2011
EMSS 2011 Proceeding
Simulation and Model Calibration With Sensitivity Analysis For Threat Detection in Brain
Authors: Keegan Lowenstein, Brian Leventhal, Kylie Drouin, Robert Dowman, Katie Fowler, Sumona Mondal
Abstract
In this study we use optimization techniques and sensitivity analyses to provide more rigorous, quantitative connectionist models of the functional interactions between the brain areas involved in detecting and orienting attention towards threat. A toolkit has been developed using flexible neural network modeling with automated parameter estimation. A sensitivity analysis is provided within the framework to identify significant model parameters and better understand model dependencies. These studies emphasize the importance of fitting the models to behavioral reaction time and brain activation data. They also show that the specific architecture of the model, and the numerical precision in the model parameters is important in determining an acceptable fit of experimental data, and that it is not the case that any model will work given the appropriate set of connection strength parameters.
I3M Scientific Sponsors
I3M Industrial Sponsors
I3M Media Sponsors
