Subscribe to the OSS Weekly Newsletter!

Register for the OSS 25th Anniversary Event

Digital Amnesia Has Been Exaggerated

The idea that we forget so many things because we rely on our computers to remember things for us has a few nagging problems

Do you know your best friend’s phone number? In the 1990s, I had my best friend’s phone number memorized, and I’m pretty sure I remembered it just now after thinking about it for a minute. But the phone numbers I use today? They don’t live in my head. They live on my phone.

The press loves to serve us alarming headlines about so-called digital amnesia and what our overreliance on computers and the Internet might be doing to our brain. A particularly egregious headline reads, “Remember how it felt to remember things?”, as if having a smartphone transformed all of us into Guy Pearce in the movie Memento, unable to form new memories.

When we cast aside the doom-mongering that warns of our devices making us dumber—or even worse, of technology giving us “digital dementia,” comparing Internet use with a head injury or the kind of cognitive decline seen in Alzheimer’s disease—we stumble upon two problems: a foundational study that doesn’t replicate very well and a scientific-sounding term coined by people who want to sell you something.

It's a reminder that, on important matters, it pays to read the primary literature.

Lest we forget where “digital amnesia” came from

The phrase “digital amnesia” means the experience of forgetting information that you believe a digital device, like a computer or smartphone, will store and remember for you. Why commit someone’s birthday to memory when your phone can send you a notification the day of?

It may sound like a scientific concept, but actually, “digital amnesia” is an example of bad PR. Imagine being the head of a vitamin supplement manufacturer who wants to increase revenue. You commission a survey, asking people if they take a multivitamin each day. The results? 70% say they do not, and of those who do, over half admit they sometimes forget to take it. You issue a press release with these findings, journalists jump all over it, and pretty soon, headlines are proclaiming that “Most Canadians do not take their vitamin pill” and “Even those who take vitamins often forget to do so. Are you one of them?” Readers become alarmed and head straight to their local drugstore. Sales of your vitamin supplements go up.

“Digital amnesia” was a term devised by the Kaspersky Lab, a cybersecurity firm, following a series of surveys they commissioned. In them, they asked people if they used their smartphone as a form of memory, if they thought they were more reliant on it than they were in the past, and, importantly, if they used antivirus software or if they created backups of their digital devices. It turns out that many people do rely on their devices to remember things for them, but would you believe that “58% use no antivirus software and only 29% back up precious information stored on their devices,” the Kaspersky Lab report informs us, “putting the majority of their memories in jeopardy should they suddenly become inaccessible due to loss, theft or cyberthreat?” Guess who sells cybersecurity solutions for home and business?

The story of “digital amnesia” is one of creating a problem by asking the right questions and positioning yourself as the solution.

That is not to say that there is no truth to the worry that we are offloading important information onto our thinking machines. In fact, before “digital amnesia” had even been dreamt up, a seminal paper had been published on what would later be christened “the Google effect,” a related phenomenon by which we forget something because we know we can Google it later.

That study, which made waves when it was published in 2011, has become a lesson in how difficult it is to do one of the most important aspects of research: trying to replicate results.

The replication devil is in the details

According to a science journalist who wrote about this paper when it was published, Betsy Sparrow had experienced something all-too familiar which prompted her research into what would become the Google effect. She was watching the 1944 movie Gaslight, which has now inspired the word “gaslighting,” which means to psychologically manipulate someone into questioning their own sanity. When Sparrow saw the actress playing the maid, she had one of those “oh, it’s that woman from… oh, what’s her name?” moments. She looked it up on her phone. It was Angela Lansbury, then 18 years old.

She went on to conduct four experiments into this phenomenon, and the results were published in this now-influential paper. The ways in which she and her team went about testing whether or not people turn to the Internet when asked to remember something may sound a bit weird when explained. Psychology studies can appear quite removed from reality, as researchers attempt to cast out all sorts of things that could influence the results in order to test a specific relationship between two variables in a laboratory setting. Sparrow and her colleagues didn’t have their participants watch Gaslight and attempt to name its actors without checking their phones. Rather, one experiment involved what is known as a Stroop task. Put simply, if I write down the name of a colour on a piece of paper, show it to you, and ask you to name it, you’ll be able to do it quickly (“Red!”). But if the name is printed in a different colour—say the name “Red” is printed in blue—you will experience a slight delay because your brain is drawn to the colour of the ink and it struggles a little bit with the mismatch. That’s a Stroop task.

