Is someone who acts as described in prospect theory always more likely or less likely to take a gamble than someone who acts as described by expected utility theory? Why? Are there conditions (such as on the weights) where you can answer this question definitively?