Poor Players Weren’t More Accurate—They Just Shot First

Few ideas in psychology have travelled as far as the “scarcity mindset”. It says that having too little narrows your attention, like looking down a tunnel.

The idea shaped a best-selling book and influenced anti-poverty policy around the world.

Now three researchers have gone back to the experiment at its heart. Their verdict: the results were real, but the explanation probably wasn’t.

What is the scarcity mindset?

The scarcity mindset is the idea that when people have too little of something, such as money or time, it changes how they think.

The theory was popularised by economist Sendhil Mullainathan and psychologist Eldar Shafir in their 2013 book, Scarcity. It makes two main claims.

The first is a “bandwidth tax”. Worry about scarce resources is said to use up mental capacity, leaving less for other things.

The second is “tunneling”. Scarcity is said to focus attention on the pressing problem, like this month’s bills.

That focus can help, but it means other important information gets neglected.

Tunneling has been used to explain why people in poverty sometimes take out costly loans. In this view, they focus so hard on today’s shortfall that they don’t notice the high interest.

This new study looks only at tunneling. It doesn’t test the bandwidth tax.

The study at the centre

The original game

In 2012, Anuj Shah, Sendhil Mullainathan and Eldar Shafir published “Some consequences of having too little” in Science.

In one key experiment, people played a game called Angry Blueberries.

They fired blueberries at targets to earn points, which raised their chances of winning a prize.

Some players were made “rich”, with plenty of shots. Others were made “poor”, with far fewer.

What the original study reported

The “poor” players spent longer aiming and earned more points per shot. The original authors read this as scarcity’s “focusing effect”.

When borrowing was allowed, the “poor” borrowed more of their budget and ended up doing worse. The authors read that as attentional neglect: focused on now, blind to the cost later.

What the new re-analysis found

Who did it

The new paper, published in Nature Communications in September 2026, is by Fiona tho Pesch, Antonia Langenhoff and Mahesh Srinivasan. Their affiliations include the University of Cologne, Max Planck Institute for Research on Collective Goods, UC Berkeley and Stanford.

They re-analysed the original data, from 68 people. They also re-analysed the original authors’ own large 2019 replication, with 974 people after exclusions.

“Better focus”? Not quite

The researchers noticed something simple. On every level, early shots score more because there are more targets left to hit.

“Poor” players only ever took the first three shots of a level. “Rich” players kept shooting at nearly empty levels, which dragged their average down.

So they compared only the first three shots, which both groups took. There, the “poor” were no more accurate.

Across all players, the average score fell from 2.81 points on the first shot of a level to 1.80 by the third. For “rich” players, it fell to 0.69 by the fifteenth shot.

In other words, the “poor” scored more per shot because of the game’s rules, not because they aimed better.

“Ignoring interest rates”? Also not quite

The re-analysis found that “poor” players simply hit the borrowing option far more often. With only three shots per level, they ran out quickly.

When they actually faced the choice, the “poor” were, if anything, less likely to borrow than the “rich”. That difference wasn’t statistically significant.

Borrowing hurt everyone’s score. The “poor” were simply exposed to the option far more often.

A new experiment

The team then ran their own pre-registered version of the game with 174 online participants.

This time, both groups got the same chances to buy insurance against a future “drought” that could wipe out points. Eight insurance deals were good value, and two were “scams”.

Tunneling predicts the “poor” would neglect the future and skip good insurance. They didn’t.

The “poor” bought about as many good deals as the “rich”. And they were significantly less likely to buy the scam insurance.

Why the result needs caution

The new study doesn’t prove scarcity has no effect on thinking. It shows that one cornerstone experiment doesn’t show what it was thought to show.

All the studies used online players on Amazon’s MTurk platform, playing a lab game. That’s a long way from real poverty.

The new experiment had fewer participants than the authors had planned, 174 instead of 200, because of budget limits. They also changed their planned analysis, which they report openly.

The original authors have defended their work in the past. The broader scarcity literature has had mixed replication results, so this debate isn’t over.

Funding came from the Center for Effective Global Action and the James S McDonnell Foundation, and the authors declared no competing interests.

What earlier research found

The original claim

The 2012 Science study argued that scarcity “leads to attentional shifts that can help to explain behaviors such as overborrowing”.

The bandwidth tax

A 2013 Science study by the same group tested farmers before and after harvest. They did worse on thinking tests before harvest, when they were poor, than after it.

That study tested the “bandwidth tax”, not tunneling, so the new paper doesn’t directly challenge it.

A big replication project

In 2018, a project in Nature Human Behaviour re-ran 21 social science experiments published in Nature and Science. Only 13 found a significant effect in the same direction, and effects averaged about half the original size.

The new paper notes that Study 1 of the 2012 scarcity paper is among those that failed to replicate.

Auditing the whole field

A 2021 review in PNAS replicated 20 studies on the psychological effects of scarcity. It found “considerable variability”, with “some strong successes and other undeniable failures”.

How it fits together

The scarcity idea isn’t dead. But its most famous evidence for tunneling now looks much weaker than the book and policy reports suggested.

How much should you trust this?

Promising. The re-analysis is careful and transparent, and it’s backed by a new experiment. But it’s still lab games, not real lives.

What makes it convincing

• It uses the original authors’ own data, including their large 2019 replication.
• The explanation is simple and easy to check: early shots score more.
• The new experiment was pre-registered, and its data and code are public.
• Peer review was handled by Nature Communications, with a published peer review file.

What makes me cautious

• Lab games may not capture real poverty, as the authors themselves say.
• The new experiment was fairly small and fell short of its planned size.
• It tests tunneling only, not the wider scarcity theory.
• The original authors may contest the new reading of their data.

What this means for you

• Be wary of “mindset” explanations for poverty. This study suggests circumstances can explain behaviour that looks like poor thinking.
• Look at the options people face. People with less money are offered high-cost loans far more often, which may matter more than how they think.
• Don’t assume tight budgets make you careless. In the new game, players with less were more careful with bad deals.
• Read big psychology claims with healthy doubt. Even well-known findings can rest on a single experiment.

In this talk, hosted by Princeton’s Woodrow Wilson School, Eldar Shafir explains the original scarcity theory:

What we still don’t know

1. Does tunneling exist at all? No study yet shows both the focus and the neglect clearly.
2. What exactly is inside the “tunnel”? The theory doesn’t define it clearly enough to test.
3. How does real poverty shape attention? Lab games can’t fully answer that.
4. What about the bandwidth tax? That part of the theory needs its own careful tests.
5. How much do circumstances explain? The authors call for more research on the different choices poorer people face.

My take: blame the game, not the brain

What I like about this paper is how simple its key insight is. Early shots score more, and the “poor” only ever took early shots.

I’m less convinced that this closes the book on scarcity. It knocks out one pillar of the theory, not the whole building.

But it’s a useful reminder. Before we explain poverty by how poor people think, we should look hard at the choices they’re actually given.