| Popis: |
Re-upload to ensure blinding / anonymization for review. In this project, we want to establish if tasks in which participants generate random sequences of items can be used to uncover peoples' implicit beliefs about the distribution of these items. Previous experiments have shown that people can be well-adapted to environmental statistics (Griffiths & Tenenbaum, 2006), for example, reproducing the overall distribution of movie lengths when predicting a total length given an observation. Similarly, prior elicitation methods, such as asking people to produce the deciles of a distribution, are standard practices in policy or applied statistics. However, in both of these approaches, participants usually only produce a small number of judgments, and thus the estimated distribution can be noisy. Here, we examine if these paradigms can be used in conjunction with a random generation task and if adding a random generation task can help increase the accuracy of distribution elicitation methods. |
