Capturing Single Neuron Activity at the Millisecond Scale
The recording of single neural activity with millisecond resolution in freely moving animals continues to rely on classical extracellular recording methods. A key step in identifying single neuron spikes in extracellular recording is “spike-sorting” which relies on signal analysis methods such as blind source separation, statistics and modern machine learning approaches. Additionally, neuroscience datasets are becoming increasingly large and there is an growing need for automated and robust data processing pipelines.
My research for the past several years has focused on improving spike-sorting methods with a particular focus on super-computer simulations and identifying the limits of spike sorting and statistically robust automation (see Publications).
Brain Machine Interfaces
In addition to identifying single neuron spikes it is also important to carry out real-time manipulations of environment stimulus, single neurons or populations of neurons.
My research is also aimed at using opto-genetic stimulation (i.e. control of specific neurons using light) to show causality between specific brain areas or neuron populations and intrinsic behaviours such as goal oriented movements.