|
Abstract:
|
Health warning labels have been found to increase awareness of the harmful effects of tobacco products. An eye tracking study was conducted to determine the optimal placement of a health warning label for hookah pipes. Three areas of interest (AOIs) were considered for comparison: water bowl, stem and hose. Participants viewed images that contained one of four waterpipes, one of three warning labels, and placed in one of the AOIs. Typically, summary statistics such as total dwell time, duration score, and number of fixations on an AOI have been the focus for determining such placement. However, these summary statistics fail to capture the complete variability of eye movement over the entire image domain. Instead, we propose heat map estimation of eye coordinates via kernel density estimation, which are nonparametric, bivariate pdfs. For statistical analysis of heat maps, we use the Riemannian-geometric framework based on the Fisher-Rao metric. This metric-based framework enables efficient comparisons of heat maps, statistical summarization and exploration of variability of a sample of heat maps through the Karcher mean and principal component analysis, and metric-based clustering.
|
