CEPD · Module 4 — Scenario, dialogue, and avatar direction
Parallel choices, distinct feedback, signed preferred path, RTL dialogue tested.
About 16 min
Practice in the lesson player
Sorting, matching, quizzes, and the module gate stay behind a free account so scored answers are not public. The teaching notes below are the same cited curriculum.
01
Check | Fail symptom
Choices parallel in length and tone | The longest or most polite option is always right
Wrong branches teach | "Incorrect" only
RTL branch layout tested | Back button on the wrong edge, options in wrong order
Consequence before delight | Celebration on wrong clicks
Help type fits the beat | Policy dump inside the mentor bubble
Disclosure present | Synthetic presenter not declared
02
Before publish, read the three branch labels aloud. If one is a paragraph and the others are four words, learners will pick the long one or the only polite one. That is a length and tone defect, not a measure of judgment. Each wrong branch needs feedback that names the gap so the next attempt can change.
03
After a pilot with 30 or more learners, look at how often each option was chosen. If over 90% choose the preferred path on the first attempt, the scenario is probably too easy or cued. If a wrong option is chosen by almost nobody, it is not plausible; replace it with a real error. If the strongest performers (by gate score) pick a wrong option as often as the weakest, the preferred path may be debatable; take it back to the SME. These are the same item statistics used in test development, applied to branches.
04
Run the QC table on your scenario script. Plan a pilot with at least 30 learners and write the thresholds from slide 4 you will use to decide.