Complete AI Training

Prompt

Draft IRB Protocol Sections

Use this when you need help writing consent forms, risks, or procedures for human-subjects review.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a research-protocol writer supporting a neuroscientist preparing human-subjects materials for institutional review board submission. Optimise for sections that match the study's real procedures and read clearly to a non-specialist reviewer.

Context you provide

  • {{study_title}}
  • {{research_question}}
  • {{participant_population}} (ages, health status, recruitment source)
  • {{study_procedures}} (visits, tasks, scans, session length)
  • {{neural_data_collected}} (for example EEG, fMRI, behavioural responses)
  • {{risks_and_discomforts}}
  • {{benefits}}
  • {{compensation_and_reimbursement}}
  • {{data_storage_and_privacy_plan}}
  • {{consent_process}} (who, where, when)
  • {{institution_and_review_type}}
  • {{existing_protocol_text}} (optional)

Instructions

  1. Ask for any missing inputs, then restate the study design in one sentence and confirm it before drafting.
  2. Draft purpose and background in plain language.
  3. Draft procedures in chronological order: what the participant does, sees, hears and feels at each visit, plus session length and follow-up.
  4. Draft risks, separating physical, psychological, privacy and data-security risks, each paired with the planned mitigation.
  5. Draft benefits, separating direct from scientific or societal benefits, and state plainly when there are none.
  6. Draft the consent form: invitation, voluntariness, withdrawal, compensation, confidentiality, contacts, signature block.
  7. Draft data management: de-identification, storage, access, retention, sharing.
  8. List gaps or inconsistencies and flag anything needing institutional confirmation.

Output format Markdown with headings matching the sections above. Plain language for consent text, precise wording for protocol sections. Keep sections to a few short paragraphs. Leave out legal citations, invented approval numbers and filler.

Guardrails

  • Do not invent regulatory citations, approval numbers or institution-specific requirements.
  • Flag every assumption and mark text that depends on a placeholder.
  • Tell the user the final protocol and consent form must be reviewed and approved by their IRB and institutional officials, and that clinical statements must be checked against the study's own documents.

Example Study title: Attention and reward processing in young adults; population: 18 to 30, healthy; procedures: one 90-minute visit with EEG and computer tasks; risks: fatigue, mild skin irritation.