Columbia Card Task

Background

The Columbia Card Task (CCT) is a classic psychological assessment test of risky decision making. The test was developed by Bernd Figner and colleagues at Columbia University in 2009 and allows researchers to study risky decision making under two conditions: 'hot' (= decision making under the influence of emotions) and 'cold' (= more deliberate, cognitive-controlled decisions making). Thus, it allows researchers to pinpoint whether a person's risk-taking is driven by impulsive emotions or more calculated reasoning.

The basic CCT task in both conditions is to decide how many cards to turn over of 32 possible ones if you know that there are (32-N) number of 'GAIN' cards (add points) but N 'LOSS' cards (subtract large number of points) and once you turn over one of the LOSS cards, the game is over. The game is played with different numbers of loss cards (manipulating the loss probability) and different gain and loss amounts.

The difference in the two versions is that the HOT version has participants turn over the cards one-by-one and provides immediate feedback by revealing either an uncovered smiley (gain) or a frowny (loss) face. This puts participants in a constant state of heightened emotional arousal (aka a "hot" state), as they experience the thrill of consecutive wins or the sting of a sudden, big loss. A variant of the HOT game procedure delays the feedback and allows participants to select all cards at once and only then reveals the smiley and frowny faces. In the COLD version on the other hand, participants select all the cards they want to turn over at once and only learn about the final outcome at the end of the game. By not revealing the outcome of individual choices, the COLD game version reduces emotional interference and allows more deliberate choices.

Figner et al results support that for adolescents the emotional influence on making risky decision tends to overshadow the making of controlled, deliberate decisions.

Task Procedure:

Under both conditions of the CCT task, participants work on 63 trials of which 54 are actual test trials and 9 trials are filler trials to hide the rigged feedback implemented in the hot condition: in test trials the hidden loss card(s) will always be the last possible card(s) that may be turned over by the participant.

Task Procedure: Hot

Under the CCT HOT condition, participants see a screen with 32 face-down cards and get information about the number of loss cards, amount won by gain cards and amount won by loss cards. They then proceed to click/touch on the cards one-by-one to reveal either a smiley or frowny face.

Task Procedure: Cold

Under the CCT COLD condition, participants see a screen with 32 face-down cards and get information about the number of loss cards, amount won by gain cards and amount won by loss cards. They also get to select how many cards they wish to uncover. Once they selected the appropriate number, the trial is over and the next one is started. No feedback is revealed until the end of the task.

What it Measures

The CCT is a measure of risk taking behavior under the influence of affective and cognitive information processing

Psychological domains

• Risky Decision Making: Making decisions that have risky consequences
• Affective (Emotional) Processing: Risky decision making under the influence of heightened emotions
• Deliberative (Cognitive) Processing: Risky decision making driven by deliberate, cognitive processes without emotional interference
• Impulsivity and Self-Regulation: It tracks the "stopping point"—the moment a participant's desire for more reward is outweighed by their fear of the penalty

Main Performance Metrics

• Average Number of Cards turned: indicator of risk propensity
• Total Score: A measure of overall task success or performance efficiency

Psychiatric Conditions

CCT performance tends to be higher or lower in the following groups:

• Adolescence
• Attention Deficit Hyperactivity Disorder (ADHD)
• Substance Abuse
• Major Depressive Disorder
• Anxiety Disorder