📊 statistics

Perceived scarcity and risky choice under high objective risk: Evidence from a hierarchical drift diffusion model

This study demonstrates that perceived resource scarcity increases risky decision-making, particularly under high objective risk, by accelerating the drift rate toward the risky option as revealed through hierarchical drift diffusion modeling.

Original authors: Hongmin Luo, Junyu Liang, Jun Chen

Published 2026-06-30

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Hongmin Luo, Junyu Liang, Jun Chen

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are standing in front of a giant, spinning "Wheel of Fortune." You have to make a choice: spin a wheel that guarantees you a small, safe prize, or spin a wheel that might give you a huge jackpot but also might give you nothing at all.

Now, imagine the rules of the game change based on how much "fuel" you have in your tank.

The Rich Tank: You start with a full tank of fuel. You can afford to be patient.
The Empty Tank: You start with almost no fuel. If you don't get more fuel soon, you're going to run out.

This is the core setup of a study by researchers at South China Normal University. They wanted to find out: Does feeling like you don't have enough resources (scarcity) make people take bigger risks? And if so, how does their brain make that decision?

Here is the story of what they found, broken down into simple concepts.

1. The Setup: The "Noise" Game

The researchers didn't just ask people to imagine being poor. They made them feel it.

The Threat: Participants had to listen to a loud, unpleasant noise (like a harsh static hiss) for several seconds.
The Fuel: They were given "resource units" (points). Each point they earned shaved one second off the noise.
The Scarcity Manipulation:
- In the Rich condition, they started with 7 points. They were already safe from most of the noise.
- In the Scarcity condition, they started with only 2 points. They were in immediate danger of hearing the full noise.

Then, they had to play the Wheel of Fortune game to earn more points to stop the noise.

2. The Three Types of Risk

The researchers tested three different "flavors" of uncertainty:

Low Risk: The risky wheel had a 75% chance of winning. (Pretty safe).
High Risk: The risky wheel had only a 25% chance of winning. (Very dangerous).
Ambiguity: The wheel was covered up. They didn't know the odds at all. They just had to guess.

3. The Big Discovery: Desperation Makes You Gamble

When the researchers looked at the results, they found a clear pattern:

When people felt "rich" (safe from the noise), they played it safe. They mostly chose the guaranteed small prize.

When people felt "scarce" (in danger of the noise), they took bigger risks. They were much more likely to spin the risky wheel.

The Twist: This effect wasn't the same everywhere.

In the Low Risk situation, everyone chose the risky wheel anyway because it was so likely to pay off. Scarcity didn't change much.
In the Ambiguous situation, people were confused, so the effect was mixed.
In the High Risk situation, the difference was huge. When people were "rich," they avoided the high-risk wheel. But when they were "scarce," they jumped at the high-risk wheel.

The Metaphor: Imagine you are walking on a tightrope. If you have a safety net (Rich), you walk carefully. If the safety net is gone and you are about to fall (Scarcity), you might decide to take a running jump to a distant platform, even though it's a long shot, because staying put guarantees a fall.

4. The "Brain Engine": How the Decision Happened

The researchers didn't just look at what people chose; they used a sophisticated computer model (called a Hierarchical Drift Diffusion Model) to look at how the decision happened in the brain.

Think of the decision-making process like a racing car trying to reach a finish line (making a choice).

The Finish Line: One side is "Safe Choice," the other is "Risky Choice."
The Engine (Drift Rate): This is how fast the car gathers evidence to move toward one side.

What they found:
When people felt scarce, their "engine" revved up specifically toward the Risky Choice.

It wasn't that they started the race closer to the risky side (no starting bias).
It wasn't that they lowered the finish line to make it easier (no lower threshold).
It was that they gathered evidence for the risky option much faster. The idea of "I need a big win right now" became the dominant thought in their minds, pushing them toward the gamble with more speed and force.

5. The "Impulsivity" Factor

The researchers also asked participants to fill out a test about how impulsive they usually are (BIS-11 scale).

They found that people who scored higher on impulsivity showed an even stronger "engine rev" toward the risky choice when they were in the high-risk, scarce situation.
Essentially, if you are naturally impulsive and you feel like you have nothing left to lose, your brain accelerates toward the gamble even faster.

Summary

The paper claims that perceived scarcity doesn't just make people "stupid" or "reckless" in general. Instead, it changes the specific mechanics of how they think when they are in a high-stakes, high-risk situation.

When resources are low and the risk is high, the brain shifts gears. It stops weighing the odds carefully and starts accelerating toward the "big win" option with intense speed, because the alternative (staying safe) feels like a guaranteed loss.

Key Takeaway: Scarcity acts like a turbocharger for risk-taking, but only when the situation is already dangerous and the potential payoff is high.

