Knock-out for descriptive utility or experimental-design error?

Using an experimental design of stating equivalent probabilities for 252 stimulus lottery pairs, Chechile and Cooke (1997) alleged to have refuted generic-utility theory which itself comprises many modern utility theories. The present paper systematically investigates the feasibility of the Chechile-Cooke experimental design using numerical methods. We examine 1,277 utility setups (involving 17 parameter sets for four probability-weighting functions and 11 parameter sets for three component utility functions) which represent ten different utility theories. Our results demonstrate that on average for more than one third of all stimulus lottery pairs no equivalent probabilities exist. That is, the Chechile-Cooke experimental design prevents subjects from stating their true probability equivalents. Therefore, they cannot claim to have refuted generic-utility theory and the members of its family.