STUDYING SINGLE NEURON COMPUTATIONS WITHIN BRAIN-WIDE CIRCUITS

NIH RePORTER · NIH · F30 · $53,974 · view on reporter.nih.gov ↗

Abstract

PROJECT SUMMARY/ABSTRACT All our sensations, decisions, and actions are mediated by signals flowing through networks of neurons, the fundamental information-processing units of the brain. Each neuron receives thousands of inputs from other neurons at synapses and converts particular patterns of inputs into output (action potentials). Understanding the operations that neurons can perform will help reveal the algorithms that the brain utilizes to produce behavior. A rich body of theoretical work has put forward several models of neural computation of varying complexity. Some propose that spatially clustered inputs are more effective at driving output whereas other models suggest that diffuse input is better. It is still unclear which models, if any, accurately describe single neuron computation in vivo and during behavior. The study of single neuron computation requires the simultaneous measurement of many inputs to, and the output from, a neuron in vivo, which has not yet been possible. In my research, I will leverage novel imaging technologies to overcome current limitations and measure the input and output signals of individual cortical neurons. I hypothesize that action potential generation in vivo is more likely after a neuron receives clustered synaptic input. I will address this hypothesis through the following Specific Aims: Aim 1: Characterize the spatiotemporal relationships between synaptic input patterns and action potentials. Aim 2: Investigate how inputs change when the brain drives activity in a specific neuron. Aim 3: Determine whether hierarchical nonlinearities are needed to model a neuron’s computation. This project will result in new models of the neuron that better represent the range of computations performed in vivo. Given that neuropsychiatric diseases, such as schizophrenia, autism spectrum disorder, and bipolar disorder, have been hypothesized to affect proteins that dictate a neuron’s computational properties, a comprehensive understanding of neural computation would allow for mechanistic explanations of how such diseases affect patients’ cognitive capabilities.

Key facts

NIH application ID
10997010
Project number
1F30MH138009-01
Recipient
JOHNS HOPKINS UNIVERSITY
Principal Investigator
Michael Everest Xie
Activity code
F30
Funding institute
NIH
Fiscal year
2024
Award amount
$53,974
Award type
1
Project period
2024-09-01 → 2027-08-31