All the classic economic models for choice that I've seen fail to consider that the perception of not only risk but reward are BOTH nonlinear. IMO, this has been an Achilles heel of classic price and game theory. The rise of behavioral economics in recent decades would seem to agree with this iconoclysm.
If I need $1 million right now or else a loved one dies, then it doesn't matter how big the reward of a riskier alternative choice may be. I take the million NOW. If the additional reward is a victim of decreasing value as that offer rises, it's only rational for the decider to show diminished interest in choosing the greater reward (even if the marginal odds are only a tiny amount less likely).
Disregarding the reward curve of the individual is going to consistently misjudge economic choice and will surely be a poor basis for any economic model.
Not sure what you mean by this. Even in the most basic rational choice analysis, where the players are agents seeking to maximize a utility function, the "reward" can be nonlinear. I'm not familiar with any economic model that unintentionally restricts the agent's utility to be linear. Sometimes you assume that agents have a linear "utility of money", but everyone in economics knows that this is a taylor expansion around a small region where the linearity assumption is reasonable, and not an actual fact about human preferences.
If I need $1 million right now or else a loved one dies, then it doesn't matter how big the reward of a riskier alternative choice may be. I take the million NOW. If the additional reward is a victim of decreasing value as that offer rises, it's only rational for the decider to show diminished interest in choosing the greater reward (even if the marginal odds are only a tiny amount less likely).
Disregarding the reward curve of the individual is going to consistently misjudge economic choice and will surely be a poor basis for any economic model.