This simulation demonstrates how digital assessments can be used in online learning environments, modeling the exact mechanics a computerized adaptive test (CAT) runs on: a bank of calibrated questions, each with its own item-response-theory difficulty and discrimination curve, and an engine that always selects the single most informative next question for the current ability estimate. Set a hidden "true" ability for the simulated learner, step through the test one item at a time (or auto-run it), and watch the 3D item bank respond — curves turn green or red as they're answered, the ability estimate θ̂ and its standard error update live by maximum-likelihood after every response, and the engine's next pick is highlighted before you administer it. A toggle lets you compare the adaptive "maximum information" selection rule against picking items at random, showing directly why adaptive scoring converges to an accurate ability estimate in far fewer questions than a fixed-form quiz.