Behavioral Task

Balloon Analogue Risk Task

A research-style behavioral task that measures how you handle decisions when risk and reward both increase with every choice you make.

30 balloons · ~6 min · Free
Start the Task
A research-inspired implementation of the Balloon Analogue Risk Task paradigm (Lejuez et al., 2002). This is not affiliated with the original researchers or any official BART distribution.

How It Works

Pump a virtual balloon to earn points. Each pump adds more, but push too far and it pops. You decide when to stop.

Pump

Each pump earns 5 points and grows the balloon. Risk of popping increases every pump.

Collect

Cash in your points before it pops. Points transfer to your permanent bank.

Pop

If the balloon pops, you lose that round's points. The hidden pop point varies each balloon.

You'll start with 3 practice balloons, then 30 scored rounds.

Practice 1 of 3 0%
Practice
0 pumps 0 pts this balloon
Your bank 0 pts

Space = Pump · Enter = Collect · M = Mute

Practice Complete

Ready for the scored rounds? You'll complete 30 balloons. Your results will show how you balanced risk and reward across all 30 trials.

Your Results

Based on 30 balloon trials

Your 30 Balloons

Each bar is one balloon. Height shows how many times you pumped.

Collected Exploded

Worth Reflecting On

  • When did you feel the pull to pump just one more time?
  • Did you change your approach after a balloon popped?
  • Think of a real situation where you face similar risk-reward choices.
  • Do you tend to lock in smaller wins, or push for bigger payoffs?

The Research Behind the BART

The Balloon Analogue Risk Task was introduced in 2002 and has been used in hundreds of studies. Here's what researchers have found, and how we built our version.

The BART is a behavioral measure of risk-taking propensity. It captures how much reward someone is willing to pursue while risk grows with every step and losses are possible.

It is not a personality test, a diagnosis, or a measure of impulsivity on its own. Meta-analytic research across 22 studies and 2,120 participants found that links between BART behavior and traits like sensation seeking (~.14) and impulsivity (~.10) are small to moderate.

The Risk-Reward Tradeoff

Expected earnings per balloon based on pump count. For a 1–128 explosion range, the mathematically optimal stopping point is 64 pumps. Most people stop well below that.

From the Research

“Associations between BART behavior and traits like sensation seeking and impulsivity are typically small to modest in meta-analytic work.”

— Lauriola, M. et al. (2013). Individual Differences in Risky Decision Making. Journal of Behavioral Decision Making.

The BART produces several behavioral metrics. No single score tells the whole story.

Metric What It Captures Range
Adjusted Avg Pumps Average pumps on collected balloons only (standard primary score) 0–127
Explosions How often choices exceeded the hidden pop point 0–30
Total Earnings Points secured across all collected balloons 0–19,050
Post-Loss Adjustment Whether pumping changes after an explosion Varies
Pump Variability How consistent your pumping is across balloons 0+

About the adjusted score: Because it excludes exploded balloons, this metric can be biased. Formal analysis has shown it systematically underestimates pumping for cautious players (who rarely explode) and overestimates for aggressive players. That's why we show multiple metrics.

Original Development Paper

“Adjusted values, defined as the average number of pumps excluding balloons that exploded.”

— Lejuez, C.W. et al. (2002). Evaluation of a Behavioral Measure of Risk Taking: The Balloon Analogue Risk Task (BART). Journal of Experimental Psychology: Applied, 8(2), 75–84.

Methodological Critique

“The BART suffers from methodological problems” including optimal-strategy logic, censoring in the adjusted score, and ambiguity-to-risk learning effects.

— de Groot, K. (2020). Burst Beliefs: Methodological Problems in the Balloon Analogue Risk Task and Implications for Its Use. Journal of Trial and Error.

Test-Retest Reliability

Test-retest stability for adjusted average pumps over approximately two weeks has been reported as r ≈ .77.

— White, T.L. et al. (2008). Test-Retest Characteristics of the Balloon Analogue Risk Task (BART). Experimental and Clinical Psychopharmacology, 16(6), 565–570.

Key Takeaways

  • The BART has been studied for over 20 years across substance use, driving behavior, financial decisions, and more.
  • Links to real-world risk behaviors are mixed and depend on the specific domain and how the task is set up.
  • No clinical cutoffs or diagnostic thresholds exist. Results are best understood as behavioral patterns, not labels.
  • A single session captures a snapshot, not a fixed trait. State, motivation, and environment all affect performance.

Task Parameters

  • Version: BART-30 (30 scored balloons, plus 3 practice)
  • Explosion range: 1–128 pumps per balloon
  • Reward: 5 points per pump (virtual, no real money)
  • Explosion schedule: Fixed list identical for all users (mean pop point = 64), hidden from participants
  • Primary score: Adjusted average pumps, with supplementary metrics shown due to known bias in the primary score

How This Differs From Lab Versions

  • Virtual rewards: Research versions often use real money (e.g., $0.05/pump). Virtual points can change behavior and reduce comparability to published norms.
  • Visible earnings: We show current balloon points on screen. The original protocol does not display the accumulating reserve. We chose to show it for clarity, and disclose this difference.
  • Web environment: Device latency, distractions, and motivation in a web setting differ from a controlled lab session.

This implementation follows the BART-30 protocol described in Lejuez et al. (2002) and the common parameterization summarized in MacLean et al. (2018). It is independently built and not affiliated with the original researchers or any official BART distribution.

About This Assessment

This is an independently built web version of the Balloon Analogue Risk Task for educational use.

This tool is inspired by the BART described in Lejuez et al. (2002). We implemented the task mechanics from the published protocol using our own code, instructions, and UI. This is not the official BART software and we are not affiliated with the original researchers.

This is an educational self-reflection tool. It is not a diagnostic instrument, not a substitute for professional evaluation, and does not predict specific real-world outcomes for individuals. No clinical cutoffs exist for BART scores.

We use virtual points (not money), display current balloon earnings on screen (the original does not), and run in a web browser rather than a controlled lab. These differences can affect behavior and mean results are not directly comparable to published lab norms.