General Manager
Before I ask what a simulation should look like, I want to know which decision would make it worth building. A polished room and a responsive character can still leave the learner with very little to do.
In Altaius's Practice-Native Learning methodology, the scenario is organised around action, consequences and feedback. The design starts with a task someone may need to perform, not with a list of features the software can display.
“Leadership under pressure” is a theme. “Decide whether to commit the team to a date before capacity has been checked” gives the designer something observable. The learner can ask for information, qualify the promise, negotiate a different commitment or ignore the uncertainty.
Here is an illustrative six-field record I would use. The number of fields is a design choice, not a research finding.
This record gives the team a common object to review. A subject specialist can challenge realism, a learning designer can challenge the target and an evaluator can challenge what the record actually supports.
The customer should not become angry simply because the learner failed to use a preferred phrase. A change in the interaction should have a plausible relationship to what happened. If the learner made an unsupported promise, the scenario might later reveal a capacity conflict. If the learner asked a relevant question, information should become available under a defined rule.
I would document those relationships before polishing the dialogue. Some decisions may have several defensible answers. The system needs to recognise the constraints and reasoning behind an answer rather than treating one script as the only route to success.
Surprise can make a scenario interesting, but arbitrary penalties make it difficult to learn. A learner should be able to review why a consequence occurred, even when they disagree with the design.
A beginner may need a worked example, an explanation or a prompt before attempting the task independently. I would make that support available and record its use. A coached attempt can be valuable practice while providing different evidence from an unassisted attempt.
The leadership-training literature supports attention to needs analysis and programme design, not just delivery media. The testing standards also require a clear account of intended score interpretations and uses. Neither source validates this particular six-field worksheet. They help explain why I want the objective and evidence boundary written down. [1, 2]
I would run several deliberately different paths: a competent approach, a plausible novice error, a reasonable alternative and an attempt that exploits the system's wording. Then I would compare the resulting feedback with the actual decision record.
If two sensible approaches receive incompatible feedback, the review should examine the rules before blaming the learner. If a character's own words appear as evidence of the participant's performance, the record needs correction. If the scenario gives no opportunity to demonstrate a skill, absence of evidence must not become a low score for that skill.
For me, a useful scenario is one in which the learner can inspect a consequential decision and try a better approach. The setting helps make that possible. It should never be asked to stand in for the learning design.
Read the Altaius definition of Practice-Native Learning
Previous in this series: Practice-Native Learning and the Evidence for Deliberate Practice
