Can Online Exams Detect Tab Switching? What the Log Records

The browser is always talking

The short answer is yes, in most cases, but the level of detection depends entirely on whether you are taking a standard quiz on a platform like Canvas or Moodle, or using specialized proctoring software like Proctorio or Respondus LockDown Browser. The mechanism is the same in both cases, and it is simpler than most students think. The browser is constantly talking to the website, and when you leave the exam tab, the browser fires a blur event. When you return, it fires a focus event. The LMS logs the exact timestamp of every blur and focus event, creating a timeline of how often you navigated away and for how long.

A plain LMS quiz logs the events and does little else. Canvas and Moodle automatically record focus events even without a webcam, which means a student who switches tabs during a quiz has already created a record, whether or not anyone looks at it. A single accidental tab switch is not an automatic fail, but frequent exits from the exam window are the primary red flag for academic integrity investigations, and the log is the evidence the investigation starts from.

What proctoring adds, and what it does not

Lockdown browsers change the question entirely. Respondus LockDown Browser locks the computer into a single-application mode: no new tabs, no copy and paste, no printing, no screenshots, no second monitor, no background software. It disables Alt-Tab, Ctrl-C and Ctrl-V, and Print Screen, blocks virtual machines and remote-desktop software, and forces kiosk mode so the exam is the only thing the machine can do until submission. Under the hood it is a fence around the screen, and a fence has no idea what is happening on the other side of the desk.

AI proctoring adds the monitoring layer the lockdown browser lacks. Proctorio, Honorlock, and Respondus Monitor record the session and run the footage through algorithms that track face, eye movement, head position, lighting changes, and background noise. They can detect multiple screens, screen-sharing software like Discord or Zoom, and background apps. But the systems flag, they do not find. An AI flag is a timestamp in a recording that a human reviews afterward. It is explicitly not a determination of academic dishonesty, and the review is where the context lives: a phone call, a notification, a child walking into the room.

What students and institutions should know

For students, the practical rule is to treat every focus event as a record, because it is one. Do not switch tabs during an exam, even to check the time, because the log does not know why you left. If something forces you away, a notification, a call, a crash, say so in the exam's comment field or to the instructor immediately, while the timestamp is fresh. The flag and the explanation travel together, and the explanation is what the reviewer reads.

For institutions, the lesson is that a flag without a review process is worse than no flag at all. A system that auto-fails students on tab switches will fail the student whose laptop lost focus to a system notification, and a system that logs everything and reviews nothing has built a haystack with no needle. The honest design is the one the better platforms already use: record the events, show the student a warning the moment they switch, and let a human decide. Nothing is auto-failed, and the record is there when it is needed. For the tools that try to watch the screen itself, the AI overlay blind spot shows what lockdown browsers cannot see, and answer pattern analysis covers what the answer sheet reveals when the screen does not.