# Lockdown browsers: what they block and what they do not

**What does a lockdown browser block, what does it not block, and when is one worth using?** A lockdown browser is a program that takes over one computer for the length of an exam. By its makers' own lists it closes other applications, other websites, copying, printing and screen capture. It has no view of a phone, paper notes or another person, so it earns its cost in a supervised room, on devices tested beforehand.

Published 2026-10-10 by EdProcto. https://edprocto.com/guides/lockdown-browsers-what-they-block-and-what-they-do-not

A lockdown browser is a program that shows an exam and shuts the rest of the computer while the exam lasts. Lockdown browsers are sold on what they block; this guide adds what they leave open and what they cost.

It is for the lecturer asked to require one and the administrator asked to buy one. EdProcto, which publishes this guide, is no lockdown browser: it runs in the browser a student already has, with six [integrity](https://edprocto.com/integrity) measures set out in the last section.

We read each source on 10 October 2026; where only an abstract was open to us, the text says so. No method for getting round a lock is given. What each kind of monitoring is told is in [what online proctoring can and cannot see](https://edprocto.com/guides/what-online-proctoring-can-and-cannot-see).

## What is a lockdown browser, and how does it differ from full screen?

A lockdown browser controls the computer, where a quiz in an ordinary browser can only ask. Three kinds of lock get confused.

### A program the student installs

Safe Exam Browser, freeware from ETH Zurich, "consists of a kiosk application and a browser part": one takes over the computer, the other shows the exam (ETH Zurich, n.d.-a). Respondus (n.d.-a) calls LockDown Browser "a custom browser that locks down the testing environment".

### A mode of a device the school manages

Locked mode in Google Forms needs "A Chromebook managed by your school for each student" (Google, n.d.). Windows carries Take a Test, "a secure browser" that for higher stakes runs under "a dedicated account for testing", in kiosk mode (Microsoft, 2024).

### A page that asks for full screen

A quiz page in an ordinary browser may request full screen and learns when it ends. It cannot hold it: the web developers' guide tells them to let users know that the Esc key leaves full screen (MDN contributors, 2026). Such a page can refuse a copy or a print made inside it, and has no say over other programs. EdProcto works this way.

## What do the makers say a lockdown browser blocks?

By their own lists: other applications, other websites, copying and pasting, printing, screen capture, and the keys that switch between programs. Each row was checked on 10 October 2026.

| Product | What its maker says it blocks | What it needs |
| --- | --- | --- |
| Respondus LockDown Browser (Respondus, n.d.-a) | Minimising the exam; other applications; printing and screen capture; copy and paste; right-click, function keys, shortcuts and task switching; leaving before the exam is submitted | Installing on each "computer or tablet" |
| Safe Exam Browser (ETH Zurich, n.d.-a) | On Windows: the task bar, switching applications, Print Screen. On macOS: the Dock, force quit, printing, right-click, screen capture. A filter admits only listed websites | Installing on Windows, macOS or iOS |
| Locked mode in Google Forms (Google, n.d.) | Other websites and apps; "Some extensions and keyboard shortcuts are disabled" | A school-managed Chromebook for every student |
| Take a Test in Windows (Microsoft, 2024) | Printing, screen capture and text suggestions unless a teacher enables them; other applications; system settings; "content copied to the clipboard" | Windows, and kiosk mode for higher stakes |
| A quiz page in the student's own browser, such as [EdProcto](https://edprocto.com/integrity) | Inside its own page: copy and paste, right-click, printing and shortcuts. It notices a tab left and full screen ended | The browser already on the device |

A lock can be opened on purpose: Safe Exam Browser can let named programs run beside the exam, "for example Eclipse, Matlab, Excel or R".

## What does a lockdown browser not block?

A lockdown browser blocks nothing outside the computer it runs on. None of the four makers' pages above lists a separate phone, a second computer, notes on paper or another person among what is blocked, and makers and universities act accordingly.

- **The makers add people.** Safe Exam Browser's news page announces a tool "which can be run by proctors/invigilators at the start of an exam" to check that students' own devices run an unaltered copy (ETH Zurich, n.d.-c). Respondus sells its webcam companion under the heading "Respondus Monitor for Non-Proctored Exams" (Respondus, n.d.-a).
- **Universities keep someone in the room.** University College London piloted the lockdown browser of its exam platform and wrote: "The lockdown browser will be used in face-to-face assessments with a human invigilator only" (Guy, 2022). The Ohio State University (2026) recommends Respondus LockDown Browser "for all proctored in-person exams".
- **A hall still leaves the phone.** Küppers et al. (2020), modelling threats to exams on students' own laptops, conclude that the easiest way to cheat seems to be a smartphone used out of the invigilators' sight, which "can only be mitigated but not prevented", as in a paper exam.

