Event Details:
Constrained Bayesian ICA for brain connectomics
Claire Donnat
Mind, Brain, Computation and Technology graduate trainee, Stanford University
Brain Connectomics is a developing eld in neuroscience which strives to understand cognitive processes and psychiatric diseases through the analysis of interactions between brain regions. However, in the high-dimensional, low-sample, and noisy regimes that typically characterize fMRI data, the recovery of such interactions remains an ongoing challenge: how can we discover patterns of co-activity between brain regions that could then be associated to cognitive processes or psychiatric disorders?
In this talk, we investigate a constrained Bayesian Independent Component Analyis (ICA) approach which simultaneously allows (a) the exible integration of multiple sources of information (fMRI, DWI, anatomical, etc.), (b) an automatic and parameter-free selection of the appropriate sparsity level and number of connected submodules and (c) the provision of estimates on the uncertainty of the recovered interactions. Our experiments, both on synthetic and real-life data, validate the exibility of our method and highlight the benets of integrating anatomical information for connectome inference.
Related Paper
[1] Constrained Bayesian ICA for Brain Connectome Inference
Human perception of intermediately-fusible visual stimuli
Guillaume Riesen
Mind, Brain, Computation and Technology graduate trainee, Stanford University
Our eyes each have their own view of the world, but we perceive a single field of vision. Differing features are either combined into an intermediate percept, or one is completely suppressed of one in favor of the other. The former is known as binocular fusion while the latter leads to binocular rivalry, where perception switches dynamically from one eye to another if highly-divergent inputs are maintained over time. Both phenomena have garnered great interest from investigators seeking to understand how sensory signals are integrated and become percepts, but they have generally been studied separately and considered mutually exclusive processes. Rather than using only very-different or very-similar binocular inputs to specifically elicit rivalry or fusion, I have parametrically varied the amount of mismatch between each eye’s view to explore the relationship between these processes. My results show that some static stimuli can result in both fusion and rivalry over time. I will present these findings and discuss their implications for models of binocular vision hoping to capture the perceptual dynamics observed.
Curriculum vitae
Related Papers
[1] Humans Perceive Binocular Rivalry and Fusion in a Tristable Dynamic State