The nervous system must continually adjust sensitivity and action as the risk of harm changes. We investigate how neural circuits organise this adaptive control, how it depends on behavioural and internal state, and how it becomes persistently altered in chronic pain.
Quantitative behaviour · two-photon imaging · electrophysiology · optogenetics · chemogenetics · neural interfaces · machine vision · closed-loop systems
Research
All research →
How neural circuits transform sensory input into coordinated behaviour, how these transformations depend on behavioural and internal state, and how injury reorganises the relationship between sensation, neural activity, and action.
Explore →
Imaging, recording, neural manipulation and closed-loop systems that connect neural activity to behaviour in naturalistic environments.
Explore →
Developing neurotechnologies that sense neural, physiological and behavioural state and respond in real time, aiming to relieve chronic pain in patients.
Explore →The challenge of chronic pain
Pain is one of medicine’s oldest problems and remains one of its largest sources of disability and suffering. Persistent pain can remain after an injury has healed or arise without one, yet it is still among the conditions least well served by existing treatments. The difficulty is fundamentally mechanistic: persistent pain is not simply a louder version of acute pain, and the changes in the nervous system that sustain it are still poorly understood. Without knowing which mechanisms have changed, treatment will inevitably remain blunt. We aim to identify those mechanisms and use them to define what, where, how and when the nervous system should be targeted.
Protective control
Pain is part of a broader problem in neuroscience: how the nervous system regulates behaviour to reduce the risk of harm. Effective protection depends not simply on detecting danger, but on combining sensory information with internal state, ongoing behaviour and competing priorities to select the right action for the circumstances. We study the neural mechanisms that organise this protective control.
Selected work
All publications →Mechanical input is embedded with movement and arousal in the organisation of S1 population activity. Inflammation reorganises this stimulus–state geometry, changing how sensory input relates to the animal’s ongoing state.
Real-time body-part tracking and remote optical stimulation target defined areas of skin as mice move through complex environments, allowing sensory input to depend on what the animal is doing and where it is.
Scanned transdermal optogenetics controls genetically defined sensory input without touching the animal, while high-speed machine vision resolves the resulting local and whole-body actions at millisecond timescales.