Our latest  work with MIT colleagues was featured in the Picower Institute's News Website. Read it here.

A new study adds evidence that electric fields in the brain help to organize and shape underlying neural activity via “ephaptic coupling.” You can access the MIT Picower news piece and paper here. Abstract below:

 The waxing and waning cortical oscillatory power correlates with function and disease. This cross-trial variability has been thought to be due to neuromodulation, uncertainty encoding, and/or changes in cortical excitability. Here, we report evidence that it is also due to fluctuations in ephaptic influences of mesoscale electric fields. We analyzed LFP data from the PFC recorded during a spatial delay saccade task. We constructed a model that describes the electric field close to the cortical patch given the neural activity that generates it. This revealed that field-to-neuron interactions (ephaptic coupling strength) were stronger than neuron-to-field, and it correlated trial-by-trial changes in oscillatory power. This suggests a form of circular causality where neural activity and extracellular electric fields continuously shape each other. These results further suggest that mesoscale ephaptic effects help drive the formation of memory ensembles, a prediction of the cytoelectric coupling hypothesis.

In a recent Biological Psychiatry paper with UCL and other colleagues, we found that reduced pyramidal cell excitability is present at baseline in clinical high risk for psychosis (CHR) converters, consistent with the hypothesis that hypofunction of pyramidal cells is a primary pathology in schizophrenia rather than a consequence of chronic illness. We also found that positive symptoms among CHR-converters may reflect compensatory downregulation of inhibition, see also here for a previous, free version of the same paper.

How does the brain processes information?
This remains one of neuroscience’s most fascinating challenges.
A growing body of research suggests that the cortex functions as a predictive system—constantly generating models to anticipate incoming input. But the story isn’t that simple.
Multiple frameworks—predictive coding, routing mechanisms, and autoencoder-based models—offer overlapping explanations. The question is: which one actually reflects what the brain is doing?

In our latest work, we tackled this directly by comparing these approaches using laminar LFP recordings across a cortical network during a visual search task.
What we found was striking:
🔹 No single model fully explains neural dynamics
🔹 Instead, the brain appears to use a hybrid strategy
🔹 Deep cortical layers align with Predictive Coding principles
🔹 Superficial layers reflect predictive routing—without explicit error computations
This points toward a more nuanced view of cortical function—where complementary mechanisms operate across layers and regions, combining top-down predictions with superficial-layer inhibition.
Rather than competing theories, these frameworks may each capture part of a larger, integrated explanation. -- work with Earl K. Miller, PhDAndre Bastos

New paper explains how the dynamics of neurons, fields and their interactions shift in depression.  This is  motivated by the cytoelectric coupling hypothesis, which suggests that efficient information processing results from mesoscale electric fields and that the re-emergence of depression symptoms might result from altered electric fields. The SCC model of depression proposes that the subgenual cingulate cortex (SCC) becomes overactive, disrupting the normal functioning of the limbic, subcortical, and prefrontal regions it connects with. Deep brain stimulation (DBS) appears to restore effective signal propagation in some patients, at least temporarily. Our work aims to understand why DBS is not universally effective and why relapse occurs. We suggest that relapse can be explained by examining whether the electrical fields generated by SCC neurons and the neurons themselves evolve in tandem over time. When they begin to diverge, this may signal a loss of coordinated control—much like an orchestra in which musicians either follow the conductor or gradually drift out of alignment.

Our work with Partha Ghose on quantum effects in the brain was featured among the 10 most-read stories in 2025 by the Institute of Physics. Read more here. 

