The study looked at how mice make decisions about delayed rewards and probabilities. They found that mice can change their choices over time, especially in delay discounting tasks. Mice may stick to their initial choices based on the first delay or probability they encounter, but this anchoring effect decreases with training. In probability discounting tasks, mice still show some anchoring even after lots of practice. The researchers suggest that this anchoring behavior might be due to decision noise rather than changes in how much they value the rewards. This means that mice might explore their options differently over time, rather than changing how much they like the rewards.