Peter Zeidman

Many congrats Nadine and Steve! This looks very elegant indeed!

Thanks so much Rimona! Really nice you were there!

And huge thanks to James Bonaiuto of @danclab.bsky.social - their fantastic laminar MEG toolbox laMEG paved the way to SPM-Python! github.com/danclab/laMEG

The beta release of SPM-Python is out now! Amazing work by Johan Medrano @johmedr.bsky.social , Yael Balbastre, Yulia Bezsudnova @ybezs.bsky.social and other members of their team. A new era for SPM! #OHBM2025

Don't miss our upcoming #SPM course for #EEG & #MEG online from 19-22 May 2025! 🧠

We'll cover preprocessing, source reconstruction, DCM & OPM analyses. Practicals will include SPM-Python!

Programme: www.fil.ion.ucl.ac.uk/spm/docs/cou...

Register by May 11th 👇 onlinestore.ucl.ac.uk/conferences-...

🎺🎺🎺 We will be hosting #MEGUKI2025 this July and hope that lots of you will be able to join us, including for the kick off day focusing on #naturalistic neuroscience.

More details and info on how to submit abstracts below 👇

Detecting early neurovascular dysfunction in Alzheimer’s Disease and Cerebral Amyloid Angiopathy (CAA) at University College London on FindAPhD.com PhD Project - Detecting early neurovascular dysfunction in Alzheimer’s Disease and Cerebral Amyloid Angiopathy (CAA) at University College London, listed on FindAPhD.com

Excited to share a new PhD studentship, using fMRI and MEG to detect early changes in Alzheimer's disease, fully funded for a UK student by @alzheimersresearchuk.org . Situated at UCL in Queen Square, London. Join us! tinyurl.com/nvmgroupphd

Don't miss out on securing your spot for our upcoming #SPM course for #fMRI and #MRI/#VBM!

Join us in London 23-25 April! Reserve your place now! 🧠⬇️

onlinestore.ucl.ac.uk/conferences-...

Brilliant work from the FIL OPM-MEG team!

UCL – University College London UCL is consistently ranked as one of the top ten universities in the world (QS World University Rankings 2010-2022) and is No.2 in the UK for research power (Research Excellence Framework 2021).

We have an opening for a 12 month Research Assistant post, funded by an ERC grant. The postholder would help with several neuroimaging projects, including MEG and 7T fMRI. Please repost and share with anyone who may be interested. www.ucl.ac.uk/work-at-ucl/... #neuroskyence #Neuroimaging

Brilliant! haha not at all, it's a badge of honour! 🤓🔬

Sounds like we have a winner!

Coming soon!

Can we find the earliest adopters of SPM on Bluesky? Full marks to @jonsimons.bsky.social for SPM96!

Happy 30th birthday year to SPM!! 🎈🍰

To get the party started, we are delighted to announce SPM 25.01, a major new version of the open-source neuroimaging software.

Read the preprint to discover what's new arxiv.org/pdf/2501.12081 and download SPM for free from Github github.com/spm/spm/rele...

I am so grateful to Eleanor Maguire for being such an amazing teacher and mentor. We have set up a web page for sharing memories of Eleanor, via the link below.

Moreover, a unique contribution of fMRI is haemodynamics, and I think DCM for fMRI will be very important for laminar fMRI, where modeling laminar vasculature is needed in order to deconvolve the superficial bias. See the beautiful work in this paper as an example doi.org/10.1016/j.ne... [5/5]

So if you want to relax those strict assumptions made by the GLM, you might want DCM for fMRI. E.g. you could test the hypothesis that an interaction in one brain region can be explained by afferents from main effects expressed in two other brain regions. [4/5]

So does DCM for fMRI still have a role? I think so. First, note that standard mass-univariate fMRI GLM analysis is a special case of the DCM for fMRI model, where interactions among brain regions are ignored, and haemodynamics are (typically) assumed to be of a fixed form across brain. [3/5]

If your question is about the directionality of neural interactions, I totally agree DCM for M/EEG data would be much better than fMRI. E.g. the oscillatory frequency gives a clue as to the direction of laminar communication. [2/5]

DCM is just a framework for estimating the statistical evidence for state-space models, which are fitted to time series data using Bayesian methods. Each model is a hypothesis. So the question is, what's the right data and models for your question. [1/5]

UCL – University College London UCL is consistently ranked as one of the top ten universities in the world (QS World University Rankings 2010-2022) and is No.2 in the UK for research power (Research Excellence Framework 2021).

🌟 Rare Opportunity Alert 🌟

The Department of Imaging Neuroscience has a lectureship position available! 🎺

Lead a research programme in Cognition and Perception, and shape the future of the Institute's new BSc in Human Neuroscience. 🧠

#AcademicJobs #Neuroscience #UCL

www.ucl.ac.uk/work-at-ucl/...

Sympathies! It's so helpful to have such specific feedback...

Out now, our new paper evaluating the suitability of common haemodynamic models for use with older participants' fMRI scans. Always a pleasure to collaborate with Rik Henson and the Cam-CAN team!

Welcome Visit the post for more.

I am recruiting a postdoc at imaginerealitylab.org! Are you interested in mental imagery and reality monitoring and do you have a background in cognitive neuroscience? Check out the ad at: www.ucl.ac.uk/work-at-ucl/...

Don't miss out on securing your spot for our upcoming
#SPM course for #fMRI and #MRI / #VBM ! Join us online 14-16 October! Reserve your place now! onlinestore.ucl.ac.uk/conferences-...

Detail of the poster abstract. (a) Picture of a neural mass model with 2 interlinked brain regions, (b) Formulae for a design matrix X that scales connection strengths beta over time. (c) Illustration of a design matrix using that uses DCT basis functions to capture slow modulation of connection strengths. Medrano et al. 2024

Introducing a new method by Johan Medrano from the FIL Methods Group, for inferring gradual changes in synaptic gain, using M/EEG data and DCM. Find out more at #OHBM2024 poster 1935 on Weds/Thurs with @johmedr.bsky.social . ww6.aievolution.com/hbm2401/inde...

The new Google Scholar PDF reader plugin for Chrome seems very good indeed. You can click a link to a reference and then press back to return to the text 🤯 Hope it lasts.

In principle, you could use a model that has the amount of spatial sparsity parameterised within it, to explicitly test punctate vs distributed hypotheses. Multivariate Bayes (MVB) is one such (decoding) model in SPM. Not so commonly used. See www.jiscmail.ac.uk/cgi-bin/weba...

A personal bugbear: reporting evidence for effect A, and no evidence for effect B, then claiming there's a difference between A and B, without having compared them