Helen C Barron
@HelenCBarron
Followers
1K
Following
849
Media
23
Statuses
289
Joined March 2016
Congratulations Xenia!! So looking forward to working on this exciting project with you
Many congratulations to Unit postdoctoral researcher Dr Xenia Grande on being awarded a Walter Benjamin Postdoctoral Fellowship by the German Research Foundation (DFG)! https://t.co/Lmk3aQ2YZV
@NDCNOxford @UniofOxford @OxfordMedSci @The_MRC
0
0
3
Here's the story in more detail:
Our paper "Memory reactivation generates new, adaptive behaviours that reach beyond direct experience" is now out https://t.co/oiTC4HH2ks ! With @AnnieBRawson, Jill O’Reilly and @HelenCBarron at @MRCBNDU @NDCNOxford 1/6
0
0
3
Congratulations to @AnnieBRawson @SumedhaNalluru @MRCBNDU @NDCNOxford and Jill O'Reilly for new work showing that TMR during rest improves inference!
New Unit paper shows that memory reactivation generates new, adaptive behaviours that reach beyond direct experience, published in @Nature Scientific Reports! https://t.co/oe55oj7Noz
@NDCNOxford @UniofOxford
0
4
12
In an incredible display of commitment to sustainable research, seven @MRCBNDU laboratories have achieved @LEAFinLabs accreditations for their efficient practices. https://t.co/eSzQ6GstXt Thank you to everyone involved for driving positive changes within our labs! @NDCNOxford
0
3
5
Thanks for organizing Ilka- really enjoyed the excellent talks and invigorating discussion!
Very excited to chair the symposium about internal world models in humans, animals, and AI at #FENS2024 at 9:45AM in Hall A. A fantastic line up of speakers is waiting for us: @HelenCBarron, Kenji Doya, @ThomasBrox, and Alla Karpova. @BrainworldsFR
0
0
1
Check out Prakriti’s amazing work here!
My first ever conference poster is up at #FENS2024 today! Work on “Codependent plasticity underlies cortical memory strengthening” in session 02 (PS02-26PM-167), done with Annalise Rawson, @ejagnes, @douglasftome, @HelenCBarron & @TPVogels . Codependency is good sometimes :)
0
1
4
Valuable PPI session - thanks to patients from @OxfordHealthNHS ! and to Valentina Mancini and @GlitzLeonie for organizing
Huge thank you to psychosis patients from Oxford Health NHS FT @OxfordHealthNHS for reviewing our ethics documents and providing critical feedback on the design of our upcoming study with Valentina Mancini, Helen Barron, and Charlotte (Charlie) Stagg @NDCNOxford @UniofOxford
0
0
1
Thanks for having me- great to take part in this vibrant meeting!
Never thought I'd ever recommend chairing a conference, but here I do: https://t.co/zEj4A4n8jl was amazing and smooth thx to excellent organizers @CECOG3 & inspiring and engaging speakers @Franklandlab @duffalab @m_schoenauer @Zoe_Chi_Ngo @HelenCBarron in a relaxed setting! (1/2)
0
0
1
At #NeurIPS2023 now! I’ll be presenting our work on sequence memory with predictive coding ( https://t.co/nXyZb9HqH6) Tuesday morning, and a work led by Tianjin Li on recognition memory at the AMHN workshop on Friday ( https://t.co/Z3utKGrHG3)
1
1
11
I’m recruiting for a post-doc! I’m looking for someone to drive the creation of new acquisition or reconstruction techniques for functional magnetic resonance spectroscopic imaging (fMRSI). Previous experience of spectroscopy is not a requirement.
1
27
22
Congratulations @mufeng_tang !!
Will bring this work to @NeurIPSConf in New Orleans this December 😀 Camera-ready version (with great feedback and suggestions from the reviewers) will soon be out on arxiv
0
0
1
I am hiring a PhD student in cognitive computational #neuroscience! 👩🏻🎓👨🏻🎓Come and help us discover how the human brain learns, decides and guides flexible behaviour in beautiful Würzburg 🧠🙌 Find out more & apply here: https://t.co/peB7eeShaQ
@bioDGPs_DGPA @FENSorg @SfNtweets
7
169
272
There is still time to get your registration in for this symposium. If you have an interest in measuring neurotransmitter and neuro-metabolic dynamics in response to brain activity, come along and join a welcoming and growing community. https://t.co/KnhI5pbK4F
0
6
9
I’m recruiting for a post-doc! At Oxford, we’re trying to develop new ways to measure functional neurochemistry across the whole brain. I’m looking for someone to create new acquisition or reconstruction techniques for functional magnetic resonance spectroscopic imaging (fMRSI).
1
53
89
Congratulations again to @mufeng_tang for all the brilliant work, to co-supervisor Rafal Bogacz, and thanks to @MRCBNDU @NDCNOxford @OxNeuro
0
0
1
To conclude, in this work we propose a plausible model of sequential memory in the brain, which presents many interesting computational (implicit statistical whitening) and biologically relevant properties (predictive processing + contextual representation).
1
0
1
And, once we have two layers in place, the model develops latent representations of contextual information, a property also observed in the hippocampus. This helps tPC disentangle identical memories repeating in a sequence by representing *time* e.g. “3-2-1-2-3”:
1
0
1
But this is not the end of the story. Like the classical predictive coding models, we can flexibly extend tPC to a multi-layer model while retaining the plausible neural implementation:
1
0
2
The normalization by whitening also makes tPC robust against sequential inputs with varying feature (e.g. pixel) values, a case where the family of AHNs easily fail:
1
0
1
How is this an interesting finding? Well we know that statistical whitening decorrelates and normalizes along the feature dimension. Practically this makes tPC particularly robust against correlated patterns in a sequence:
1
0
1