Dr. Omer Yuval

Research interests

The neurons that control living organisms are incredibly complex machines. Despite popular analogies to artificial neural networks, biological neurons show a great variety of dynamics across species, throughout the network, as well as across time-scales and contexts within the same neuron. To name a few, spiking, graded responses, multi-stability, desensitisation, synchronisation, rebound and phase-resetting. However, the importance of these mechanisms for information encoding and animal behaviour remains unclear.

Biological neural networks were fine tuned and optimised through an evolutionary process within a wide range of temporal and spatial scales (e.g., subcellular, cellular and network level) for controlling a physical body that operates within a physical environment. Hence, neurons do not function in isolation, and their dynamics are modulated by a variety of sensory stimuli, such as chemosensation and proprioception.

While increasingly larger connectomes are being mapped, single-neuron- and network-dynamics are still largely missing. Understanding the importance of these dynamics has far-reaching implications both for understanding human and animal behaviour, and for developing artificial systems that can interact with the real world.

I am interested in the dynamics and patterns generated by individual neurons and small neural networks (typically up to 10) and how they operate in their natural context across a range of time scales. I combine experimental techniques, computational tools and mathematical modelling to develop biologically-grounded models that capture fundamental phenomena in the brain. These can be used to characterise and make testable predictions regarding information encoding and animal behaviour, as well as implicated in artificial systems.