Procrastination and learning about self-control

We study a model of task completion with the opportunity to learn about own self-control problems over time. While the agent is initially uncertain about her future self-control, in each period she can choose to learn about it by paying a non-negative learning cost and spending one period. If the agent has time-consistent preferences, she always chooses to learn whenever the learning is beneficial. If the agent has time-inconsistent preferences, however, she may procrastinate such a learning opportunity. Further, if her time preferences exhibit inter-temporal conflicts between future selves (e.g., hyperbolic discounting), the procrastination of learning can occur even when the learning cost is zero. The procrastination also leads to a non-completion of the task. When the agent has multiple initially-uncertain attributes (e.g., own future self-control and own ability for the task), the agent's endogenous learning decisions may be misdirected - she chooses to learn what she should not learn from her initial perspective, and she chooses not to learn what she should.