The Iowa Gambling Task (IGT) is one of the most widely used paradigms in decision-making and clinical neuropsychology. Participants repeatedly choose from four decks of cards (A, B, C, D) over 100 trials. Each card yields a monetary gain; some cards also carry a penalty. Participants are not told which decks are advantageous — they must learn through experience.

The four decks differ in their long-run payoff structures:

Decks A and B (disadvantageous): high immediate gains ($100 per card) but large penalties that produce a net loss of $250 per 10 cards.
Decks C and D (advantageous): lower immediate gains ($50 per card) but smaller penalties that produce a net gain of $250 per 10 cards.

The two disadvantageous decks differ in their penalty frequency: Deck A delivers frequent small-to-moderate penalties across 5 of every 10 cards; Deck B delivers a single large penalty ($1,250) once every 10 cards. The advantageous decks are analogous: Deck C has frequent small penalties; Deck D has one moderate penalty ($250) per 10 cards.

Participants begin with a $2,000 loan and see each card's win and net outcome immediately after choosing. Healthy adults gradually shift towards Decks C and D across the 100 trials; patients with ventromedial prefrontal cortex (vmPFC) damage, and individuals with high trait risk-taking or substance dependence, often continue to prefer the high-reward/high-loss decks.

The primary dependent measure is the net score: (C + D) − (A + B) choices, computed over the full 100 trials or across five successive blocks of 20 trials to track learning over time.