Bandit

Present a grid of colored squares representing multiple “bandits.” The participant clicks a square to choose one option. The stimulus records the selected index (choice) and a corresponding value (e.g., payout/label). The layout adapts to the number of bandits, and a short post-response delay can drive simple animations before the trial ends.

When To Use

• Multi-armed bandit tasks with several simultaneous options.
• Reward/payout choice screens where each option has an associated value.
• Visual grids of clickable alternatives.

Examples

A three-armed bandit task for reinforcement learning. The task is to choose between options, each with a different value.