Skip to main content

How do I run a study with sensitive or disturbing content?

You can run studies with sensitive or disturbing content on Prolific as long as you follow the rules in this article. If you have questions about your study, contact our support team using the button at the bottom right of this page.

You must never include content in your study that infringes a third party's copyright, rights in confidential information, image rights, privacy, or other rights. Your content also can't be defamatory, threatening, or harassing, can't encourage any crime, and can't be otherwise unlawful.

Checklist for researchers

Use this checklist for every study with sensitive or disturbing content:

  1. Get ethics approval before you publish if your institution or company requires it, such as from an Institutional Review Board (IRB), research ethics committee, or independent commercial ethics board

  2. Add content warnings to your study details, description, and consent form

  3. Use the Harmful Content audience filter so only willing participants see your study, and the Deception filter if your study uses deception

  4. Get informed consent in your study platform, such as Qualtrics or Gorilla, before participants share any data

  5. If you use generative AI, label tasks by type and severity, and let participants skip tasks without penalty

  6. Debrief participants before they complete their submission, and point them to support resources

Plan your study

What counts as sensitive or disturbing content?

This article focuses on content that participants view. For help with asking questions about sensitive topics, read our guidance on asking sensitive questions. For our rules on collecting personally identifiable information, read our policy on collecting personally identifiable information.

Sensitive content covers deeply personal topics or potentially distressing issues. This can include scenarios, stories, or images about health, trauma, or other upsetting subjects.

Disturbing or explicit content includes anything you show participants that's graphic, violent, sexual, or egregious. It can take many forms:

  • Text, such as derogatory words

  • Images and videos, such as sexually explicit content

  • Audio, such as profanity

Check your study for sensitive content

To check whether your study includes sensitive content, preview it and review all of its materials. We recommend checking your study against common content policies, such as Google's content policies.

Sensitive content is often obvious, but it’s easy to get used to material you’ve designed yourself. After seeing it many times, you might not notice that it could upset someone.

Review every part of your study as if you’re seeing it for the first time, or ask a colleague or peer to look it over and give you feedback.

Weigh the benefit against the potential harm

Take extra care with sensitive or disturbing content: review anything that may distress participants before you publish it, and weigh the benefit of your research against the potential harm to participants.

We want every participant to have a safe experience. If your study includes sensitive or disturbing content, follow the guidance below so it doesn't offend or upset participants.

Get ethics approval

If your institution or company requires approval from an IRB or equivalent ethics board, make sure it's in place before you publish. Your consent process is usually part of what the board reviews, so make sure the consent form participants see matches the one the board approved.

Studies with sensitive content often need extra scrutiny. Include your content warnings, filters, and debrief plan in your submission so the board can review the full participant experience.

If you change your consent form or process after approval, check with your ethics board first. Your ethics board may require you to resubmit your study for approval if you change what participants see or how you collect consent.

Our blog post on How to write the perfect consent form for research can help. If you're unsure about your consent process, contact our support team.

Set up your study

Add content warnings

A content warning tells participants there may be upsetting material before they see it, so they can decide whether to take part. You must add one whenever your study includes sensitive content or there's any risk of explicit or disturbing content.

Add your warning in the study setup, under "Labels" in study details. There are two parts:

  1. Select a content warning type: "Exposure to explicit or disturbing content" or "Sensitive topics"

  2. Add content warning keywords to say exactly what participants will see

Eligible participants see the warning on the study details page, so they can make an informed choice about taking part.

Write content warning keywords

The content warning type tells participants the category of content. Keywords tell them what's actually in your study. The keywords box is optional in the study setup, but we expect you to use it whenever your study includes explicit or disturbing content. A category on its own doesn't give participants enough detail to decide if they want to participate.

To add keywords, type them in the keywords box and separate each one with a comma. For example: graphic injuries, animal abuse, sexually explicit images.

Follow these rules when you write your keywords:

  • Name the specific content participants will see or hear

  • Describe severe content, such as graphic injuries or animal abuse, explicitly

  • Use plain words that participants will understand immediately

  • Keep each keyword to a word or short phrase

A vague content warning isn't enough. "Moral and immoral behavior, emotional video clips" doesn't tell participants what they'll see, so they can't make an informed decision about whether to take part.

Add the warning to your study description

Include the warning in your study description too. Aim for a balance: give participants enough detail to understand the warning, but not so much that the warning itself becomes upsetting. In your description, you can keep the warning broader. For example: "In this study, you will see violent images." Follow this with a more detailed warning in your consent form.

In short, your keywords are specific, your description is broader, and your consent form is the most detailed.

For more help, read our blog post on How to write the perfect study description.

Edit content warnings

You can edit the content warnings on your study whether it's active or paused.

To do this:

  1. On the project page, click the three-dots next to the study you'd like to edit

  2. Select "Update content warnings"

  3. Select a content warning type, and add keywords separated by commas

  4. Click "Update content warnings" to save and apply your changes

If we pause your study

Make sure you've added your content warning before you publish. If we find your study contains sensitive content but doesn't have one, we may pause it and put your account on hold. You'll need to add a warning to your study and let our support team know, so that your study can continue.

Use audience filters

Harmful Content filter

If your study includes sensitive or disturbing content, use our Harmful Content audience filter. You'll find it in the "Participation on Prolific" category. The filter asks participants: "Are you willing to participate in studies which may contain harmful, graphic or upsetting content?"

