Guido Meijer

doi.orgSerotonin drives choice-independent reconfiguration of distributed neural activitySerotonin (5-HT) is a central neuromodulator which is implicated in, amongst other functions, cognitive flexibility. 5-HT is released from the dorsal raphe nucleus (DRN) throughout nearly the entire f...

5 Aug 2025

🚨Pre-print alert🚨 We stimulated serotonin with optogenetics while doing large-scale Neuropixel recordings across the mouse brain. We found strong widespread modulation of neural activity, but no effect on the choices of the mouse 🐭 How is this possible? Strap in! (1/9) 👇🧵 doi.org/10.1101/2025...

Tonight Europe is going to wintertime, one hour of sleep more for everybody! Unless you have young children, then you get one hour of sleep less. Lovely.

thetransmitter.orgHow about adding an element of chance to grant review?A process that uses a lottery system saves time and accepts that, above a quality threshold, luck has always played a part in science funding.

Distributing grants through a lottery has its advantages but I'm afraid it will result in even more AI generated grant applications. A lottery is a good idea but it has to be taken one step further to really work: no more grant proposals at all anymore. www.thetransmitter.org/funding/gran...

16 Sept

Let me tell you, teaching a course to first year undergrads is a sure-fire way of making you feel really, really old.

5 Aug

It was great to see the RIKEN institute in Japan and give a talk there. Thanks to Muhang Li and @tommchugh.bsky.social for inviting me!

31 July

If you're attending the NEURO2026 conference in Kobe, Japan ⛩️ don't miss our symposium on Sunday morning about the latest from the IBL @intlbrainlab.bsky.social. My talk on how neuromodulators shape decision-making starts at 09:55 🧠

biorxiv.orgSerotonin induces DOWN states across the anesthetized mouse forebrainSerotonergic neurons in the dorsal raphe nucleus (DRN) project extensively throughout the forebrain, yet their influence on global network states remains controversial. Serotonin is implicated in regu...

21 July

I am happy to announce my very first senior author paper! Serotonin is known to abolish UP-DOWN state dynamics, which occur during sleep and anesthesia. There are several outstanding questions: What happens at short latencies? How widespread across the brain is it? 🧠 www.biorxiv.org/content/10.6...

Computational modelling revealed that the cortical DOWN state is global. The visual cortex does not receive direct dorsal raphe input, but is still transitioned into a DOWN state. The 5-HT induced DOWN state is widespead enough that the visual cortex is pulled into it by the rest of the brain.

21 July

As always all the data are openly available through ONE (int-brain-lab.github.io/ONE/). This work was spearheaded by Rosa Großmann (@rosacg.bsky.social) with help from Jorge Meijas (@jorgefmejias.bsky.social), Zach Mainen (not on bluesky), and the IBL collaboration (@intlbrainlab.bsky.social).

To answer these questions we did large-scale Neuropixel recordings across the anesthetized mouse forebrain while optogenetically stimulating dorsal raphe serotonergic neurons. We found that spiking activity across the brain was suppressed by serotonin stimulation.

21 July

Looking at the spike rasters across the entire probe we saw that 5-HT stimulation resulted in a DOWN state. To quantify this effect we fit a two-state HMM and found 5-HT induced DOWN states everywhere except the midbrain.

16 July

Now that my daughter can say a few words I feel like being a servant to a neanderthaler. I'm constantly being shouted at "food!", "give!", "yes good, more more!"

10 July

I wonder which layer of hell Dante reserved for the developers of the FENS app

7 July

Thanks to all who visited my poster for your interest and lively discussions 😃

6 July

I'm on my way to Barcelona for #FENS2026! 🌞🧠Come see my poster tomorrow morning: PS01-07AM-292 - DYNAMIC UPDATING OF REWARD PREDICTIONS BY THE PARA-HIPPOCAMPAL NETWORK I did simultaneous Neuropixel recordings in CA1 and 5 cortical regions to find out how the brain switches between cognitive maps 🗺️

Guido Meijer reposted

Carl Zimmer

@carlzimmer.com

1 July

Meet SpudCell, a synthetic cell made from lifeless ingredients that feeds, grows, divides, and experiences selection. If it doesn't have all the hallmarks of life, it has a lot! Here's my story. Gift link: nyti.ms/4vCbTih

28 June

Idea for an app!💡 World Cup football matches are often in the middle of the night for us Europeans; the Netherlands is playing at 3AM 💤 I want an app that wakes me at 5:30AM to see the penalty shoot outs, but only if there are any. Can someone who doesn't have a 2 year old develop this? Thanks! 🙏🏻

30 June

Turned out I didn't need an app and was up at 5:30 anyway due to said toddler 🙃

12 June

I wonder how many unnecessary figure panels have been made just to make the entire figure nice and rectangular.

science.orgReward magnitude determines reinforcement learning efficiencyStandard animal learning studies minimize individual reward magnitudes to maximize the repetitions of reinforced behaviors. We investigated how reward magnitude influences initial learning across five...