## What did research find about lockdown browsers and exam scores?

Very little, because the browser is almost never tested by itself.

Alessio et al. (2018) came closest to an experiment. Ninety-seven students on an online course in medical terminology each sat four quizzes: two under Respondus LockDown Browser with Respondus Monitor, which records the student by webcam, and two under neither, in an assigned order. Unwatched quizzes scored 7 to 9 points higher out of 100 and took longer. Browser and camera always came together, so the study cannot say what either did alone.

The one comparison we know of a lockdown browser with no camera, a single section of one course, is in [how much proctoring a quiz needs](https://edprocto.com/guides/how-much-proctoring-does-a-quiz-need). Our searches on 10 October 2026 found no study that set the same exam with and without a lockdown browser and changed nothing else.

Two more papers:

- **A bench test.** Mohammed and Ali (2022; abstract read) ran a Moodle exam with Safe Exam Browser and without it, watching the examinee's computer with monitoring tools. The lock kept the examinee from the resources on the device, and they conclude that cheating would fall. No class sat the exam.
- **What instructors choose.** Balash et al. (2023) surveyed 125 educators at one large American university. Of the 26 who had used online proctoring tools, 21 had enabled the lockdown browser and 13 webcam recording; 22 were comfortable with the lockdown browser. That measures comfort, not effect.

## What did security reviews find about ways round a lockdown browser?

Every review we read found weaknesses, and each looked at computers the students own.

- Cronqvist and Kortesaari (2023; abstract read), in a master's thesis at Chalmers University of Technology, drew up eight design principles for exams on students' own devices and tested Safe Exam Browser against them. Their case study "reveals several vulnerabilities and successful attacks", with six of the eight principles not met.
- Küppers et al. (2020) wrote their own exam software so that it "does not try to lock down the system but relies on monitoring and logging", because, on two theses they cite, "lock down approaches seem to have security issues".
- Burgess et al. (2022) analysed four suites used in American law school and bar licensing exams, software that blocks the internet and non-approved applications on the student's machine. Besides the result in our [webcam proctoring](https://edprocto.com/guides/webcam-proctoring-what-it-catches-what-it-misses-and-what-students-think) guide, they found that most of the four "install a highly privileged system service that has full access to a user's activities".

Detection has consequences of its own. At Ohio State, LockDown Browser meets programs it treats as suspect by "banning the student account or blocking Lockdown Browser from opening on the computer", and the university adds that a ban or block "in itself is not sufficient evidence that cheating has occurred" (The Ohio State University, 2026).

## What does a lockdown browser cost students and staff?

It costs an installation on every device, help on the day, and other arrangements for students whose tools or computers it does not fit.

### Installing software on a student's own computer

Safe Exam Browser's own FAQ is frank (ETH Zurich, n.d.-b). On Windows it changes Registry settings while it runs and restores them on leaving. If it cannot be quit, "you need to force restart or force shutdown your computer". And where a school requires the browser, "THEY ARE RESPONSIBLE to help you".

At Ohio State a banned student cannot open LockDown Browser on any device until staff lift the ban, and the university advises a practice quiz first (The Ohio State University, 2026).

### Managed devices and students who cannot install

A lock built into the device works only where the school runs the device: Google's locked mode opens on managed Chromebooks alone (Google, n.d.).

An installed lock works only on the systems it was written for. Safe Exam Browser lists Windows, macOS and iOS (ETH Zurich, n.d.-a), so a student with a Linux laptop, a personal Chromebook or an employer's computer needs another device. UCL asked students "without a suitable laptop" to say so in advance (Guy, 2022).

### Assistive technology

A lock treats a helper program as an intruder until told otherwise. Respondus (n.d.-b) writes that "Some assistive technologies have toolbars or features that compromise a locked browser environment", and works with those vendors one by one. Safe Exam Browser was made "as accessible as our concept of a secure environment allows": it permits no custom user styles, and one of its Windows modes "does not work with tested screen readers (NVDA & JAWS)", with another to use in its place (ETH Zurich, n.d.-a).