If you select "Yes," we only show your study to participants who've said they're comfortable viewing this type of content. That keeps your study in front of willing participants. When you add your filters, you'll see a live count of participants who match your criteria. If fewer than five qualify, we hide the exact number to protect participant privacy.

Sexually explicit content

The filter question doesn't mention sexually explicit content. Participants who answer "Yes" may not have considered it when they chose their answer, so the filter alone isn't enough.

If your study includes sexually explicit content:

  • Use the Harmful Content filter to reach willing participants

  • Name the content in your content warning keywords, for example: sexually explicit images

  • Mention it in your study description and consent form

This makes sure participants know exactly what they'll see before they decide to take part.

Deception filter

Deception means intentionally withholding or misrepresenting information about your study's purpose, procedures, or participant interactions. If your study uses deception, you must use our Deception filter. You'll find it in the "Participation on Prolific" category, alongside the Harmful Content filter.

The filter asks participants: "Would you be happy to take part in a study where you are intentionally given inaccurate information about other participants and the study? You would be debriefed after the study." Participants who say "Yes" should be comfortable with deception.

The Deception filter doesn't replace content warnings. Participants who say "Yes" have agreed to receive inaccurate information, not to see explicit or disturbing content. If your study includes this content, you still need a content warning and the Harmful Content filter.

If telling participants about your study design would affect your results, you can hold back details of your study's purpose until the debrief. This counts as deception, so you must use the Deception filter. You can't hold back the warning about the type of content they'll see.

Run your study

Get informed consent

Make sure participants understand your study and agree to take part. This matters for all research, and even more for studies with sensitive content.

In every study on Prolific, ask participants to consent in your study platform before they share any data, including demographic questions. A simple way to do this is to put your consent form on the first page of your study.

Informed consent means participants fully understand what you're asking of them. That goes beyond agreeing to take part in a research project. For sensitive content, participants need to know exactly what they'll do and see, and whether it could upset them. The UK government has a useful resource on what you need to gain informed consent.

Use generative AI with care

Take particular care with generative AI because, unlike traditional content, you can't fully control what it shows participants.

If your study uses generative AI, follow these steps:

  • Add a content warning to your consent form explaining that participants may see text, video, or audio containing sensitive or disturbing content

  • Label tasks by type and severity

  • Let participants skip any task without penalty

  • Debrief participants thoroughly after the study, before they complete their submission

Labeling by type and severity tells participants what they might see and how intense it could be. For example, you could label a task by type, such as violent or sexual content, and by severity, such as mild, moderate, or extreme. Participants then choose which tasks to complete, and skipping a task must never count against them.

If telling participants the content is AI-generated would affect your study design, you need the Deception filter.

You're still responsible for any disturbing or explicit content that generative AI shows participants.

Debrief participants

You must debrief participants after every study, before they complete their submission. A good debrief gives participants closure. It tells them what the study was for, what happens to their data, and where to go if the content affected them.

Cover the key points

Your debrief should cover four things:

  • The purpose of your study

  • How participants' data will contribute to your research

  • Who participants can contact with questions about the study

  • Where to find support if the content upset them

Write each point in plain language, in a few short sentences. Avoid academic or research terms that participants may not understand. For help with the last point, see Point participants to resources below.

Explain the purpose of your study

Tell participants what your study was for and how their data contributes to your research. If they saw sensitive or disturbing content, explain why it was necessary for the research.

Reveal what you held back

If your study used deception or withheld details, such as its true purpose, reveal what you held back and explain why. Participants should understand the reason, not just the fact. If you used generative AI, tell them the content was generated and that you couldn't fully control what it showed.

Explain what happens to data

Tell participants what happens to their data. Cover how you'll store it, how long you'll keep it, and whether you'll anonymize it. Make sure this matches your consent form. If participants can ask you to remove their data, explain how.

Share who to contact

We recommend asking participants to contact you through Prolific's internal messaging system, which protects their anonymity. You should also give them an independent contact outside your research team.

If you’re an academic researcher, this is the Institutional Review Board (IRB) or ethics board that approved your study. A board gives participants a neutral contact if they have questions, concerns, or complaints about their rights, safety, or treatment during the study. Include the board’s name, the contact details it provides for participants, and your study title or reference number.

If you’re an industry researcher, include the name, role, and contact details of someone outside your research team. This could be the ethics board that approved your study, a department lead, or a team lead.

Point participants to resources

Some participants may feel upset after your study, even if they chose to take part and you gave them a clear content warning. A few minutes of support at the end can make a real difference.

Include these in your debrief:

  • A reminder that it's normal to react to difficult content

  • Links to support resources that fit the content in your study

  • Your contact details, or a link to message you through Prolific

  • Details for your ethics board, if one approved your study

You don't need to share every resource, only the ones that fit your content. For example, a study with images of violence needs different support from a study about bereavement. Check the participant version of this article in our Help Center for resources you can share. You can also add other trusted resources you know of, such as national helplines or mental health charities.

Keep your wording calm and supportive. Participants should feel cared for, not alarmed. Make sure your resources work for participants in every country you recruit from, as helplines and services differ by region.

Get more help

If you have any further questions about sensitive or disturbing study content on Prolific, contact our support team using the button at the bottom right of this page. We're here to help.

Did this answer your question?