Skip to main content

BART (Balloon Analogue Risk Task)

Version: v1 (current)

The Balloon Analogue Risk Task (BART) is an ecologically valid behavioral measure of risk-taking propensity. Participants pump up virtual balloons to earn rewards, but each pump increases both the potential payoff and the probability of explosion, creating a dynamic risk-reward decision-making paradigm.

Overview​

Risk-taking is a fundamental aspect of decision-making with implications for real-world behaviors ranging from financial choices to substance use. Unlike hypothetical risk assessments (e.g., questionnaires), the BART requires participants to make repeated consequential decisions where they experience direct gains and losses.

On each trial, participants decide how many times to pump a balloon. Each pump increases earnings but also brings the balloon closer to its (unknown) explosion point. If the balloon explodes, all earnings for that trial are lost. If the participant stops pumping and collects, they bank the accumulated reward. The task measures individual differences in risk tolerance, learning from negative feedback, and impulsivity.

The BART has demonstrated strong ecological validity. Performance predicts real-world risk behaviors including substance abuse, gambling, unsafe sexual practices, and financial decision-making better than self-report measures. It's widely used in addiction research, developmental studies, and investigations of decision-making under uncertainty.

Scientific Background​

Key Findings:

  • Ecological Validity: BART performance correlates with real-world risk behaviors (substance use, unprotected sex, risky driving) more strongly than questionnaire measures (Lejuez et al., 2002)
  • Primary Measure: Adjusted average pumps (pumps on non-exploded balloons) best captures risk-taking propensity while controlling for explosions
  • Learning Effects: Most participants show risk reduction across trials after experiencing explosions
  • Neural Correlates: Risky pumping engages ventral striatum (reward processing) and anterior insula (risk/uncertainty), with individual differences in activation predicting risk behavior (Rao et al., 2008)

Theoretical Framework: The BART creates a conflict between approach motivation (potential rewards) and avoidance motivation (potential loss). Unlike static risk assessments, it captures dynamic decision-making where participants update their risk estimates based on experience.

Seminal References:

  • Lejuez et al. (2002): Original BART development and validation
  • Hunt, Hopko, Bare, Lejuez, & Robinson (2005): Construct validity and real-world correlates

Why Researchers Use This Task​

Researchers choose the BART to:

  • Assess risk-taking propensity in adolescents, substance users, and clinical populations
  • Predict real-world risk behaviors (smoking, drugs, unsafe sex)
  • Study decision-making under uncertainty and probabilistic outcomes
  • Examine learning from negative feedback (explosions)
  • Investigate individual differences in impulsivity and sensation seeking
  • Evaluate interventions targeting risk behavior (e.g., substance abuse treatment)

Task Parameters​

Balloon Configuration​

ParameterTypeDefaultDescription
Bart Practice EnabledbooleanFalseEnable practice trials with separate balloons.
Bart Practice Trialsarray[]Practice balloon configurations (see Trial Structure below).
Bart Display Pump CountbooleanFalseShow current pump count to participant (recommended: false, to increase uncertainty).

Visual Customization​

Each balloon can be configured with:

ParameterTypeDefaultDescription
Balloon Color'blue' | 'orange' | 'yellow' | 'red' | 'green' | 'purple' | 'pink''blue'Visual appearance of balloon.
Scale Incrementnumber0.025How much balloon grows per pump (0.01–0.05 typical).

Label Customization​

All labels can be customized (defaults shown):

ParameterTypeDefault
Bart Label Total Earningsstring"Total Earnings"
Bart Label Current Balloonstring"Current Balloon"
Bart Label Pumpstring"Pump"
Bart Label Collectstring"Collect"
Bart Label Pumpsstring"pumps"
Bart Label Popstring"Pop!"
Bart Label Loststring"Lost"
Bart Label Collectedstring"Collected!"
Bart Label Earnedstring"Earned"

Instructions​

ParameterTypeDefaultDescription
Main InstructionsHTMLDefault textInstructions shown before task starts.
Hint InstructionsHTMLDefault textQuick reference shown via "?" button during task.
Practice InstructionsHTMLDefault textInstructions shown before practice trials begin.
Trials InstructionsHTMLDefault textInstructions shown after practice, before main trials.

Trial Configuration​

Each trial in bart_trials (and bart_practice_trials) is defined by:

FieldTypeRequiredDescription
Balloon ColorstringYesColor: Blue, Orange, Yellow, Red, Green, Purple, Pink.
Explosion PointnumberYesExact pump number at which balloon explodes (set by researcher).
Reward Per PumpnumberYesMonetary value earned per pump (e.g., 0.05 = 5 cents).
Scale IncrementnumberNoBalloon growth rate (default: 0.025).
BlockstringNoOptional block label for grouping trials.
Feedback Duration (ms)numberNoFeedback display duration: 0 = click Next to advance (default), >0 = auto-advance after delay.

