Project Reference: TRG27-PDN-LD1

Project title: Cortico-subcortical interactions in auditory statistical learning

Project outline

Sounds from sources all around us are constantly reaching our two ears. The brain, largely automatically, makes sense of this sensory input, detecting patterns (statistical learning) and identifying auditory objects through a combination of processes. Some processes act on the sensory input itself (bottom up), such as denoising, background suppression, or sound localization. Other processes make use of expectations and memory (top down), such as pattern detection, filling in sound gaps through predictions, or stream segregation. Combined, these processes result in statistical learning of the input patterns, something fundamental for perception. To date it is still not clear what the combined and separate contribution of sensory subcortical structures and primary cortices is and how these cortico-subcortical loops stabilize over adolescence. 

We study, in rodents, how subcortical structures represent the complex incoming input, what information they convey to cortex and, what the cortex projects onto subcortical structures depending on the current context. We are interested in how these processes work in healthy adults, in the growing brain, and in neurodiverse brains. We use a combination of naturalistic behaviours, state-of-the-art electrophysiology, and genetic manipulations in adult and developing mice. Behaviours can either be volitional group behaviours in large arenas or more controlled reinforced behaviours. To sample brain activity from several structures simultaneously, we use Neuropixels probes in combination with optogenetics. This is followed by computational analysis of this rich and comprehensive dataset, comprising the activity of neurons across several structures in awake behaving animals, as well as recordings of facial and body movement. Ultimately we are interested in understanding how the brain decodes the patterns in the soundscape and makes sense of it, through processes that use minimum energy but extract information with sufficient precision.

Principal Supervisor

Dr Livia de Hoz

Email address: ld749@cam.ac.uk

Department / Partner Institute: Physiology, Development and Neuroscience (PDN)

Website: https://www.dehozlab.org/

BBSRC strategic theme

Understanding the rules of life