"Thus in prospect theory choice is a two-phase process, with prospects ''edited'' in the first phase using different decision heuristics, and choices then made in the second phase from a restricted or reformulated class of prospects...

The two-phase analysis makes it possible to look at gains and losses relative to reference points and ultimately introduce well-observed phenomena at odds with standard framework predictions such as diminishing sensitivity and loss aversion (Tverysky and Kahneman 1991). More generally, prospect theory opens the door to the investigation of a variety of descriptive and procedure invariance failures that cast doubt on the traditional idea that individuals possess stable and coherent preferences. Taken as a whole, then, the ''one feature common'' to all nonconventional theories that distinguishes them from those approaches meant to preserve a revised standard expected utility framework is that ''none of them can be reduced to, or expressed purely in terms of, a single preference function V(.)'' (Starmer 2000: 339)...

Researchers refer less to exprimental results as violations and anomalies, and now consider the possibility that standard expected utility theory is more likely a special case within a more general alternative theory rather than a general theory itself."

--- John B. Davis, Behavioral economics and identity