Example Trial Sheet​

balloon_color | explosion_point | reward_per_pump | scale_increment | block | feedback_duration_ms
--------------|-----------------|-----------------|-----------------|-------|---------------------
blue          | 8               | 0.05            | 0.025           | main  | 0
orange        | 32              | 0.05            | 0.025           | main  | 1000
yellow        | 16              | 0.05            | 0.025           | main  | 0
red           | 64              | 0.05            | 0.025           | main  | 0
green         | 12              | 0.05            | 0.025           | main  | 0

Note: feedback_duration_ms of 0 means the participant must click "Next" to advance. A value greater than 0 (e.g., 1000) means the task auto-advances after that delay.

Design Tips:

  • Explosion Point Range: Use 4–64 pumps for realistic uncertainty (too low = always explode, too high = never explode)
  • Distribution: Uniform distribution (e.g., 4, 8, 12, 16, 24, 32, 48, 64) or skewed (more low-risk balloons)
  • Reward Structure: Fixed per-pump reward (e.g., $0.05) is standard; variable rewards can test sensitivity to payoff magnitude
  • Color: Different colors can signal different risk distributions (e.g., red = high risk) if explicitly instructed

Participant Experience​

Unmoderated/Moderated Mode​

  1. Main Instructions: Participant reads task instructions and clicks "Start"
  2. Practice Phase (if enabled):
    • Practice instructions appear
    • Practice balloons to familiarize with pump/collect mechanic
    • Total earnings do not carry over from practice
  3. Trials Instructions (if practice was enabled): Brief reminder before main trials
  4. Main Trials:
    • Balloon appears (visual representation with color)
    • Participant decides: Pump (increase earnings but risk explosion) or Collect (bank earnings safely)
    • Pump: Balloon grows larger, earnings increase, explosion risk increases
    • Collect: Earnings added to total, proceed to next balloon
    • Explosion: Balloon pops, lose all earnings for this balloon, proceed to next balloon
  5. Feedback: After each trial, show outcome (collected amount or explosion)
  6. Completion: Task ends after all trials, final total earnings displayed

Decision Process Per Trial​

1. Balloon appears → Current balloon: 0 pumps, $0.00
2. Participant pumps → 1 pump, $0.05
3. Participant pumps → 2 pumps, $0.10
4. Participant pumps → 3 pumps, $0.15
5. Participant collects → Earned $0.15, Total: $0.15
   (OR: Balloon explodes → Lost $0.15, Total: $0.00)

Data Output​

Participation Log Events​

Trial Start (bart_trial_start):

{
  "trial_index": 1,
  "stimulus_id": "bart_0_1",
  "balloon_color": "blue",
  "explosion_point": 16,
  "reward_per_pump": 0.05,
  "block": "main"
}

Pump Event (bart_pump):

{
  "trial_index": 1,
  "pump_number": 5,
  "current_earnings": 0.25,
  "latency_ms": 1234
}

Collect Event (bart_collect):

{
  "trial_index": 1,
  "pumps": 8,
  "earnings": 0.40,
  "total_earnings": 3.45,
  "latency_ms": 987
}

Explosion Event (bart_explosion):

{
  "trial_index": 2,
  "pumps_at_explosion": 16,
  "lost_earnings": 0.80
}

Summary Artifact​

At task completion, a JSON summary is generated (bart_summary_<taskIndex>.json):

{
  "task_kind": "bart",
  "task_index": 0,
  "total_trials": 30,
  "total_earnings": 8.45,
  "summary_statistics": {
    "adjusted_average_pumps": 12.5,
    "overall_average_pumps": 10.2,
    "number_of_explosions": 8,
    "explosion_rate": 0.27,
    "average_decision_latency_ms": 1456,
    "non_exploded_trials": 22,
    "exploded_trials": 8
  },
  "by_balloon_color": {
    "blue": {
      "total_trials": 6,
      "adjusted_average_pumps": 11.8,
      "explosions": 2,
      "explosion_rate": 0.33,
      "total_earnings": 1.90
    }
  },
  "trials": [...]
}

Key Metrics:

  • Adjusted Average Pumps: Average pumps on non-exploded balloons. This is the primary measure because it controls for explosion censoring.
  • Overall Average Pumps: Average across all trials (biased by explosions)
  • Explosion Rate: Proportion of trials ending in explosion (indicates risk tolerance)
  • Total Earnings: Cumulative reward (motivational check)
  • Decision Latency: Time between pumps/collect decisions (impulsivity indicator)

Design Recommendations​

Trial Counts​

  • Minimum: 15 trials for basic assessment
  • Standard: 30 trials (original BART)
  • Extended: 90 trials for individual differences research

Explosion Point Distributions​

Uniform Distribution (recommended for first use):

Explosion points: 4, 8, 12, 16, 24, 32, 48, 64 (repeat 3–4 times each)
Average explosion point: 26 pumps

High-Risk Distribution (for studying risk sensitivity):

Explosion points: 2, 4, 6, 8, 10, 12, 14, 16 (more frequent explosions)
Average explosion point: 9 pumps

Low-Risk Distribution:

Explosion points: 32, 40, 48, 56, 64, 72, 80, 96 (fewer explosions)
Average explosion point: 61 pumps

