This research aims to better understand human decision-making in circumstances involving risk. It is widely accepted that when uncertainty is involved, human agents do not always act in accordance with expected utility maximization theories (EU). Further, when people make choices, they are able to distinguish both which choice they prefer and how strongly they prefer it. While baseline models simply look to model which choice a person selects, there is room to improve inference and illuminate decision-making processes by utilizing strength of preference ratings. Using Kahneman and Tversky’s cumulative prospect theory (CPT) as a baseline, this project aims to understand how the CPT model can be improved by incorporating strength of preference ratings in a Bayesian model estimation. CPT models allow for psychological concepts including loss aversion and asymmetric risk perceptions to better capture potentially irrational human responses to risk. To build upon this model, a study was run to collect choices from 225 unique gamble choice pairs, each offering a choice between two dollar-probability combinations. 60 participants completed the study, providing a robust dataset with choices, strength of preference ratings, attractiveness ratings, and buy ratings for each gamble. Research and model implementation is ongoing, however, there is strong evidence that after implementing the baseline CPT model, inference can be improved by incorporating these strength of preference ratings.