Simulation components: a quick reference
CESURA Simulation Components Reference Guide
This guide provides an overview of the core building blocks of CESURA simulations. Understanding these components helps you review simulations generated by Professor CESURA, build custom simulations, optimize learning outcomes, and ensure quality across all elements.
A CESURA simulation consists of interconnected components organized around three phases: character and scenario creation, assessment and feedback mechanisms, and post-simulation reflection.
Part 1: Persona and Scenario
The foundation of any simulation is the persona (character) and scenario (context). Together, they create the world in which learners engage and practice.
Persona — The Character
The persona is the character with whom participants interact. A well-crafted persona is believable, contextually appropriate, and serves the learning objectives.
A complete persona includes basic information (name, age, gender, and optionally how hard the character is to persuade), a detailed character profile describing the person's background, profession, role, personality, and communication style, and example sentences that demonstrate the character's distinctive voice and patterns of speech.
When creating a persona, ground it in a specific role and context rather than creating a generic character. Include psychological depth—strengths, vulnerabilities, decision-making patterns, and areas where the character might be defensive. Make the persona believable through specific details, not stereotypes. The character should have internal logic that explains their behavior, even if they are difficult or challenging.
The example sentences are crucial. They should demonstrate the character's vocabulary, sentence structure, values, and priorities. They show how the character handles different topics and emotional situations. They guide the AI in generating authentic dialogue that sounds like a real person, not a generic response.
Scenario — The Situation
While persona answers "Who is the character?" the scenario answers "What is the situation?" and "Why does it matter?"
A complete scenario includes a clear title that captures the core situation, a scenario story that explains the immediate context (when, where, what triggered the conversation), the stakes (why this conversation matters, what could be lost or gained), the character's position (what they expect, hope for, assume, or worry about), and the participant's challenge (what they need to accomplish, what makes it difficult, what success looks like).
The scenario should establish urgency and relevance. Rather than abstract practice ("practice giving feedback someday"), create a specific context ("it's the quarterly review meeting, and this person's performance is below expectations"). The best scenarios create genuine ambiguity—there is no single obvious "right" answer, which drives meaningful learning.
Both the participant and the character should have something at stake. The persona is not just there to be practiced upon; they have their own concerns, hopes, and vulnerabilities.
Theory — Professional and Academic Context
The theory component provides the intellectual foundation for the simulation. It establishes what "good" looks like from a professional standpoint and provides a framework for evaluating performance.
Theory should include relevant models and frameworks that apply to the scenario, professional standards and best practices, common pitfalls and learning objectives, and evidence or research supporting the approaches being taught.
Provide enough theory to orient learners but not so much that it overwhelms. Make theory directly relevant to the scenario and feedback criteria. Explain the reasoning behind best practices, not just what they are. Balance idealism with acknowledgment that real situations are messy. Provide a lens for evaluating decisions rather than a script.
Documents — Supplementary Resources
The documents component allows you to upload files that give the AI additional context and knowledge during the simulation. These might include organizational materials (policies, job descriptions), historical data (previous reviews, email chains), or reference materials (guidelines, regulatory requirements).
Documents add realism and ground the character's answers in real material. Learners do not see these files; to hand learners a document, add a Document step in Activity steps instead. Limit documents to essential materials that directly support learning objectives. You can upload documents once the simulation has been saved.
Introduction — Pre-Simulation Content
The introduction is made of the steps you place before the Simulation marker in Activity steps. It prepares learners before the simulation begins. It can include context-setting information, background needed to understand the scenario, explanation of what will happen, baseline assessment of knowledge or attitudes, and psychological safety building.
Structure introductions clearly. Use markdown to organize content into sections. Include learning objectives so learners understand what they will practice.
Part 2: Assessment and Feedback Mechanisms
Feedback Report Criteria
The feedback report uses criteria that the AI evaluates learners against. Each criterion should be observable in the simulation transcript, specific and focused on one skill or behavior, measurable (the AI can determine how well it's demonstrated), relevant to learning objectives, and clearly defined so the AI understands what to assess.
