Cian O'Donnell
@cianodonnell.bsky.social
Computational neuroscientist.
Senior Lecturer at Ulster University in the Great City of Derry, Northern Ireland.
"not articulate enough"
https://odonnellgroup.github.io
Senior Lecturer at Ulster University in the Great City of Derry, Northern Ireland.
"not articulate enough"
https://odonnellgroup.github.io
also applies to academia.
The principle I go by is to treat every person like a person. You never know who will be successful, nor do you know who will become a dear friend. So just be kind and engaged.
The only people who hate being treated as normal are nightmares. Not clicking with them is a secret bonus.
The only people who hate being treated as normal are nightmares. Not clicking with them is a secret bonus.
Talking to one of these types at a con who asked how I 'got in' with so many 'famous writers' who I'd been talking to and I said "we made friends before anyone knew who we were decades ago by striking up interesting conversations and enjoying each other's company" and he kept re-asking the question.
December 26, 2025 at 10:30 PM
also applies to academia.
Reposted by Cian O'Donnell
The principle I go by is to treat every person like a person. You never know who will be successful, nor do you know who will become a dear friend. So just be kind and engaged.
The only people who hate being treated as normal are nightmares. Not clicking with them is a secret bonus.
The only people who hate being treated as normal are nightmares. Not clicking with them is a secret bonus.
Talking to one of these types at a con who asked how I 'got in' with so many 'famous writers' who I'd been talking to and I said "we made friends before anyone knew who we were decades ago by striking up interesting conversations and enjoying each other's company" and he kept re-asking the question.
They have come to the convention hoping to "break in" and then find themselves -- shocker! -- a "nobody in the book world." Yeah no shit! We all are! I have gently told guys like this a million times that it takes most of us *years* of hard work but they do not believe that could be true for them. +
December 26, 2025 at 10:07 PM
The principle I go by is to treat every person like a person. You never know who will be successful, nor do you know who will become a dear friend. So just be kind and engaged.
The only people who hate being treated as normal are nightmares. Not clicking with them is a secret bonus.
The only people who hate being treated as normal are nightmares. Not clicking with them is a secret bonus.
Reposted by Cian O'Donnell
Friendly neighborhood biologist here. I see a lot of people are talking about biological sexes right now. Lots of folks make biological sex seem really simple. Well, since it’s so simple, let’s find the biological roots, shall we? Let’s talk about sex...[a thread]
(Reposted from the other site)
(Reposted from the other site)
August 9, 2024 at 10:34 PM
Friendly neighborhood biologist here. I see a lot of people are talking about biological sexes right now. Lots of folks make biological sex seem really simple. Well, since it’s so simple, let’s find the biological roots, shall we? Let’s talk about sex...[a thread]
(Reposted from the other site)
(Reposted from the other site)
Reposted by Cian O'Donnell
Mp3 players fucking rock we fucked up so bad putting everything on phone
December 25, 2025 at 7:34 PM
Mp3 players fucking rock we fucked up so bad putting everything on phone
Reposted by Cian O'Donnell
New Perspective from myself, Sarah Heilbronner and @myoo.bsky.social . “Rethinking the centrality of brain areas in understanding functional organization” in Nature Neuroscience. 🧵
rdcu.be/eVZ1A
rdcu.be/eVZ1A
Rethinking the centrality of brain areas in understanding functional organization
Nature Neuroscience - Parcellation of the cortex into functionally modular brain areas is foundational to neuroscience. Here, Hayden, Heilbronner and Yoo question the central status of brain areas...
rdcu.be
December 23, 2025 at 1:02 PM
New Perspective from myself, Sarah Heilbronner and @myoo.bsky.social . “Rethinking the centrality of brain areas in understanding functional organization” in Nature Neuroscience. 🧵
rdcu.be/eVZ1A
rdcu.be/eVZ1A
Reposted by Cian O'Donnell
Japan is joining Horizon Europe, the EU’s flagship research and innovation programme.
Openness and international cooperation can shape a bright future for science and technology.
Through science, we can build bridges, strengthen competitiveness, and accelerate the green and digital transitions.
Openness and international cooperation can shape a bright future for science and technology.
Through science, we can build bridges, strengthen competitiveness, and accelerate the green and digital transitions.
December 22, 2025 at 9:53 AM
Japan is joining Horizon Europe, the EU’s flagship research and innovation programme.
Openness and international cooperation can shape a bright future for science and technology.
Through science, we can build bridges, strengthen competitiveness, and accelerate the green and digital transitions.
Openness and international cooperation can shape a bright future for science and technology.
Through science, we can build bridges, strengthen competitiveness, and accelerate the green and digital transitions.
