The Iowa Gambling Task and Risky Decision Making

In my last post, I wrote that people who’re primed to think about free will tend to make riskier decisions. This is true, but like many things in psychology, it’s not quite as simple as it sounds at first.

The problem is that “risky decisions” aren’t a tangible thing that’s easy to quantify. When I say that people with free will are more likely to make risky decisions, what I mean is that they’re more likely to behave a certain way on a laboratory task.

In this case, the task is something called the Iowa Gambling Task (IGT). The IGT is a simple game where you win or lose money, and different people play the game with different strategies.

I’m going to talk briefly about how the game works, what it tells us about how people make decisions and evaluate risk, and how the strategy you use might say something about how you make decisions.

However, before I do this, you might want to try playing the game yourself. There’s an iOS app and an Android app available. If you want to try it firsthand, you should stop reading now and do it – I’m about to give away the secret to how the game works, after which point you won’t be able to play anymore.

OK, don’t say I didn’t warn you!

Iowa Gambling Task

Picking a $120 reward card from deck A in the IGT

The IGT consists of four stacks of cards, and you play by turning over cards from any of the stacks. At any point, you choose which stack you want to pick a card from.

Some of the cards give you rewards, and you gain money. Other cards penalize you, and you lose money. The goal, of course, is to end up with as much money as possible.

When you flip over a given card, it’s up to chance whether that card’s going to reward you or penalize you. But there’s one thing you do know that makes the game work: the four decks are stacked differently, so some decks are better overall than others.

That’s all you know when you start. What you don’t know is the specific rules of how the decks are stacked (spoiler alert!):

  • The cards in decks A and B have larger rewards than the cards in decks C and D
  • The penalty cards in deck A are more frequent than the penalty cards in deck B, but the penalties in deck B are larger than in deck A. Likewise, the penalty cards in deck C are more frequent than the penalty cards in deck B, but the penalties in deck D are larger than in deck c.
  • Altogether, the penalties in decks A and B outweigh the rewards, but the rewards in decks C and D outweigh the penalties. In other words, choosing cards from the decks with the higher rewards loses money over time, and choosing cards from the decks with the lower rewards gains it.

Therefore, the basic idea is that there’s a conflict between short-term and long-term rewards. If you keep picking the cards with the higher short-term rewards, you’ll lose money in the long term.

When researchers started having test subjects play the IGT, they found that people with damage to the prefrontal cortex, a part of the brain involved in decision making, don’t adapt to choose from the “good” decks that win money over time. They keep going for the decks that lose money but have larger immediate rewards.

The initial assumption was that most healthy people explore all the decks in the early stages of the game and grow to prefer decks C and D, learning to avoid the “bad” decks.

But it turns out that there’s actually quite a bit of variety in the approaches people take. Some people opt for the larger immediate rewards even if it means playing a suboptimal long-term strategy. Many people also develop a preference for the decks that have the most frequent rewards – these people choose more cards from deck B, which has very infrequent but very large penalties.

Overall, strategies that emphasize immediate rewards but tend to lose money over time are riskier. These strategies are linked with risk-taking and impulsivity although the exact nature of the connection isn’t clear.

So when I said in my last post that people who’re thinking about free will make riskier decisions, what I meant was that they play riskier strategies on the Iowa Gambling Task – that is, strategies that give larger short-term rewards but lose money in the long run.

And these risky strategies correlate with a variety of risky real-world behaviors, including everything from reckless driving to getting tattoos. Stop and think for a minute about how what strategy you use on a simple game of chance predicts these other things about you.

More generally, the research that’s been done using the Iowa Gambling Task gives us some insight into how people make decisions. For example, it tells us:

  •  Sleep can improve decision making: Players who sleep between games choose deck B (the one with high, frequent rewards but occasional devastating losses) less often.
  • Hunger can improve decision making: People who are hungry actually perform better on the IGT. Apparently being hungry doesn’t increase your appetite for risk.
  • Too much and too little anxiety both lead to bad decisions: People with a moderate amount of anxiety are the only ones who win money on the IGT as a group. People with low levels of anxiety preferred deck B because they were overly focused on rewards while people with high levels of anxiety preferred deck A for reasons that aren’t entirely clear.
  • Working memory helps decision making: People with high working memory (the kind of memory you use to hold things in your mind and, well, work with them) do better on the IGT than people with low working memory. One possibility is that making good decisions means being able to hold all the relevant factors in your mind and take them into consideration.

So this is one of the things that’s cool about psychology: something as simple as how someone plays this basic card game ties into all these other factors.

But that’s one of the things that makes psychological studies so open to interpretation too. Going back to the study of free will, you could say that people who are thinking about free will make “riskier decisions.” But you could also say they have worse decision making abilities. Or they’re less sensitive to punishment. Or they’re just worse at math. Some of these things are related, but they’re different ways of framing the same data.

In fact, there’s still a lot of disagreement over how to interpret the way people play the IGT. After the IGT was introduced, it took a while before researchers realized just how often “normal” people prefer decks with frequent rewards that still lose over time (especially deck B, which turns out to be way more popular than it has any right to be!).

Therefore, the IGT is probably a good example of how you should look at results from any psychology study: half with wonder at what well-designed experiments can reveal and half with skepticism over how the results are being interpreted.

Dice image: Henning Buchholz

IGT image: Wikipedia/PaulWicks


  1. Michelle on January 29, 2020 at 4:57 pm

    I think if this game were being played with real money you would see different outcomes. Since there were no real world consequence to losing why not take a greater risk?