Reward Structure​

  • Standard: $0.05 per pump (scales to $3.20 for 64 pumps)
  • High Stakes: $0.10–$0.25 per pump (increases motivation)
  • Variable: Different reward rates per balloon color (tests payoff sensitivity)

Instructions Considerations​

  • Emphasize: "Each balloon has a different explosion point"
  • Do NOT reveal: Specific explosion points or distribution
  • Clarify: Explosions lose all earnings for THAT balloon only, not total earnings
  • Motivate: "Try to earn as much as possible"

Practice Trials​

  • Recommended: 3–5 practice balloons with varied explosion points (e.g., 8, 16, 32)
  • Purpose: Familiarize with pump/collect interface and explosion mechanic
  • Feedback: Same as main trials (explosions and collections)

Common Issues and Solutions​

Issue: All Participants Pump to Explosion​

Symptoms: Explosion rate >60%, very high adjusted average pumps

Possible Causes:

  • Explosion points too high (e.g., all >50 pumps)
  • Participants not understanding loss consequence

Solutions:

  • Lower explosion points (use 4–32 range for more moderate risk)
  • Emphasize in instructions: "Explosions lose ALL earnings for that balloon"
  • Include practice trials where some explode early

Issue: No One Takes Risks (Very Low Pumps)​

Symptoms: Adjusted average pumps <5, explosion rate <10%

Possible Causes:

  • Explosion points too low (frequent early explosions)
  • Instructions overemphasize risk

Solutions:

  • Increase explosion point range (e.g., 8–64 instead of 4–16)
  • Balance instructions: "Try to earn as much as possible" alongside risk warning
  • Check if rewards are motivating (may need higher per-pump values)

Issue: No Learning Across Trials​

Symptoms: Pumps remain constant across trials, no reduction after explosions

Solutions:

  • Increase trial count to 30+ (learning requires experience)
  • Use more varied explosion points to allow pattern learning
  • Analyze early vs. late trials separately

Issue: Participants Confused by Color Changes​

Symptoms: Questions about whether color matters

Solutions:

  • Use single color if colors have no meaning
  • If using colors to signal risk: Explicitly instruct (e.g., "Red balloons are riskier")
  • Make color-risk mapping consistent and obvious

Example Study Configurations​

Configuration 1: Standard BART​

Purpose: Replicate original task for risk-taking assessment

Settings:

  • Total trials: 30
  • Explosion points: Uniform distribution 4, 8, 12, 16, 24, 32, 48, 64 (each appears 3–4 times)
  • Reward per pump: $0.05
  • All blue balloons
  • Practice: 3 trials

Expected Performance:

  • Adjusted average pumps: 20–35 (varies widely by individual)
  • Explosion rate: 25–40%

Configuration 2: Substance Use Screening​

Purpose: Predict substance use risk behaviors

Settings:

  • Total trials: 30
  • Same as Configuration 1
  • Additional measures: Correlate adjusted average pumps with substance use history/intention

Expected Relationship: Higher adjusted pumps → greater substance use

Configuration 3: Color-Coded Risk Levels​

Purpose: Examine sensitivity to risk information

Settings:

  • Green balloons: Low risk (explosion points 32–64)
  • Yellow balloons: Medium risk (explosion points 12–24)
  • Red balloons: High risk (explosion points 4–12)
  • 10 trials per color, randomized
  • Explicit instructions about color-risk mapping

Hypothesis: Participants should pump more on green, less on red

Configuration 4: Developmental Study​

Purpose: Compare risk-taking across age groups

Settings:

  • Simplified instructions for children
  • Visual rewards (points/stars instead of money)
  • 20 trials (shorter for younger attention spans)
  • Explosion points: 8–48 (narrower range)

Expected Age Effects: Adolescents show higher risk-taking (more pumps) than children or adults

References​

  1. Lejuez, C. W., Read, J. P., Kahler, C. W., Richards, J. B., Ramsey, S. E., Stuart, G. L., ... & Brown, R. A. (2002). Evaluation of a behavioral measure of risk taking: The Balloon Analogue Risk Task (BART). Journal of Experimental Psychology: Applied, 8(2), 75–84.

  2. Hunt, M. K., Hopko, D. R., Bare, R., Lejuez, C. W., & Robinson, E. V. (2005). Construct validity of the Balloon Analogue Risk Task (BART): Associations with psychopathy and impulsivity. Assessment, 12(4), 416–428.

  3. Rao, H., Korczykowski, M., Pluta, J., Hoang, A., & Detre, J. A. (2008). Neural correlates of voluntary and involuntary risk taking in the human brain: An fMRI study of the Balloon Analog Risk Task (BART). NeuroImage, 42(2), 902–910.

  4. Aklin, W. M., Lejuez, C. W., Zvolensky, M. J., Kahler, C. W., & Gwadz, M. (2005). Evaluation of behavioral measures of risk taking propensity with inner city adolescents. Behaviour Research and Therapy, 43(2), 215–228.

Related Tasks: Go/No-Go Task (impulsivity), Iowa Gambling Task (decision-making under risk)