This UKRI-funded Future Leaders Fellowship project (2022-26) is based on the idea that in psychotic disorders there is an imbalance between levels of excitation and inhibition in the brain. It may be that different kinds of psychosis have different 'imbalances', e.g., not enough excitation, or not enough inhibition. If so, then we should be trying to use different treatments in these different groups, instead of giving everyone the same drug (boosting inhibition, for example). We are trying to find a way of using a computational model of EEG data to work out whether someone has too much inhibition relative to excitation, or vice versa. If we can do this, we will try to use this measure to predict the outcome of giving an excitation-boosting drug to people with psychosis.

This project (funded by the UCL Institute of Mental Health, 2020-24) looks at associative inference in psychosis. An example of associative inference is that if you learn that A pairs with B, and B pairs with C, then you might also infer that A and C go together. A brain region called the hippocampus is known to be important for these inferences. It also interacts with prefrontal cortex during this process. People with psychosis can make abnormal inferences sometimes (e.g. thinking A and D go together, even if no connection between them has been experienced), and hippocampus and prefrontal cortex are known to have some neural abnormalities in psychosis too. A previous study we did found that abnormal hippocampal-prefrontal coupling affects spatial memory in psychosis. This project is trying to determine whether this abnormal coupling also affects non-spatial inferences in psychosis too.

This Wellcome Mental Health Award project (2025-30) is a collaboration between me, Dr Mohamady El-Gaby and Dr Matt Nour (Oxford University) and Dr Maria Eckstein (UCL/Deepmind). Mohamady recently found evidence for how prefrontal cortex plans steps towards a goal. We want to see if this planning algorithm is disrupted in people with psychosis, in genetic mouse models of psychosis, and in neural networks that have perturbations similar to those found in brain circuits in psychosis. We will soon be recruiting post docs and RAs to work on this project!