Reposted by Cian O'Donnell
Includes me exposing the myth that Cambridge somehow earned its way into its leadership position in UK life sciences. It didn't. Leeds was in the lead, and then the Medical Research Council in London picked Cambridge and London as winners because it was their mates.
December 12, 2025 at 1:53 AM
Includes me exposing the myth that Cambridge somehow earned its way into its leadership position in UK life sciences. It didn't. Leeds was in the lead, and then the Medical Research Council in London picked Cambridge and London as winners because it was their mates.
Reposted by Cian O'Donnell
I went to a dinner with the AI Minister and I read a paper on Lovelace Institue --- a proposal from Tony Blair's Institute on how we could fund breakthrough science better. And I wrote that all up into one incoherent mess for you to enjoy. tomforth.co.uk/nationalpurp...
National purpose on AI.
Lovelace Institutes, why I like the idea, and how I think they could work.
tomforth.co.uk
December 12, 2025 at 1:48 AM
I went to a dinner with the AI Minister and I read a paper on Lovelace Institue --- a proposal from Tony Blair's Institute on how we could fund breakthrough science better. And I wrote that all up into one incoherent mess for you to enjoy. tomforth.co.uk/nationalpurp...
Lovely time today at the 2nd All-Ireland Computational Neuroscience Symposium at Queens U Belfast. It's a small but growing community on the island. imo the talks were all genuinely great and I'm really excited for the future
Round 3 back in Dublin next August apparently if anyone wants to join! ☘️🧠
Round 3 back in Dublin next August apparently if anyone wants to join! ☘️🧠
December 19, 2025 at 5:08 PM
Lovely time today at the 2nd All-Ireland Computational Neuroscience Symposium at Queens U Belfast. It's a small but growing community on the island. imo the talks were all genuinely great and I'm really excited for the future
Round 3 back in Dublin next August apparently if anyone wants to join! ☘️🧠
Round 3 back in Dublin next August apparently if anyone wants to join! ☘️🧠
Reposted by Cian O'Donnell
I think it's also important to recognize that LLM "summaries" are not epistemically grounded in the ideas of the source material. When they're correct, it's more because *other* writing (in the training corpus) contains similar language. That makes it especially bad when applied to novel results.
December 17, 2025 at 6:18 AM
I think it's also important to recognize that LLM "summaries" are not epistemically grounded in the ideas of the source material. When they're correct, it's more because *other* writing (in the training corpus) contains similar language. That makes it especially bad when applied to novel results.
Reposted by Cian O'Donnell
Bichan Wu (@bichanw.bsky.social) & I wrote a tutorial paper on Reduced Rank Regression (RRR) — the statistical method underlying "communication subspaces" from Semedo et al 2019 — aimed at neuroscientists.
arxiv.org/abs/2512.12467
arxiv.org/abs/2512.12467
Reduced rank regression for neural communication: a tutorial for neuroscientists
Reduced rank regression (RRR) is a statistical method for finding a low-dimensional linear mapping between a set of high-dimensional inputs and outputs. In recent years, RRR has found numerous applica...
arxiv.org
December 17, 2025 at 2:06 AM
Bichan Wu (@bichanw.bsky.social) & I wrote a tutorial paper on Reduced Rank Regression (RRR) — the statistical method underlying "communication subspaces" from Semedo et al 2019 — aimed at neuroscientists.
arxiv.org/abs/2512.12467
arxiv.org/abs/2512.12467
Reposted by Cian O'Donnell
Fully-funded International Neuroscience Doctoral Programme🧠 Champalimaud Foundation, Lisbon, Portugal 🇵🇹
Deadline: Jan 31, 2026
fchampalimaud.org/champalimaud...
Research program spans systems/computational/theoretical/clinical/sensory/motor neuroscience, neuroethology, intelligence, and more!!
Deadline: Jan 31, 2026
fchampalimaud.org/champalimaud...
Research program spans systems/computational/theoretical/clinical/sensory/motor neuroscience, neuroethology, intelligence, and more!!
December 16, 2025 at 7:20 PM
Fully-funded International Neuroscience Doctoral Programme🧠 Champalimaud Foundation, Lisbon, Portugal 🇵🇹
Deadline: Jan 31, 2026
fchampalimaud.org/champalimaud...
Research program spans systems/computational/theoretical/clinical/sensory/motor neuroscience, neuroethology, intelligence, and more!!
Deadline: Jan 31, 2026
fchampalimaud.org/champalimaud...
Research program spans systems/computational/theoretical/clinical/sensory/motor neuroscience, neuroethology, intelligence, and more!!
Reposted by Cian O'Donnell
New paper for #neurips2025!
AI models adjust millions of internal settings to get better at a task. But how are these adjustments determined? For decades, we've mostly figured this out through trial & error.