9 June

Want your mice to learn your task faster? 🐭 According to this paper giving a huge reward (10x typical size) in a subset of trials strongly shortens training times: www.science.org/doi/10.1126/...

Guido Meijer reposted

Kenneth Harris

@kenneth-harris.bsky.social

21 May

Introducing the International Brain Lab AI Agent: a tool to help researchers analyze neural activity from across the mouse brain, using AI coding agents. Please try it! We would love your feedback. github.com/int-brain-la...

15 May

zetapy v4.0 is live! Due to various optimizations of the code it now runs 4x faster⚡Upgrade now and get your significant neurons in no time! pip install zetapy==4.0

atlas.internationalbrainlab.orgIBL Brain Atlas

7 May

Are you interested in serotonin receptors? I made a website which visualizes the expression of different 5-HT receptors! Data from the Allen Brain Atlas (@alleninstitute.org) website by the IBL Brain Viewer (@intlbrainlab.bsky.social). atlas.internationalbrainlab.org?buckets=meij... Instructions 👇

7 May

Select a receptor subtype in the dropdown menu on the left ("Select a feature"). You can scroll through the views with the sliders underneath. You can also scale the colormap by going to "Control Panel" on the top right and changing the sliders sliders underneath "colormap range".

I've been watching Fireman Sam with my daughter (the new animated version) and I must say I'm deeply worried for the residents of Pontypandy, the small Welch village with a population of apparently 10 people. A thread 🧵

The residents of Pontypandy make terrible decisions leading to a high incidence of forest fires and the likes. My fear is that the entire budget for fire and rescue services of Wales is being spend on saving the ten people who live their over and over.

What can we do? The current approach is to accept that these people need constant rescuing. This policy needs to shift to a more preemptive approach, why do these people make decisions like a five year old? How can we help them become better at this? Any suggestions are welcome.

The fire and rescue services of Pontypandy are extremely advanced. Besides regular fire engines they have two helicopters, a tiltrotor, a hovercraft, drones, and several boats. Who is paying for all this? The tax burden for these poor people must be astronomical to afford all this kit.

Then there is the issue of mass surveillance around the town. When something goes wrong, an AI system knows exactly what is happening and where. Even when nobody calls them. This is only possible through ubiquitous camera coverage with facial recognition software, these people have no privacy!

thetransmitter.orgERC backtracks on stricter grant resubmission rulesThe swift reversal came after more than 1,000 scientists signed an open letter protesting the stricter rules last week.

I'm very happy that fairness and reason has prevailed and the ERC has backtracked on this. Thanks to everyone who signed the open letter to the ERC for making this happen 🙏🏻 www.thetransmitter.org/policy/europ...

Guido Meijer@guidomeijer.com

The ERC is changing the rules retroactively! A "B" grade now carries a 2-year timeout instead of 1, even for those who submitted under the old rules. Researchers (incl me) make massive career decisions based on these timelines, you can't change them after the fact 😵‍💫 erc.europa.eu/news-events/...

erc.europa.euERC President explains stricter application measures amid rising demand for fundingThe text of an open letter sent by ERC President Maria Leptin to ERC panel members, grantees and other stakeholders on 16 April 2026.

Guido Meijer reposted

Pierre Le Merre

@pierrelemerre.bsky.social

1 May

So I (well Claude mostly) made a @julialang.org port of the ZETA test originally developped by Jorrit Montijn (published here: elifesciences.org/articles/71969). The port is faithful to zetapy (Python version by @guidomeijer.com and @alexanderheimel.bsky.social). Enjoy github.com/PierreLeMerr...

github.comGitHub - PierreLeMerre/ZetaJuContribute to PierreLeMerre/ZetaJu development by creating an account on GitHub.

