How much time did you spend on your phone this week?

Take a moment to estimate and check to see if you were right. On an iPhone, you can check by going to Settings, scrolling down to “Screen Time,” and then selecting “See All App & Website Activity.” On an Android phone, go to Settings, scroll to “Digital Wellbeing and parental controls” and select the icon of a bar graph in the top right.

How far off were you?

A group of scientists explored this idea with a simple test. They installed an app on 137 adolescents’ phones which would track how many times they checked their phone and the total time their phone screens were being used. Every evening for two weeks, participants reported how much time they thought they had spent on their phones. At the end of the experiment, they were also asked how much time they felt they had spent on their phones on an average day.

The researchers found that, on average, adolescents underestimated their time spent on devices by more than half an hour each day. When asked about a typical day, they underestimated by nearly an hour. Some participants underestimated by more than two hours.

Similar studies have found that accuracy depends on how people use their phones (e.g. Instagram vs. Snapchat) and that adults tended to overestimate their time on devices.

Some evidence suggests that people are even worse at estimating time spent gaming than using screens in general. Studies have demonstrated that people tend to underestimate time spent playing video games more compared to reading and using the internet.

This is a major problem for researchers. Many pivotal studies rely on participants’ ability to accurately report how much time they spend on screens. The most common way to research the effects of screen time is to ask people about their use of screens and then look for a correlation with another variable.

For example, participants in a study might fill out a questionnaire asking how much time they spend playing video games in an average week and then answer questions related to symptoms of depression. If people who estimate that they spend more time gaming also report more symptoms of depression, researchers could conclude that there is a correlation between video games and depression. But, based on the information found in this new study, there would be significant reason to doubt that the data researchers use are accurate.

For example, some studies which rely on accurate self-reporting have found that:

• Gamers are better at focusing attention.

• Time spent gaming does not correlate with attention problems or GPA.

• Surgeons who play more video games perform laparoscopic surgeries faster and with fewer mistakes.

• Autistic children* spend more time playing video games than their allistic (non-autistic) peers.

• People who play more video games have worse overall health and academic performance.

• Teens who play more video games have higher rates of obesity, diabetes, high blood pressure, and high cholesterol.

These studies might be accurate, but they also might actually be uncovering something more nuanced. Perhaps boys are more or less accurate than girls at estimating their time on screens. Maybe lonely kids overestimate their time on screens. Children whose parents impose strict limits might pay more attention to how many minutes they’ve expended each day. Maybe teens who have recently talked to a doctor are more likely to include time spent listening to music on a phone in their estimate.

Researchers need to take these new data seriously to ensure that their findings are accurate. I believe that several variations of the current study need to be conducted to continue challenging our assumptions.

• Repeat the study with tablets, computers, and gaming consoles.

• Repeat the study with a large, diverse group of people and see how different variables affect participants’ accuracy.

• Repeat the study, asking about screen time in different ways. “How much time do you spend on screens in an average week?” might be answered less accurately than “How much time do you spend on screens in an average week, including time spent listening to music, watching videos, scrolling through social media, watching television, and playing games?”

Hopefully, we can use this information to re-examine previous studies and find out what we know for sure.

* Although many refer to autistic people as “people with autism” or “people with autism spectrum disorders,” almost 90% of autistic adults prefer “autistic person.” This language is used here to respect that preference.