It’s obvious when we think about lead athletes, like gymnasts, that people are capable of remarkable feats of coordination. But in fact, keeping our bodies balanced and moving different parts of the body relative to each other in a coordinated manner while remaining stable are complex problems that the brain solves all the time. How does the brain control our movements and what happens when the underlying neural circuits malfunction?

The Neural Circuits and Behaviour Lab studies the cerebellum, a brain area that is critical for coordinated motor control and motor learning. The well-described cerebellar circuit is conserved across species, which enables the researchers to study it in mice, a powerful animal model that offers an array of genetic tools. Using these tools, the researchers are able  to measure and manipulate activity in specific populations of neurons in the cerebellum. In some cases, these manipulations mirror neural conditions that exist in humans who suffer damage to the cerebellum through illness or injury.  Among the recent advances in the lab was the development of LocoMouse, an open source tool that allows researchers to observe the fine details of movement with near-millisecond and millimeter resolution. The team developed this tool as the first step in their large-scale project to reveal the neural circuits that generate coordinated walking, for which the lab has received significant support for from the European Research Council.

Main Interests

How the brain generates and controls coordinated movement


Quantitative behavioural analysis, Electrophysiology, Optogenetics

Models and Regions


The cerebellum is composed of many cell types, including Purkinje cells, labelled here in yellow.


Cerebellar contributions to coordinated locomotion in mice

Gait ataxia, or uncoordinated walking, is one of the most prominent symptoms of cerebellar damage, but the mechanisms through which the cerebellum contributes to coordinated locomotion are not well understood. Both ataxic mouse mutants and the sophisticated genetic tools available for manipulating neural circuits in mice have the potential to help shed light on this problem. However, analyses of mouse gait have typically lacked the kind of detail about the precision and timing of limb movements that would be required for a full analysis of coordination. We have built a custom video tracking system (LocoMouse) for measuring and analysing overground locomotion in freely walking mice. The LocoMouse system automatically detects the position of paws, snout, tail, and body centre in all three spatial dimensions with high spatiotemporal resolution. We have used this system to generate a comprehensive description of mouse gait parameters, including continuous paw, snout, and tail trajectories in time. We are applying the LocoMouse system to quantify the coordination deficits of ataxic mouse lines with cerebellar dysfunction, such as the Purkinje cell degeneration (pcd) mouse. These experiments are helping us to understand how the cerebellum contributes to specific elements of coordinated movement.


Neural mechanisms of locomotor adaptation

Locomotor patterns are constantly adapted for changing environments but the neural mechanisms underlying this basic form of learning are not well understood. Locomotor adaptation has been studied in humans using a motorized split-belt treadmill in which the limbs on opposite sides of the body move at different speeds. Subjects adapt to split-belt walking over time by changing spatial and temporal gait parameters, which show negative after- effects in post- adaptation. This type of motor learning is thought to involve the cerebellum, as previous studies have indicated that patients with cerebellar lesions cannot adapt to the perturbation (Morton & Bastian, 2006). However, the circuit mechanisms within the cerebellum that support this adaptation are not known. We have built a split-belt treadmill for mice and are using it in combination with genetic and electrophysiological tools to investigate the neural basis of locomotor adaptation.


Dissecting the role of endocannabinoids in eyeblink conditioning

Delay eyelid conditioning is a simple form of classical conditioning that depends critically on an intact cerebellum. Multiple synaptic plasticity mechanisms within the cerebellum have been identified and proposed as cellular substrates of learning for this behaviour. One class of molecules that appears to be important is endocannabinoids. Both cannabis users and cannabinoid receptor (CB1) knockout mice have been shown to exhibit impairments in delay eyelid conditioning. However, endocannabinoids are important for multiple plasticity mechanisms at many synapses, and it is not clear exactly where or how they act to modulate eyeblink conditioning. We are taking a genetic approach to this problem, by deleting CB1 receptors selectively from identified cell types within the brain. Through behavioural and electrophysiological experiments in these mice, we aim to constrain both the candidate sites and mechanisms of action for CB1 receptors in eyelid conditioning.