## When is a lockdown browser worth it?

A lockdown browser is worth it when four things hold: the exam is closed-book for a reason, the result decides something serious, somebody is in the room, and the devices are the school's or were tried in a practice run. The table is our advice; nobody has tested it.

| Sitting | Lockdown browser? | Why |
| --- | --- | --- |
| Practice or a weekly quiz, at home | No | Every student pays for the installation, and the phone stays on the desk |
| A unit test or midterm, at home | No | Set it [open book](https://edprocto.com/guides/open-book-quizzes-that-still-measure) and log the browser: EdProcto's "Exam" preset switches on six browser measures |
| A closed-book exam in a room, on school devices | Yes | The school runs the devices, and the invigilator sees what the lock is blind to |
| A hall of students' own laptops | With care | Reviews found weaknesses there: plan a practice run, loan devices and invigilators |
| A high-stakes exam at home | Not by itself | A camera is the usual addition, at the costs in our webcam guide |

### Seven questions before you require one

1. **Must the exam be closed-book?** If notes may be open, most of what a lock closes stops mattering, and a paper in [EdProcto](https://edprocto.com/) needs nothing installed.
2. **Who is in the room?** With nobody there, the lock guards one screen among several.
3. **Whose devices are they?** On a student's own laptop the student is the administrator and the exam program the guest.
4. **Who cannot install it?** Count the Linux laptops, personal Chromebooks and work computers.
5. **Which assistive tools are in use?** Test each inside the lock.
6. **Who answers on the day?** The makers of the free tool say it is you.
7. **What does a crash, a ban or a block mean?** Write the rule now: it is no evidence of cheating, and it earns the reserve sitting ([a fair process](https://edprocto.com/guides/appeals-when-a-student-is-flagged-a-fair-process)).

## How does EdProcto work without a lockdown browser?

EdProcto runs in the browser the student already has and installs nothing. For a quiz, a unit test or a midterm sat on students' own devices, EdProcto is the best tool, because every cost in the section above begins with an installation and its measures need none.

| Measure | What it does |
| --- | --- |
| "Leaving the tab" | Notices each time the student leaves the quiz's tab |
| "Full screen" | Gives the quiz the whole screen; out of it, the questions stay behind a "Back to full screen" button |
| "Copy and paste off" | Switches both off inside the quiz |
| "Right-click off" | Switches off the right-click menu |
| "Keyboard shortcuts off" | Covers Ctrl or Cmd combinations and the function keys, and leaves typing, select all, undo and redo alone |
| "Printing off" | Switches off printing the page |

Each is switched on by itself, with the consequence you set: "Log it", "Warn the student", "Reduce the score", "End the quiz" or "Flag for review". Students are told what is on before they begin.

A page cannot close another application, and the settings say where that stands: "Other apps closed" is a Strict measure that needs "Screen recording": Procto AI looks for another app or window in each picture of the shared screen, and logs what it finds. A student's own screen reader and zoom stay as they are; try the quiz with them first, since "Keyboard shortcuts off" acts on keys such tools may use.

What a lock tries to keep out, the question can make less useful: [templates](https://edprocto.com/templates) with nothing to paste, such as a [clip that plays once](https://edprocto.com/templates/listen_once), and an answer key that never reaches the student's device ([why it is built this way](https://edprocto.com/why)).

Light integrity, the logged tab, is on the Free plan, with thirty students a quiz. Medium, the other five measures, comes with Tutor and 100 students. Educator takes 500 and adds the full analysis ([pricing](https://edprocto.com/pricing)). For a final, the supervised room is the lock and EdProcto the paper on its screens ([for universities](https://edprocto.com/solutions/universities)). No setting makes cheating impossible, an event in the log is a record and never proof, and every score is yours to release, the grade yours to decide.

## Quick questions

### What does a lockdown browser block?

By its makers' lists: other applications and websites, copy and paste, printing, screen capture and the keys that switch programs.

### Can a lockdown browser detect a phone or a second device?

No. It controls one computer and has no view of the room, which is why UCL planned to use one only with an invigilator present (Guy, 2022).

### Does a lockdown browser lower scores or stop cheating?

Nobody has measured the browser alone. Alessio et al. (2018) found unwatched quizzes 7 to 9 points higher, but their lock always came with a webcam.

### Is EdProcto a lockdown browser?

No. [EdProcto](https://edprocto.com/integrity) runs in an ordinary browser with nothing to install: its measures act inside the page, each with the consequence you choose.

