Risk Related Brain Regions Detected with 3D Image FPCA

Speaker(s): 
Ying Chen (National University of Singapore)
Date: 
Monday, June 22, 2015 - 2:00pm
Location: 
Spandauer Straße 1, Room 23

Risk attitude and perception is reflected in brain reactions during an RPID experiment. Given the fMRI data, the question is how to detect the risk related regions and explain the relation between people's risk preference and brain activity. Conventional methods are often insensitive to the original spatial patterns and interdependence of the fMRI data. In order to cope with this fact we propose a 3D Image Functional Principal Component Analysis (3D Image FPCA) method that converts the brain signals to fundamental spatial common factors and subject-specific temporal factor loadings via proper orthogonal decomposition. A simulation study and real data analysis show that the FPCA method improves the quality of spatial representations with accurately detected risk related regions. The selected regions carry explanatory power for subjects' risk attitudes. In the application of risk classification, the proposed method reaches 100% accuracy for in sample analysis. In out-of-sample cross validation, it achieves 82% overall accuracy, with 100% correctly classifying strongly risk averse subjects, and 55% for weakly risk averse subjects.