Paper here. Abstract follows:

Autism Spectrum Disorder (ASD) is characterized by certain difficulties in emotion-related processing. Recent research using electroencephalography (EEG) to measure somatosensory evoked potentials during emotion perception has shown reduced embodiment of emotional expressions in autistic compared to neurotypical individuals, independently from differences in visual processing. However, the underlying neural dynamics are not clear. In this study, we use Dynamic Causal Modeling (DCM) on EEG data to investigate whether reduced embodiment during emotion processing in ASD individuals is caused by changes in intrinsic connectivity within the somatosensory cortex, or by top-down modulatory effects from higher-order frontal areas. We constructed a model involving the primary and secondary right somatosensory cortex, the right supplementary motor area and the right inferior frontal gyrus, and tested effective connectivity during emotion or gender discrimination tasks in two groups of ASD and typically developing (TD) participants (n = 38, male and female, 2 females). Our results reveal that task-related differences in electrocortical activity between the emotion and gender tasks are causally explained by changes in intrinsic activity within the right primary somatosensory cortex (rS1) in both TD and ASD. Importantly, these intrinsic changes in rS1 are significantly different between TD and ASD groups and individual task-related changes in rS1 significantly correlate with alexithymia traits. Our study provides novel evidence on the neural dynamics underlying difficulties in emotion processing in ASD individuals, highlighting that differential intrinsic activations of the rS1 are causally involved in such difficulties, and suggests that they are mediated by alexithymia.

It was very inspiring to listen to some of UK's brightest minds in theoretical and mathematical neuroscience for a full-day workshop at City St George’s, University of London in December — and what a lineup it was! The schedule was packed with cutting-edge talks, lively poster sessions, and rich discussions on how we model the brain in health and disease. 

Such a diverse range of topics — from theoretical principles of brain function and AI to computational methods grounded in data and neuroscience. Many thanks to Claudia Clopath Dan Goodman Thomas Nowotny Li Su Sean Froudist-Walsh Vassilis Cutsuridis Karl Friston Rick Adams Conor Houghton Peter Grindrod CBE Enrico Amico for sharing their ideas and work in such engaging ways! And many thanks to Coombes Stephen for helping make this happen!

The UK Theoretical Neuroscience Workhop will bring together theoretical and mathematical neuroscientists, as well as students from across the UK, to discuss current challenges and innovations in modeling brain activity in both health and disease. 

Register at: https://lnkd.in/gS9J55_F

We are delighted to invite you to the UK Theoretical Neuroscience Workshop, taking place on December 16, 2025, at City St George’s, University of London.

This one-day workshop in central London will bring together researchers from across the UK and beyond to share ideas, present cutting-edge work, and spark discussions on computational and theoretical neuroscience. Please find the schedule and details below.

Highlights:

An exciting lineup of invited talks covering diverse areas of theoretical and computational neuroscience.

A Poster Session, offering a platform to showcase your research and engage with peers.

A Poster Award, recognizing outstanding contributions and presentations.

Schedule:

09:05–09:40 — Claudia Clopath — Semantic extraction via systems consolidation

09:40–10:15 — Dan Goodman — TBA

10:15–10:50 — Thomas Nowotny — Harnessing the adjoint method for gradient descent in spiking neural networks

10:50–11:15 — Break & Posters

11:15–11:50 — Li Su — Digital twin models that talk the talk and walk the walk

11:50–12:25 — Sean Froudist-Walsh — TBA

12:25–13:00 — Vassilis Cutsuridis — Super memory retrieval in the hippocampus

13:00–14:00 — Lunch & Posters

14:00–14:35 — Karl Friston — Active inference and artificial curiosity

14:35–15:10 — Rick Adams — TBA

15:10–15:45 — Conor Houghton — TBA

15:45–16:10 — Break & Posters

16:10–16:45 — Peter Grindrod — The insights from neuroscience driving next generation neuromorphic AI and information processing

16:45–17:20 — Enrico Amico — Higher-order connectomics of human brain function

17:20–17:30 — Poster Award & Closing Remarks

Call for Posters (with Poster Award):

We invite poster submissions across theoretical and computational neuroscience. Present your work, spark new collaborations, and compete for the Poster Award.

Venue: City St George’s, University of London (Auditorium B200)

Date: Tuesday, December 16, 2025