Before you decide to take part in this study, it is important for you to understand why the research is being done and what it will involve. Please take time to read the following information carefully and take time to decide whether or not you wish to take part.

Purpose of the study

This research aims to investigate whether simple and quick training can improve people's performance in detecting deepfake facial images. The study will be completed by summer, 2026.

Why have I been chosen?

You are invited to participate in this follow-up study because you previously took part in an earlier study conducted by our research team 5 months ago. We are now evaluating whether the effects of that earlier training are maintained over time.

Do I have to take part?

Taking part is entirely voluntary. You are free to decide whether or not to take part. If you do decide to take part, you will be asked to give your consent. You are free to withdraw at any time before you submit the final page of the study, including after viewing your performance summary, without giving a reason. Once you have submitted the study anonymously, we may not be able to identify and remove your data because no personal identifiers are collected. A decision to withdraw, or a decision not to take part, will not affect you in any way. Refusal or withdrawal will involve no penalty or loss, now or in the future.

What will happen to me if I take part?

If you agree to take part, you will complete an online study on the website you are currently visiting. The study will take approximately 15-20 minutes in total and will proceed as follows:

• Deepfake Detection Session 1: You will participate in the first detection task, which involves identifying real and fake facial images presented one-by-one and side-by-side. These images will include faces of different genders, ages, and ethnicities, including some images of children. After this session, you will be asked to report your confidence and stress level.
• Break: There will be a short break between the two detection sessions.
• Deepfake Detection Session 2: You will participate in the second detection task, which involves identifying real and fake facial images presented one-by-one.
• Follow-up Questionnaires: You will be asked to answer a few questions about your experience at the end of the study.
• Final Results: At the end of the study, you will receive a private performance summary based on your results from the two detection sessions.

The disadvantages and/or risks in taking part are considered low and comparable to those encountered in everyday life.

• Time Commitment: 15-20 minutes of your time.
• Low to Mild Psychological Stress: The task of detecting deepfake images could potentially cause low to mild frustration or self-doubt, especially if your performance is not as high as you might hope. Your performance summary will be shown to you privately on the study website. You are free to withdraw at any point before submitting the final page of the study.

What are the possible benefits of taking part?

By taking part in the study, you will learn more about deepfake facial images and how to better distinguish them from real ones. Such skills can help you avoid becoming a victim of online harms generated by deepfake facial images.

You will receive a base payment of £3 for taking part in the study. In addition, participants whose performance ranks in the top 30% will receive a performance-based bonus: those in the top 30% will receive an extra £3, while those in the top 10% will receive an extra £6.

Beyond the benefits to yourself, your participation will help the researchers understand the effectiveness of the training on deepfake detection, which can benefit a wider group of people and organisations.

Will my taking part in this project be kept confidential?

We will not collect your name, contact details, or other directly identifying information through the study website. The data collected for analysis will be anonymous and identified only by a study code. Because the study data are anonymous, members of the research team will not be able to link the analysed responses back to you personally.

To comply with data protection legislation, the University of Cambridge provides general information about how personal data is used: https://www.information-compliance.admin.cam.ac.uk/data-protection/research-participant-data.

Anonymous data will be collected via the testing website, initially stored on its server based in the UK, and then transferred to secure research storage accessible to the research team. Fully anonymous data may also be shared with other researchers or released as a public research resource, as described below.

What will happen to the results of the research project?

The results of the research are expected to be published, presented and communicated at different venues such as scientific journals and conferences, preprint servers, talks and websites of members of the research team. Results are normally presented in terms of groups of individuals. If any individual data are presented, the data will be totally anonymous, without any means of identifying the individuals involved.

A copy of the published results and links to any future publications will be made available on the websites of members of the research team. As we are not collecting your contact information, we cannot inform you directly, so you are encouraged to check the websites in the future for updates. The researchers' website can be found via Google searches.

The anonymous research data may be released as a public resource for researchers, educators, policymakers and other stakeholders to use. This is to ensure transparency and allow for further research.

Who is organising and funding the research?

The research is organised by a team of international researchers from University of Cambridge (UK), Xinjiang University (China), Hunan University (China), Max Planck Institute for Security and Privacy (Germany), and University of Kent (UK). The whole research team includes the following researchers:

• Luning Sun and David Stillwell, University of Cambridge, UK
• Zhiqing Guo, Lixin Jia, Guofeng Yu and Liejun Wang, Xinjiang University, China
• Gaobo Yang, Hunan University, China
• Xiaowei Chen and Thorsten Holz, Max Planck Institute for Security and Privacy, Germany
• Shujun Li, University of Kent, UK

All researchers are funded by their own institute to collaborate on the research, which receives funding from Xinjiang University, China and Max Planck Institute for Security and Privacy, Germany for the experiments.

Ethical review of the study

The project has received a favourable opinion from the Research Ethics Review Group, Cambridge Judge Business School, University of Cambridge, UK (Reference Number: 25-46).

Contact for further information

For questions about the research, please contact the researcher: Dr Luning Sun, University of Cambridge (l.sun@jbs.cam.ac.uk).

Thank you for taking the time to read this information sheet and for considering taking part in this research.