Iowa Gambling Task Modified for fMRI

As described in previous studies⁷, all participants completed an fMRI-adapted version of the IGT that included five blocks of decision-making conditions interleaved with five blocks of non-gambling control conditions. Each decision-making block consisted of 20 card selections. During each trial, subjects had three seconds to make the selection, followed by a three-second feedback screen during which they learned the amount won or lost for that trial. The task was programmed and presented using E-Prime Software.

The decision-making conditions involved a computerized task in which the participant saw four decks of cards on a screen. For each trial, the participant selected a card from any of the four decks. Each card selection yielded a gain, but it could also yield a loss. The amounts won and lost were then displayed, and the display also included the overall cumulative payoff, which was updated with each trial. The reward/penalty schedules were predetermined: Deck A and B yielded high immediate rewards, but carried a risk of much higher long-term penalties, that resulted in a total loss in the long run (disadvantageous decks); Decks C and D yielded low immediate rewards, but smaller long-term penalties, that resulted in a long-term gain (advantageous decks).

The visual stimuli and motor demands for non-gambling control conditions were as identical to the decision-making conditions presentation as possible: four card decks were presented, just as in the decision-making conditions. Subjects were instructed to select a card from a specific deck (e.g. “Select Deck B”). Thus, the non-gambling control conditions contained all the visual and feedback characteristics of the experimental tasks, but without the requirement to make a complex decision, as required in the decision-making conditions.