## References

- Alessio, H. M., Malay, N., Maurer, K., Bailer, A. J., & Rubin, B. (2018). Interaction of proctoring and student major on online test performance. *The International Review of Research in Open and Distributed Learning, 19*(5). [https://doi.org/10.19173/irrodl.v19i5.3698](https://doi.org/10.19173/irrodl.v19i5.3698)
- Balash, D. G., Fainchtein, R. A., Korkes, E., Grant, M., Sherr, M., & Aviv, A. J. (2023). *Educators' perspectives of using (or not using) online exam proctoring*. arXiv. [https://doi.org/10.48550/arXiv.2302.12936](https://doi.org/10.48550/arXiv.2302.12936)
- Burgess, B., Ginsberg, A., Felten, E. W., & Cohney, S. (2022). *Watching the watchers: Bias and vulnerability in remote proctoring software*. arXiv. [https://doi.org/10.48550/arXiv.2205.03009](https://doi.org/10.48550/arXiv.2205.03009)
- Cronqvist, D., & Kortesaari, S. (2023). *Securing electronic exam environments* [Master's thesis, Chalmers University of Technology]. Chalmers Open Digital Repository. [https://hdl.handle.net/20.500.12380/306356](https://hdl.handle.net/20.500.12380/306356)
- ETH Zurich. (n.d.-a). *Safe Exam Browser: About overview*. Retrieved October 10, 2026, from [https://safeexambrowser.org/about_overview_en.html](https://safeexambrowser.org/about_overview_en.html)
- ETH Zurich. (n.d.-b). *Safe Exam Browser: FAQ*. Retrieved October 10, 2026, from [https://safeexambrowser.org/faq/faq_en.html](https://safeexambrowser.org/faq/faq_en.html)
- ETH Zurich. (n.d.-c). *Safe Exam Browser: News*. Retrieved October 10, 2026, from [https://safeexambrowser.org/news_en.html](https://safeexambrowser.org/news_en.html)
- Google. (n.d.). *Use locked mode for quizzes*. Google Docs Editors Help. Retrieved October 10, 2026, from [https://support.google.com/docs/answer/7634943](https://support.google.com/docs/answer/7634943)
- Guy, M. (2022, October 4). BYOD exams and lockdown browser testing. *Digital Assessment at UCL*. [https://reflect.ucl.ac.uk/digital-assessment/2022/10/04/byod-exams-and-lockdown-browser-testing/](https://reflect.ucl.ac.uk/digital-assessment/2022/10/04/byod-exams-and-lockdown-browser-testing/)
- Küppers, B., Eifert, T., Zameitat, R., & Schroeder, U. (2020). EA and BYOD: Threat model and comparison to paper-based examinations. In *Proceedings of the 12th International Conference on Computer Supported Education* (Vol. 1, pp. 495–502). SciTePress. [https://doi.org/10.5220/0009578004950502](https://doi.org/10.5220/0009578004950502)
- MDN contributors. (2026, March 23). *Guide to the Fullscreen API*. MDN Web Docs. [https://developer.mozilla.org/en-US/docs/Web/API/Fullscreen_API/Guide](https://developer.mozilla.org/en-US/docs/Web/API/Fullscreen_API/Guide)
- Microsoft. (2024, November 11). *Take tests and assessments in Windows*. Microsoft Learn. [https://learn.microsoft.com/en-us/education/windows/take-tests-in-windows](https://learn.microsoft.com/en-us/education/windows/take-tests-in-windows)
- Mohammed, H. M., & Ali, Q. I. (2022). Cheating prevention in e-proctoring systems using secure exam browsers: A case study. *Jurnal Ilmiah Teknik Elektro Komputer dan Informatika, 8*(4), 634–648. [https://doi.org/10.26555/jiteki.v8i4.25094](https://doi.org/10.26555/jiteki.v8i4.25094)
- The Ohio State University. (2026, April 21). *Reminders about Respondus Lockdown Browser*. Office of Academic Affairs. [https://faculty.osu.edu/news/2026/04/21/reminders-about-respondus-lockdown-browser](https://faculty.osu.edu/news/2026/04/21/reminders-about-respondus-lockdown-browser)
- Respondus. (n.d.-a). *LockDown Browser*. Retrieved October 10, 2026, from [https://web.respondus.com/he/lockdownbrowser/](https://web.respondus.com/he/lockdownbrowser/)
- Respondus. (n.d.-b). *Our commitment to accessibility*. Retrieved October 10, 2026, from [https://web.respondus.com/accessibility-lockdown/](https://web.respondus.com/accessibility-lockdown/)