My research is comprised of many closely related projects. While the experimental protocols vary significantly, the ultimate aim of all the projects is to gather converging evidence revealing principles of neural information processing.
My working hypothesis is that changes in spike and field-potential correlations reflect brain networks reconfiguring themselves to dynamically process information, and to engage different brain regions for specific computational needs.
I am looking for these signature changes in all our data now. If I confirm these changes are occurring, the data will help refine computational models of neural information processing that I am currently developing.
These models include mechanistic details of how the brain can control the dynamic reconfiguration of groups of neurons to perform useful computation.
Put simply, these results could point the way to computers that, rather than just follow programmed instructions, can actually think!
Project 1
Title:
Unifying neural dynamics and brain function – controlling the flow and transformation of information in the brain.
Research Theme:
Neural Computation
Lead Investigator:
Stratton, Peter G.
Project Summary:
Many researchers now believe that complex patterns of oscillatory activity are somehow used by the brain to support all forms of neural computation. Currently though, we lack any unifying framework or specific quantified theories for understanding how computation is rooted in these neural dynamics. I propose a new theory called neural computation with competition, resonation, self-organisation and reward (Neural Compressor theory), explaining how the combination of neural connection patterns, activity oscillations and learning could be used to control the flow of information in the brain, and to activate and deactivate brain regions as required, in order to perform computation.
Publications:
– Computing with metastability: competitive cross-coupling (CXC) in neural circuits (submitted to Frontiers in Systems Neuroscience)
– Global segregation of cortical activity and metastable dynamics (submitted to Frontiers in Systems Neuroscience)
Project 2
Research Theme:
Deep Brain Stimulation (PPN)
Title:
Role of the PPN in gait control.
Lead Investigator:
Sah, Pankaj
Project Summary:
The pedunculopontine nucleus (PPN) is a part of the mesencephalic locomotor region. Lesions of the PPN induce gait deficits, and the PPN has therefore emerged as a target for deep brain stimulation for the control of gait and postural disability. We try to understand the role of the PPN in gait control using extracellular single-unit recordings taken in patients during surgery. Results suggest that changes in gait initiation in Parkinson’s disease may result from disrupted network activity in the PPN.
Publications:
Tattersall, T. L., P. G. Stratton, et al. (2014). “Imagined gait modulates neuronal network dynamics in the human pedunculopontine nucleus.” Nature Neuroscience17(3): 449-454.
Project 3
Research Theme:
Deep Brain Stimulation (GPi)
Title:
Comparing and contrasting GPi activity in Dystonia and Tourette’s syndrome.
Lead Investigator:
Sah, Pankaj
Project Summary:
Dystonia and Tourette’s syndrome are both movement disorders which respond to DBS of the GPi. However the symptoms of each disease are distinct, and the specific DBS targets within the GPi differ. This gives us the opportunity to compare GPi activity in both diseases simultaneously, with each acting as a quasi-control for the other. Since we cannot invasively record from healthy human brains (or from healthy parts of diseased brains), this is a unique opportunity to identify similarities and differences in brain activity in distinct conditions.
Publications:
in preparation.
Project 4
Confidential.
Lead Investigator:
Silburn, Peter A.
Project 5
Research Theme:
Sensory response in amygdala
Title:
Amygdala in-vivo single- and multi-cell response to auditory stimuli.
Lead Investigator:
Sah, Pankaj
Project Summary:
Auditory fear conditioning is a widely used paradigm to study the physiology of associative memory. However, amygdala neurons’ activity entrained by auditory stimulation and the potential ensuing network dynamics remain unexplored. We are finding that single-unit responses to auditory stimuli are unreliable and vary from day to day. Results indicate that networks of neurons, rather than any single unit, represent frequency responses in the amygdala.
Publications:
Windels, F., Stratton, P. and Sah, P. (2014) “Auditory stimulation modulates amygdala network dynamics.” BMC Neuroscience 15, p.52. (Abstract only)
Project 6
Research Theme:
Fear conditioning
Title:
Amygdala networks during fear conditioning.
Lead Investigator:
Sah, Pankaj
Project Summary:
Publications:
mind2mojo 