Technical Summary: Perceived Scarcity and Risky Choice Under High Objective Risk

Problem Statement
Prior research indicates that resource scarcity alters decision-making processes, yet findings regarding its specific impact on risky choice are heterogeneous. Some theories suggest scarcity increases risk aversion due to the high cost of potential losses (vulnerability), while others propose it increases risk-seeking behavior as a desperate attempt to cross a sufficiency threshold (desperation). Furthermore, it remains unclear whether these effects depend on the informational structure of the choice environment (known risk vs. ambiguity) or which specific cognitive components of the decision process (e.g., evidence accumulation, response caution, starting bias) are modulated. This study addresses these gaps by investigating how perceived scarcity influences risky choice across high-risk, low-risk, and ambiguous conditions, and by characterizing the underlying decision processes using computational modeling.

Methodology
The study employed a within-participants experimental design ( $N=39$ ) utilizing an adapted Wheel-of-Fortune (WOF) task.

Resource Context Manipulation: Scarcity was induced on a trial-by-trial basis by varying the initial resource endowment (2 units vs. 7 units). These resources were directly linked to the avoidance of an aversive white-noise stimulus; participants earned resource units to reduce the duration of the noise. This approach aimed to maintain resource pressure within the task while minimizing variance from autobiographical recall.
Risk Conditions: The task included three uncertainty conditions:
1. Low Risk: High probability of gain ( $p=0.75$ ) with lower magnitude.
2. High Risk: Low probability of gain ( $p=0.25$ ) with higher magnitude.
3. Ambiguity: Probabilities and outcome compositions were initially concealed ( $p=0.50$ fixed, but unknown to the participant at onset), allowing for learning over trials.
- Expected value was matched nominally across options in the known-risk conditions.
Measures: Participants' choices and response times (RTs) were recorded. Post-task, participants rated their perceived resource availability, confidence, and stress. Trait impulsivity was assessed using the Barratt Impulsiveness Scale–11 (BIS-11).
Analysis:
- Behavioral: Binomial generalized linear mixed models (GLMMs) analyzed choice probabilities.
- Computational: A Hierarchical Drift Diffusion Model (HDDM) was fitted to joint choice and RT data to estimate latent parameters: drift rate ( $v$ ), decision threshold ( $a$ ), nondecision time ( $t$ ), and starting bias ( $z$ ). The model tested whether these parameters varied by resource context and risk type.

Key Results

Manipulation Check: Participants reported significantly lower perceived resource availability, confidence, and excitement in the scarcity context compared to the rich context. Stress levels did not differ significantly between contexts.
Behavioral Findings:
- The resource-scarcity context significantly increased the odds of making a risky choice compared to the resource-rich context.
- This effect was moderated by risk type. The increase in risky choice under scarcity was largest in the high-risk condition, intermediate in the ambiguous condition, and smallest in the low-risk condition.
- Exploratory analyses showed that higher BIS-11 impulsivity scores were positively associated with risky choice in the high-risk condition, though this interaction did not reach significance after multiple comparison corrections.
HDDM Findings:
- The best-fitting model (M3) indicated that the scarcity effect was primarily driven by changes in drift rate ( $v$ ) rather than decision thresholds or starting biases.
- Scarcity was associated with a higher drift rate toward the risky-choice boundary.
- Crucially, the interaction between scarcity and risk type mirrored the behavioral results: the scarcity-induced increase in drift rate was significantly stronger in the high-risk condition and weaker in the low-risk condition relative to the ambiguity condition.
- Exploratory individual-differences analysis revealed a positive correlation between BIS-11 scores and the drift-rate coefficient for the high-risk condition (relative to ambiguity), suggesting that impulsive individuals showed a stronger evidence accumulation toward the risky option under high-risk scarcity.

Significance and Claims
The authors claim that these findings provide process-level evidence that perceived scarcity selectively increases risky choice under conditions of high objective risk. The study supports a computational account where scarcity alters the quality or strength of evidence accumulation (drift rate) toward the risky option, rather than simply shifting a global risk preference or response caution.

The paper positions these results as a refinement of the "desperation threshold" model, suggesting that scarcity effects are nonlinear and context-dependent, emerging most strongly when resources are low and the stakes (risk level) are high. The authors explicitly note limitations, including that the task structure (aversive noise avoidance) may introduce urgency or relief-seeking motives alongside perceived scarcity, and that the ambiguous condition became learnable over time rather than remaining static. Consequently, the conclusions are framed specifically within the context of the current task structure and manipulation strength, offering a more specific characterization of scarcity-related decision processes without claiming universal applicability across all forms of risk or scarcity.