Come join us in the Neuroimaging Laboratories within the School of Medicine. It's a wonderful, collaborative environment with brilliant students, tons of resources, and some of the very best science done anywhere.
@boeslab
Exactly & you’re saying it with MRIs. Human neuroscience has a ton of options besides BWAS. So if you don’t want to deal with 10,000 participants, … you don’t have to. With the right approach N=1 can generate great insights.
Confusion over functional network names runs much deeper than just the AMN and Salience, of course. And many … especially the great
@LucinaUddin
… have been battling that confusion. Several previously proposed networks such as the MDS (multiple demand system), extrinsic mode
@dillannewbold
did the best job telling the pink cast study science story: behind
and there's excellent write ups from
@sourwine
: and
@maxdkozlov
: . So I was gonna tell the story story
A testable hypothesis is that regions which process pain stimuli should include both CON and SCAN. Here we show a publicly available map of pain processing (from ). Both CON and SCAN were activated by this pain stimulus.
Missed the last flight back to St. Louis on December 23rd, 2014, because I was talking about the idea for
#FIRMM
with
@DrDamienFair
. So cool that NOUS Imaging now has a brilliant CEO - Ken Bruener - bringing camera-free real-time MRI QC to patient care.
In network space, the AMN is closest to the Salience network, which likely processes the salience/reward that shapes action, and the recently described SCAN (somato-cognitive action network) , which sits closer to the output end and implements planned
Outstanding summary in Nature Reviews Neuroscience by Jake Rogers
If you want to figure out if and/or how a therapy (drug/device/other) affects the human brain - this is a good way to do it quickly & cost-effectively (n = 7).
Lesion studies by
@foxmdphd
and
@realJoshSiegel
and others have mapped deficits of voluntary goal-directed action (e.g. abulia) to the AMN.
@josef_parvizi
and others have shown action-mode relevant state changes with stimulation/seizures that map to the AMN.
@WiringTheBrain
Why don't you like 'Mind-Body Interface' we worked on the name for a year with endless debates. It's my baby. The mind is what the brain does, the body is the muscles and the organs. The MBI is the set of physical nodes in the neuroaxis; a circuit that actualizes plans.
Over the last few years, several results made us realize that our prior conceptualization of the AMN as a purely executive/cognitive control network wasn’t spot on … for one, it was hard to square with the finding that a motor plasticity manipulation selectively changed
Naming the brain’s functional networks is hard. The stickiest, least contested network names (default-mode, salience) have been the ones that make an accurate functional ascription. So we mulled over the CON’s new name (AMN) for a long, long time. In choosing ‘AMN’ we
A bunch of things in neuroscience and medicine that didn't make sense to me before discovering the SCAN ... now seem a lot less strange. For example: The reported benefits of 'motor cortex' stimulation and exercise/acitivity for chronic pain.
Well, it finally happened. This
#ICE
International
#StudentBan
got me back into the “organizing effective lobbying action via twitter” game.
Want to make sure our lovely international students can keep their visas? Here’s how to do it:
THREAD.
1/14
Next came the discovery of the Somato-cognitive Action Network (SCAN) for arousal, holistic whole-body skeletal & smooth muscle and endocrine control in the service of implementing goal-directed behavior: . The SCAN is strongly functionally connected to the
@PessoaBrain
@ndosenbach
here is a similar figure that I created for the correlation b/w BMI and HbA1c in the NHANES dataset over 1000 sampling runs. population r=0.21, need n of ~400 before you have 95% confidence in being +/- .1 from true r. hard to imagine fMRI will ever be better than this...
There’s a competitive
#T32
for this position, that guarantees 3 yrs of funding, with great perks like travel & computer funds. If you don’t get the T32 there’s still generous, guaranteed funding.
@TuringMedical
's first BullsAI product: the next step towards making personalized precision-targeted neuromodulation for brain disorders available for downloading anywhere. Amazing effort by the greatest team including but not limited to
@Drdamienfair
@max_bertolero
Hans
@ndosenbach
and I founded
@TuringMedical
to turn two decades of advanced MRI collaboration and friendship into real-world impact. We handed the keys over to the incredible team
@TuringMedical
and they are bringing that vision to life. Huge congrats on reaching this milestone! 👇🏾
Amazing mouse work by
@DavidAMcCormick
and others has subdivided waking behavioral states into substates from quieter to more active, and revealed that movement, heightened arousal and behavioral performance are intertwined in the brain.
@KarlDeisseroth
revealed a global cortical
Some functional network motifs were shifted, sometimes by centimeters, with relative positions preserved. Together the findings suggest that absolute spatial localization of brain activity, the focus of group-averaged neuroimaging, may not be as important as previously assumed.
@arokem
@ChrisFiloG
@ndosenbach
It kind of amazes me that this hasn't happened yet - would be a ton of value in producing a big, complex, naturalistic stimulus set with all kinds of event timings included
Great work by
@NicoleSeider
showing the effects of
#diffusion
#MRI
data amount on results for BaMM, BedpostX, DSI-QBall, MRtrix3. Ouchies: some methods get overconfident w/ increasing data & errors increase, some only return their priors & angle estimation is a mess. BaMM wins!
I’m proud to announce we’ve published our paper “Accuracy and reliability of diffusion imaging models”! The models tested are sensitive to priors and vulnerability to overfitting. We need more data per subject and avoid overconfident prior-driven models.
I had a feeling that the great
@bealuna55
would give the best quotes.
@WUSTLmstp
co-production with
@dillannewbold
and strong support from Tim Laumann and Anish Mitra.
This push ‘Towards Reproducible Brain-Wide Association’ studies with
@DrDamienFair
: only made possible by the Fair lab’s ABCD-BIDS pipeline and ABCD-3165 data collection. If you want ABCD to jive, use 3165.
1. Study The Brain from Inside Out, as per
@brainrhythms
. Start with brain organization and see what we can derive about how the brain drives behavior. Do not look for the things we can see (fingers, legs) or the words we made up (motor, cognitive control) in the brain.
@PessoaBrain
@russpoldrack
@ajshackman
It seems like calcium imaging isn't so straightforward either. And in mice, there's the whole ... it worked in mice, but not in humans. I'm very bullish on MRI. It's 'just an engineering problem'. We need to bring down costs and measurement error and work together better.
The somato-cognitive action network (SCAN) helps transform our motivations/plans into actions. The SCAN cuts across many domains previously thought of as separate: motivation, planning, arousal, physiology, movement, control of internal organs and pain.