Guido Meijer reposted

Fabian Sinz

@sinzlab.bsky.social

24 Apr

🧠Introducing OmniMouse: One of the largest datasets in neuroscience along with a systematic study of scaling properties of brain models Co-led by 🤩 K. Willeke, @pollytur.bsky.social and @alexrgil14.bsky.social Models trained on up to 3M neurons, >150B tokens

Guido Meijer reposted

Kate Jeffery

@katejj.bsky.social

21 Apr

Job alert! We are seeking a rodent in vivo electrophysiologist to work on coordination between anterior thalamus and hippocampus in rats using Neuropixels. Beautiful Scotland, leading university (Glasgow) and vibrant intellectual setting. Post is 23 months but hopefully extensible. Details here 👇

Guido Meijer reposted

PCI Neuroscience

@neuro.peercommunityin.org

18 Apr

Do you record spikes in vivo? Here are two recently recommended preprints on toolboxes for extracellular electrophysiology: -MLIB doi.org/10.1101/2025... by Maik Stüttgen. -Power Pixels doi.org/10.1101/2025... by @guidomeijer.com and @battaglialab.bsky.social.

The Power Pixels ⚡ pipeline is out now in Peer Community Journal! Are you doing Neuropixel recordings but don't have a pre-processing pipeline yet? Check out this end-to-end pipeline designed to simplify preprocessing of Neuropixel recordings 🧠 peercommunityjournal.org/articles/10....

peercommunityjournal.org Power Pixels: a turnkey pipeline for processing of Neuropixel recordings

Guido Meijer reposted

Pierre Le Merre

@pierrelemerre.bsky.social

20 Jan

Main postdoc study out! We can redefine prefrontal cortex regions with single-unit activity! Grateful to @carlenlab.bsky.social and @weltgeischt.bsky.social who made this crazy project real. Thanks to all co-authors, collaborators, and reviewers. www.nature.com/articles/s41...

nature.comA prefrontal cortex map based on single-neuron activity - Nature NeuroscienceThe authors mapped spontaneous and choice activity across mouse prefrontal cortex. The activity maps aligned with intrinsic connectivity rather than anatomical subregions, suggesting that connectivity...

Guido Meijer reposted

Valentin Riedl

@vavatin.bsky.social

16 Dec 2025

fMRI signals “up,” but neural metabolism might be going “down.” In our @natneuro.nature.com paper, we demonstrate that about 40% of voxels with robust BOLD responses exhibit opposite oxygen metabolism, revealing two distinct hemodynamic modes. rdcu.be/eUPO8 funds @erc.europa.eu #neuroskyence 🧵:

11 Nov 2025

The Power Pixels pipeline just keeps getting better! The latest update comes with: - OpenEphys support - AP_Histology support - automatic high-frequency noise reduction - NWB format export option - zarr compression of raw data And, most importantly: a cool new logo! github.com/NeuroNetMem/...

31 Oct 2025

Halloween isn’t my thing. My idea of a scary night is reading the plots of horror movies on Wikipedia.

25 Oct 2025

I never understood why everyone always complains about the clocks changing. Then I had a kid.

Want to make publication-ready figures come straight from Python without having to do any manual editing? Are you fed up with axes labels being unreadable during your presentations? Follow this short tutorial including code examples! 👇🧵

The workflow: Python plots single panels, saved as pdfs. Create an empty figure of 7 inch (180 mm) wide in Illustrator and place the panels as links. Add the panel letters (a,b,c). Now, if a new version of a panel is generated you just have to click update links in Illustrator and it's there. 🤩

colab.research.google.comGoogle Colab

I hope you found this useful! Here is a Google Colab with example code of what was discussed in this thread: colab.research.google.com/drive/1UlCET...

Full width figures typically have four panels next to each other, a good rule of thumb is that a single panel is 1.75 by 1.75 inch. Create two figure panels like this: f, axs = plt.subplots(1, 2, figsize=(1.75 * 2, 1.75), dpi=300) ⚠️ If it's too small increase the dpi, not the figsize!

A note about colors 🌈 Imagine you made eight figures and your PI asks you to change the color of an element you've consistently plotted green into orange. If you define your colors in the plotting function you only have to change it there and rerun the plotting scripts, this save a lot of time!

Journals have criteria figures have to adhere to. Generally, a full-width figure should be 7 inch wide and font sizes cannot be smaller than 7 pts. If you get into the habit of always creating figures that fit these criteria this will save you a lot of time when submitting.

The font size of each plot element can be easily set with seaborn. Write a function like this that sets the font size and any other style element you like to have in all your plots and the function each time before plotting (link to the code in the last post).

8 Oct 2025

I was doing the second-to-last recording of my project and was happily thinking to myself that I haven't broken a single Neuropixel in the entire two year project. You can guess what happened next... Now I have to install a new probe just for one final recording 🙃