Good criteria have these characteristics: "Active Listening" is concrete and observable, while "Good Communication" is too vague. "Does the learner ask clarifying questions to understand the other person's perspective?" is specific and measurable. "Recognizes emotional cues and responds appropriately" is focused on one skill. "Asks about the customer's budget constraints" is directly observable in the transcript.
Each criterion is scored independently. A learner can score well on one criterion while scoring lower on another. This is intentional—it shows learners they can excel in some areas while having growth opportunities in others.
Most simulations work well with 3-5 criteria. Fewer criteria (2-3) work for simple scenarios. Six to eight criteria suit complex scenarios with longer interactions. Avoid more than 9 criteria unless the simulation is very long; too many becomes overwhelming.
Report Display
The Report Display page controls whether learners see the feedback report and how it is displayed. These settings affect the learner's view only; instructors always see the full report. Use Show to learner to show or hide the feedback report entirely. Show Summary Grade and Show All Criteria Grades each take a display mode: No grades (qualitative feedback only), Numeric grade (0-100), Clustered grades with positive names, or Pass/Not Passed.
Configure whether criteria appear expanded (all visible immediately) or collapsed (expandable sections). The Feedback Chat, which lets learners ask follow-up questions about their feedback, is switched on separately (see The Feedback Chat below).
Report Steps
Report steps structure the feedback experience as a sequence. They are the steps below the Simulation marker in Activity steps. Built-in steps include Start (opening message and context), Feedback (evaluation results), Evaluator Chat (the Feedback Chat, for deeper analysis through dialogue), and End. You can add custom steps with your own markdown content, including reflection questions, next steps, resources, or encouragement, as well as quizzes and forms, and drag them into order.
A typical flow is Start (orientation), Feedback (results), custom reflection step (deeper learning), and optional Evaluator Chat (extended dialogue). More than 4 steps becomes lengthy.
Part 3: Post-Simulation Reflection and Dialogue
The Feedback Chat
After completing a simulation, learners can engage in a processing conversation (the Feedback Chat) with an evaluator persona. This creates a reflective dialogue where learners can discuss their performance, ask questions, and receive deeper analysis.
On the Feedback Chat page, configure the evaluator's profile (Evaluator Name, Profession, Gender), provide Feedback Chat Instructions that guide the conversation, and turn on Enable Feedback Chat Phase if you want learners to be able to ask follow-up questions about their feedback.
When to enable: for complex skills requiring nuanced discussion, advanced learners wanting deeper analysis, scenarios emphasizing continuous improvement, and programs where coaching is a learning objective. When to disable: for simple assessments, limited time for learner reflection, younger or less experienced audiences, and compliance training with straightforward criteria.
Best Practices for All Components
Align all components with your learning objective. Each component should directly support what you want learners to improve. Avoid elements that are "nice to know" but not central to your goals.
Prioritize depth over breadth. A simulation with one focused scenario and clear learning objectives is more effective than one trying to cover multiple situations. Single-scenario focus allows you to create realistic complexity and meaningful learning.
Ground components in reality. Personas should have believable backgrounds and motivation. Scenarios should reflect actual workplace or professional situations. Theory should connect to real professional practices.
Make components work together coherently. The theory should inform feedback criteria. Feedback criteria should directly assess what the theory prepares learners for. The scenario should create situations where the theory and criteria matter.
Write for clarity. Avoid jargon without definition. Use specific, observable language. Describe behaviors and outcomes, not abstract qualities. Make descriptions concrete rather than theoretical.
Test your simulation. Complete it as a learner would. Review the feedback generated. Verify that it is fair, clear, and helpful. Ask pilot learners for feedback on clarity and helpfulness.
CESURA AI — User Guide | v1.0