Cake Gambling Task
AKA: Cambridge Gambling Task
Background
The Cake Gambling Task (CakeGT) is a computerized cognitive assessment test of decision-making and risk-taking behavior outside of a learning context. It was introduced by Linda Van Leijenhorst and colleagues in 2008 as a child-friendly gambling task similar to the Cambridge Gambling Task (CGT). Like the CGT, the CakeGT evaluates how well a person makes choices and places wagers using visible odds without having to memorize or infer rules from past choices.
In the CakeGT, participants are shown a cake with 6 slices of either a strawberry or chocolate flavor, e.g. a cake can consist of 5 strawberry slices ("pink") and one chocolate one ("brown"). They are told that the computer will randomly select one of these slices, and participants are asked to place bets on which flavor the computer will choose. Below the cake, participants find two bet boxes, one for the strawberry and one for the chocolate slices, that contain a different number of coins that participants can bet on their choice. For example, they can bet one coin that the computer will select a strawberry piece or four coins that the computer will select a chocolate one. If they are correct, they win the coins. If they are wrong, they lose the coins. In the example described, the more likely event (the random selection of a strawberry piece with the probability of 5/6) is far less risky but also doesn't earn as much if won. Betting on the unlikely chocolate piece (probability of 1/6) is much riskier but would bring in more coins.
Using the CakeGT with participants of various age groups (starting with children age 8), Van Leijenhorst et al found that behavioral patterns in children and adolescents did not differ from adult patterns, showing a similar ability for probability estimation and reward evaluation.
Task Procedure
In the Millisecond implementation of the CakeGT, one flavor is selected as the 'low risk' flavor and the other one as the 'high risk' one. The selection is counterbalanced by groupnumber assignment. Bet amounts vary. The 'low risk' flavor allows only a low bet of 1 coin whereas the 'high risk' flavor allows variable bets from 1,3,5,7, up to 9 coins. The probability of selecting a 'low risk' flavor is always the same (3:3), or higher (4:2 and 5:1) than selecting the 'high risk' flavor.
In each trial, participants are presented with a cake and the two bet boxes. After a 3-second delay, a "?" appears between the bet boxes which signals that participant can make their choice by selecting one of the two bet boxes. Participants get 2 seconds to make their choice, otherwise the script will assume that the losing flavor would have been chosen. A computer animation plays the cake slice selection, and participants' totals are updated according to their choices.
Participants run 15 practice trials to get familiar with the task. The test runs 270 trials, each combination of the 3 slice distributions (3:3, 4:2, 5:1) by the five different variale risks (1,3,5,7,9) is run 18 times. A rest trial is run after 90 trials.
What it Measures
The Cake Gambling Task (CakeGT) measures decision-making, risk-seeking, and impulsivity under explicit risk conditions
Psychological domains
- Decision-making: Response to potential rewards and losses over time
- Risk-taking: Preference for high-reward/high risk or low-reward/low-risk
- Executive Control: The ability of our prefrontal cortex to override automatic, reward-seeking behavior
Main Performance Metrics
- Proportion Of Risky Choices: the proportion of selecting the risky choice depending on probability
Psychiatric Conditions
The Cake Gambling Task has been studied with the following patient groups
- Attention-Deficit/Hyperactivity Disorder (ADHD)
- Autism Spectrum Disorder (ASD)
- Orbitofrontal Cortex (OFC) Damage
The Cake Gambling Task, a child-friendly variation of the Cambridge Gamble Task developed by Van Leijenhorst et al (2008).
References
Search Google Scholar for peer-reviewed, published research using the Inquisit Cake Gambling Task.
Van Leijenhorst L, Westenberg PM, & Crone EA. (2008). A developmental study of risky decisions on the cake gambling task: age and gender analyses of probability estimation and reward evaluation. Dev Neuropsychol, 33(2):179-96
Van Leijenhorst L et al (2010). Adolescent risky decision-making: Neurocognitive development of reward and control regions. NeuroImage, 51, 345–355