PretectalStart here
Open the storyboard template. Write one sentence describing what this audience should be able to explain, then enter one supported claim in the evidence map before drafting slides.
Your finished output: Your package includes a source-linked storyboard, checked chart values, feedback for every choice, a static fallback, and a review record tied to the delivered version.
Pause and revise when: A result cannot be traced to its original location, the question goes beyond the study, or a learner cannot explain the finding and its limitation after the rehearsal. Resolve the gap before presenting that claim.
Before you start
Start with a question your audience needs to answer. Build a short teaching sequence around the evidence that changes that answer. Keep the study's limitations visible when you make the result easier to understand.
For: Clinical educators, speakers, and medical education teams.
The deliverable: A storyboard, claim-to-source map, and interactive evidence segment.
Gather these inputs
- The original paper and relevant supplementary material you may use
- One audience, one learning objective, and the presentation time available
- A blank evidence map and the provider's presentation requirements
The workflow
01Define the decision
Write the sentence: after this segment, the audience should be able to explain or decide ___. Choose a specific misunderstanding to address. Separate background context from the finding you intend to teach.
- AI can help
- Suggest three teachable questions using only the supplied material and audience description.
- You decide
- Choose the question yourself. Confirm the study actually addresses it; a plausible question is not necessarily an answerable one.
- Keep
- One objective and one audience prediction.
02Build the evidence map
Record each proposed claim alongside its exact table, figure, page, or section. Include the studied population, outcome, unit, comparison, and limitation. Mark missing support as unresolved.
- AI can help
- Draft an extraction table and flag details it cannot locate. Ask for source locations, not just a bibliography.
- You decide
- Open every cited location. Check adjusted versus unadjusted values, time points, confidence intervals, and whether the sentence overstates the evidence.
- Keep
- A claim-to-source map with explicit open questions.
03Create the teaching sequence
Sequence the segment as question, prediction, evidence, interpretation, and transfer. Give each slide or scene one job. Put qualification beside the finding it qualifies.
- AI can help
- Generate alternative outlines and shorten draft copy while retaining the linked claims and limitations.
- You decide
- Read the narrative aloud. Remove any transition that implies causality, generality, or certainty the source does not establish.
- Keep
- A storyboard with speaker notes and source IDs.
04Build one useful interaction
Let learners make a prediction, select an outcome, or compare interpretations. Use fixed, checked study values for an evidence explorer. Put fictional or synthetic material in a clearly separate exercise.
- AI can help
- Draft chart code, keyboard controls, and feedback using an explicit data table and answer key.
- You decide
- Check every value and interaction state. Verify labels, signs, units, zero points, accessible alternatives, and the feedback a wrong answer receives.
- Keep
- An interactive segment with a static fallback.
05Rehearse and hand off
Run the complete sequence with someone from the intended audience. Ask them to explain the finding and its limit. Record confusion, corrections, and the final approved version.
- AI can help
- Organize review notes and draft a change log or handout from approved content.
- You decide
- The presenter signs off on the final claims and delivery. Record remaining limits and the actual AI contribution.
- Keep
- Editable presentation, evidence map, disclosure, and review record.
A useful starting prompt
This prompt scaffolds a task. It does not verify its own output or replace the review described above.
Use material you are authorized to share. Replace the bracketed inputs and review the result against the original sources.
What changes in practice
A result slide shows a large benefit percentage without its comparison, timeframe, or population.
A learner predicts which outcome was measured, inspects the published comparison, and explains one limit on interpretation.
The change gives the audience a reasoning task and keeps the claim attached to its evidence. See the BLINK worked example for an implementation.
Inspect the BLINK worked example ↗Ready for handoff
- Each numerical claim resolves to the original source location.
- The displayed chart can be read without the animation or interaction.
- The answer key and every feedback state have been reviewed.
- The presenter can explain the methods, result, and limitation without relying on AI.
Sources and further reading
Source links checked October 2, 2026. Our production workflows are editorial recommendations; provider requirements apply to their own activities.
Walline et al. · JAMA · 2020
Primary source for the worked example; Table 2 contains the adjusted outcomes and contrasts.
ACCME · January 30, 2026
Guidance on disclosure, oversight, and independence in ACCME-accredited education.
PhET · University of Colorado Boulder
A design reference for inquiry, multiple representations, and feedback; not evidence that Pretectal improves learning.