We took a different approach...🧵 (1/6)
🔗 openreview.net/forum?id=oMi...
AI models adjust millions of internal settings to get better at a task. But how are these adjustments determined? For decades, we've mostly figured this out through trial & error.
We took a different approach...🧵 (1/6)
🔗 openreview.net/forum?id=oMi...
December 16, 2025 at 7:29 PM
New paper for #neurips2025!
AI models adjust millions of internal settings to get better at a task. But how are these adjustments determined? For decades, we've mostly figured this out through trial & error.
We took a different approach...🧵 (1/6)
🔗 openreview.net/forum?id=oMi...
AI models adjust millions of internal settings to get better at a task. But how are these adjustments determined? For decades, we've mostly figured this out through trial & error.
We took a different approach...🧵 (1/6)
🔗 openreview.net/forum?id=oMi...
Reposted by Cian O'Donnell
Would repost this a million times if I could!
It‘s the entitlement inherent to many of these complaints that makes them so hard to take seriously.
It‘s the entitlement inherent to many of these complaints that makes them so hard to take seriously.
this is an experience that I understand deeply, as I imagine many other people who are not white men can as well, and in the hands of a different writer it might serve as a foundation for solidarity rather than bitterness and implicit permission to participate in a reactionary political project
December 16, 2025 at 3:58 PM
Would repost this a million times if I could!
It‘s the entitlement inherent to many of these complaints that makes them so hard to take seriously.
It‘s the entitlement inherent to many of these complaints that makes them so hard to take seriously.
Reposted by Cian O'Donnell
“Basic neuroscience hasn’t produced new drugs.” 💊
Not true - zuranolone (PPD), suzetrigine (pain), gepants (migraine), and more... were born out of a long arc of studies in the lab.
I wrote a Perspective on why this matters. @thetransmitter.bsky.social
www.thetransmitter.org/drug-develop...
Not true - zuranolone (PPD), suzetrigine (pain), gepants (migraine), and more... were born out of a long arc of studies in the lab.
I wrote a Perspective on why this matters. @thetransmitter.bsky.social
www.thetransmitter.org/drug-develop...
How basic neuroscience has paved the path to new drugs
A growing list of medications—such as zuranolone for postpartum depression, suzetrigine for pain, and the gepants class of migraine medicines—exist because of insights from basic research.
www.thetransmitter.org
December 15, 2025 at 3:15 PM
“Basic neuroscience hasn’t produced new drugs.” 💊
Not true - zuranolone (PPD), suzetrigine (pain), gepants (migraine), and more... were born out of a long arc of studies in the lab.
I wrote a Perspective on why this matters. @thetransmitter.bsky.social
www.thetransmitter.org/drug-develop...
Not true - zuranolone (PPD), suzetrigine (pain), gepants (migraine), and more... were born out of a long arc of studies in the lab.
I wrote a Perspective on why this matters. @thetransmitter.bsky.social
www.thetransmitter.org/drug-develop...
Come for the great overview of dimensionality reduction, stay for the funky visualisations
I’ve been exploring a lot of ideas that are new to me recently and one that’s stuck in my brain is dimensionality reduction. How do we visualise points in hundred or thousand dimensional space?
This is a wonderful and approachable piece on the topic: colah.github.io/posts/2014-1...
This is a wonderful and approachable piece on the topic: colah.github.io/posts/2014-1...
December 16, 2025 at 11:27 AM
Come for the great overview of dimensionality reduction, stay for the funky visualisations
Reposted by Cian O'Donnell
🧠🪰 The adult Drosophila brain connectome now gives us a complete wiring diagram of ~140k neurons. But a wiring diagram alone isn’t understanding.
How is this massive network organized?
Our paper tackles that question by mapping community structure across the entire fly brain. 1/
How is this massive network organized?
Our paper tackles that question by mapping community structure across the entire fly brain. 1/
New lab paper - will say more about this in a little while
December 15, 2025 at 3:11 PM
🧠🪰 The adult Drosophila brain connectome now gives us a complete wiring diagram of ~140k neurons. But a wiring diagram alone isn’t understanding.
How is this massive network organized?
Our paper tackles that question by mapping community structure across the entire fly brain. 1/
How is this massive network organized?
Our paper tackles that question by mapping community structure across the entire fly brain. 1/
I still don't get this. The journal is asking for my opinion, not GenAI's. We don't even get any public credit for doing reviews... it's as close to a thankless task as you can get in academic research. So if someone don't have time to do a review solo, then why not just decline?
53% of researchers say they have used AI when peer reviewing articles, according to a survey by Frontiers. www.nature.com/articles/d41...
