You’re in the middle of a texting conversation when the other person suddenly stops for a long pause before responding.

What does it mean? Maybe they got a call, got distracted by something else, or their thumbs needed a break. But if your last missive was about something heady, it’s also possible they’re taking the time to cook up a lie.

Researchers from Brigham Young University (BYU) asked more than 100 students to respond to 30 questions each that were generated by a computer and texted to the participants. In half of their responses, the students were asked to lie. They found that when the students were lying, it took them 10% longer to send the text message, and they made more edits than usual.

According to the researchers, people are only able to detect when someone is lying to them in person about 54% of the time — there are behaviors that some people swear are giveaways of mendacity, like not being able to look people in the eye, or fidgeting. So it’s incredibly difficult to identify when someone is doing so online or over the phone. During texting conversations, you can’t hear the other person’s voice or see their facial expressions — which makes it easier to get away with an untruthful response. (A 2011 study found that people are more likely to lie while texting than in face-to-face conversations or video conferences for just this reason, and Manti Te’o’s catfish experience highlighted how easily it’s done on social media.)

But while digital conversations are easy to manipulate, the results suggest that some patterns, such as the delay in texting fibs, could become cues for reading into the hidden meanings of voice or text communications. Granted, it was a small study, and not a real-life setting, but the researchers say their findings raise interesting questions about how the validity of other online transactions, such as communications on social media feeds and chat rooms, might be interpreted when it comes to issues of security and personal safety. “We are starting to identify signs given off by individuals that aren’t easily tracked by humans. The potential is that chat-based systems could be created to track deception in real-time,” said study author Tom Meservy, a BYU professor of information systems in a statement about the study.

If you can’t look them in the eye, maybe it’s worth looking at a stopwatch.