Figure 1.

Experimental procedure. A, We used giftcards to manipulate quality and quantity. Cards from different shops had different qualities, depending upon the subjective value of money that can be spent at that shop alone. Quantity varied as the amount of money (number of £) depicted on the card. B, Following an initial behavioral session in which we mapped value functions for different giftcards, a subset of participants was invited to return for an fMRI session. C, Behavioral experiment. The first task involved an auction (BDM procedure). Participants were offered different cards with varying amounts of money on them and indicated the maximum amount they would be willing to pay for that card. In the second (paired-choice) task, subjects made choices between pairs of giftcards with equal quantities (£20). D, fMRI experiment. On most trials (six of seven), participants saw only a single giftcard from one of three different shops, with a randomly varying quantity (amount of money). On decision trials (one of seven), a second giftcard was displayed 2 s after the first, and participants had 4 s to make a choice between the two giftcards. ITIs were normally distributed at ∼1.5 s.