More than half of researchers now use AI for peer review — often against guidance
Policies should reflect the ‘new reality’ of researchers’ increasing reliance on tools that can summarize manuscripts and draft reports.
www.nature.com
December 15, 2025 at 10:41 PM
I still don't get this. The journal is asking for my opinion, not GenAI's. We don't even get any public credit for doing reviews... it's as close to a thankless task as you can get in academic research. So if someone don't have time to do a review solo, then why not just decline?
Reposted by Cian O'Donnell
Merriam-Webster’s human editors have chosen ‘slop’ as the 2025 Word of the Year.
December 15, 2025 at 2:07 PM
Merriam-Webster’s human editors have chosen ‘slop’ as the 2025 Word of the Year.
Reposted by Cian O'Donnell
"The board is implied. If you show the board, you limit the imagination. If you don’t show the board, anyone could be playing Mastermind at any time. In a limo. In a bunker. On a yacht while plotting the downfall of a minor principality."
The Mastermind Box Cover: What the Hell Were They Thinking?
- - -INVICTA GAMES, LTD. Packaging Team — Official Minutes Project: Mastermind / New Cover Presentation MARTIN SMITH (Marketing Senior Vice Presi...
buff.ly
December 14, 2025 at 2:00 AM
"The board is implied. If you show the board, you limit the imagination. If you don’t show the board, anyone could be playing Mastermind at any time. In a limo. In a bunker. On a yacht while plotting the downfall of a minor principality."
Reposted by Cian O'Donnell
The ZSL Scientific Medal is also awarded to Professor Stephen Montgomery, @bristolbiosci.bsky.social whose research has provided insights into the relationships between ecological niche specialisation, brain structure, and cognitive function 🏅 @ebablab.bsky.social
December 10, 2025 at 4:28 PM
The ZSL Scientific Medal is also awarded to Professor Stephen Montgomery, @bristolbiosci.bsky.social whose research has provided insights into the relationships between ecological niche specialisation, brain structure, and cognitive function 🏅 @ebablab.bsky.social
Reposted by Cian O'Donnell
GWAS has been an incredible discovery tool for human genetics: it regularly identifies *causal* links from 1000s of SNPs to any given trait. But mechanistic interpretation is usually difficult.
Our latest work on causal models for this is out yesterday:
www.nature.com/articles/s41...
A short🧵:
Our latest work on causal models for this is out yesterday:
www.nature.com/articles/s41...
A short🧵:
Causal modelling of gene effects from regulators to programs to traits - Nature
Approaches combining genetic association and Perturb-seq data that link genetic variants to functional programs to traits are described.
www.nature.com
December 11, 2025 at 5:54 PM
GWAS has been an incredible discovery tool for human genetics: it regularly identifies *causal* links from 1000s of SNPs to any given trait. But mechanistic interpretation is usually difficult.
Our latest work on causal models for this is out yesterday:
www.nature.com/articles/s41...
A short🧵:
Our latest work on causal models for this is out yesterday:
www.nature.com/articles/s41...
A short🧵:
Reposted by Cian O'Donnell
Artificial hibernation uncovers distinct synaptic engram architecture for memory retention https://www.biorxiv.org/content/10.64898/2025.12.09.692927v1
December 11, 2025 at 8:15 PM
Artificial hibernation uncovers distinct synaptic engram architecture for memory retention https://www.biorxiv.org/content/10.64898/2025.12.09.692927v1
Reposted by Cian O'Donnell
The point is that most papers we all write are incremental -- based on well established methods and ideas -- and have an audience of subfield experts. These are judged by expert colleagues when they read the work. These don't, in my view, need traditional peer review.
December 11, 2025 at 7:26 PM
The point is that most papers we all write are incremental -- based on well established methods and ideas -- and have an audience of subfield experts. These are judged by expert colleagues when they read the work. These don't, in my view, need traditional peer review.
Reposted by Cian O'Donnell
Peer review is important and useful but we should focus our efforts on a small number of papers that matter (making big claims, using new approaches) and let the vast majority of work live on a preprint server to be judged by their utility over time to domain experts.
Springer-Nature statement
“Whilst the details of peer review are confidential, we can confirm that the article underwent two rounds of review from two independent peer reviewers, supporting an accept decision.”
How am I now expected to believe that two people looked at the paper twice and DGAF?
“Whilst the details of peer review are confidential, we can confirm that the article underwent two rounds of review from two independent peer reviewers, supporting an accept decision.”
How am I now expected to believe that two people looked at the paper twice and DGAF?
Riding the Autism Bicycle to Retraction Town
Does anyone *really* know their Factor Fexcectorn?
nobreakthroughs.substack.com
December 11, 2025 at 7:16 PM
Peer review is important and useful but we should focus our efforts on a small number of papers that matter (making big claims, using new approaches) and let the vast majority of work live on a preprint server to be judged by their utility over time to domain experts.