Decision Theory x Statistical Learning

Just as how human cognition inspired the creation of neural networks, I appreciate the elegance of using statistical techniques to understand the human mind - and vice versa.

Recently, my work has been about using Case-Based Decision Theory and VC-dimension to voting. The central question is: Can in-sample knowledge alone determine the best (presidential) candidate in terms of populational preferences? If so, can we establish meaningful bounds for the potential error in terms of dataset size and number of candidates? Another question I am dealing with is the estimation of utility and probability weighting functions in risky decision making using a Neural Network hereby circumventing traditional parametric assumptions.