Where does loss aversion collide?

The finding that losses loom larger than equivalent gains, and the origin from which the rest of this map descends. In Prospect Theory: An Analysis of Decision under Risk, Kahneman and Tversky replaced the utility function over final wealth with a value function defined on gains and losses relative to a reference point: it is concave for gains, convex for losses, and steeper for losses than for gains, so the pain of a loss exceeds the pleasure of an equal gain. Tversky and Kahneman's 1991 reference-dependent model carried the idea into riskless choice and put the loss/gain ratio near two. This is the referent construct: every other field here applies the same kink to a different object.[1][2]

Unit

A person's value function around a reference point

Recovery means

Not applicable

Measured by

Choices between gambles fit to a value function, with a loss/gain coefficient estimated near two

Loss aversion in how buyers respond to price: a price above a shopper's reference price is felt as a loss and depresses demand more than an equal price below the reference lifts it. Hardie, Johnson, and Fader fit a brand-choice model in Modeling Loss Aversion and Reference Dependence Effects on Brand Choice, coding each brand's price and quality against a reference formed from prior purchases, and found the response to losses steeper than the response to gains. Here the reference point is the price a shopper has come to expect rather than a price already paid, and the same asymmetric kink shapes the demand curve.[3]

Unit

A shopper's response to price relative to a reference price

Recovery means

Not applicable

Measured by

Brand-choice models with separate coefficients for price gains and price losses from the reference

Loss aversion read from the brain: the neural response to potential losses is stronger, or falls off faster, than the response to matched gains. Tom, Fox, Trepel, and Poldrack scanned people weighing mixed gambles in The Neural Basis of Loss Aversion in Decision-Making Under Risk and found a set of regions, including the ventral striatum and prefrontal cortex, whose activity rose with the size of a potential gain and fell more steeply with the size of a potential loss, with the neural asymmetry tracking each person's behavioural loss aversion. The referent is a pattern of blood-oxygen signal, not a coefficient in a choice model, but it is the same kink measured in tissue.[4]

Unit

A neural signal's asymmetric response to losses and gains

Recovery means

Not applicable

Measured by

fMRI activity to mixed gambles, with gain and loss slopes compared within each person