A number of recent experimental studies have focused on fitting parameterized functional forms to cumulative prospect theory's weighting function. This paper examines the behavioral implications of the functional forms and the estimated parameters. We find that none of the parameterizations can simultaneously account for gambling on unlikely gains and the Allais paradox behavior. Parameter estimates that lead to reasonable amounts of insurance and gambling behavior tend to also generate large risk premia. Taken as a whole, the analysis suggests that the functional forms proposed in the literature are not suitable for application to many settings.