The undergraduate students who participated in the first of Sparrow’s experiments went through a similar test. They were asked trivia questions of the sort that might make you wish you could Google the answers. Then, they were shown words printed in either blue or red and had to quickly press the key corresponding to that colour. When the word was related to computers (like “screen,” “Google,” or “browser”), they pressed the key a little bit slower than when the word had nothing to do with computers (like “Coca Cola,” “book,” or “television”). This made the scientists think that those trivia questions had made people long for access to a computer, and those computer-related words had captured their attention and caused a delay in the task of choosing the right colour. There were other experiments, to which I will return, that taken together led the researchers to conclude, based on this preliminary evidence, that the Internet had become a form of transactive memory. Like a husband turning to his wife when asked to remember a friend’s birthday, we turn to the Internet, Sparrow and colleagues argued, when we need to tap into our memory. We don’t remember the answer, but we remember where to find the answer.

Scientific findings are more reliable when they can be reproduced by an independent team, however, and this was just one study. Cue an international team of scientists who, in 2018, published the results of a massive endeavour: they had tried to replicate 21 experimental studies in the social sciences whose results had been published in the preeminent journals Science and Nature. They reached out to the researchers behind these studies to make sure they would follow their protocol to the letter. The team found a similar result for only 62% of the studies they tried to replicate, though the magnitude of the effect was on average only half of what had been originally reported. One of the studies that did not replicate was Sparrow’s first experiment.

You may think that the case is closed. Sparrow’s experiment could not be replicated, ergo her results were a fluke. But it unfortunately gets more complicated.

The team in charge of the replication was unable to reach Sparrow and her team to ensure they had every detail of the protocol right. Researchers have to describe their protocol in their papers, but often details are missing or the writing leads to ambiguity. When the replication team’s results were published, Sparrow pointed out that she had conducted her Stroop test slightly differently. Also, while her study was published in 2011, the tests were done in 2006. Some of the computer-related words she had used were “Altavista” and “Lycos,” search engines that early adopters of the Internet will faintly recognize. In 2018, the replication team should have used more relevant words. Two years later, a German team took Sparrow’s comments into account and also attempted to replicate her first experiment, and their results failed to show the famed Google effect.

But the Sparrow paper had also reported on three other experiments that had nothing to do with the Stroop test. Instead, participants had had to type trivia statements into a computer and had been asked to remember them. The researchers had toyed with whether the typed text was saved or erased by the computer, and whether participants knew in advance what would happen. The conclusions from these tests were that knowing that the computer will not save this information enhances your memory of it, and that when the information is saved, people may not remember the information word for word but they’ll remember in which folder it was saved. In 2021, researchers at the University of California, Santa Cruz tried to replicate this part of Sparrow’s experiment, but they only found the same result when participants were first put through a practice run where they saw that the computer’s saving function was indeed reliable. The fly in the ointment, according to these researchers, is that there was no practice run in Sparrow’s study, so why was it necessary here to find the same result?

If you are sighing in despair, you understand the problem. Scientific research is messy and reliable answers don’t emerge from preliminary data. The study of exactly how our use of modern technology is influencing our memory and our thinking is still very much in its infancy, and it is hard for scientists to keep up with the pace at which our use of the Internet evolves. These laboratory experiments are far removed from our reality. We remember things we find interesting. A trivia statement like “the international telephone dialing code for Antarctica is 672,” which was used in Sparrow’s study, may not be seen as engrossing or useful enough to justify a slot in our brain’s memory bank in the first place. (A college student’s thesis did test for this in trying to replicate Sparrow’s findings and did not find that interest in a statement was important in remembering it, but the test was only done in 20 people.)

The use of our phone’s camera to capture an event we attend has also been the subject of scientific studies, and there is evidence that it may distract us from the smells and sounds and the overall experience. Then again, our own memories are not as reliable as we’d like to believe, and our ability to remember often degrades with age, so the ability to take photos and videos using a smartphone can be a boon.

Concerns over how we use modern technology are justified, but we shouldn’t veer into a moral panic. Some of our ancestors famously fretted over wax tablets and printing presses, and how these new-fangled technologies might ravage our memory. Even if the Google effect is common, it may be more of an adaptation than a devolution. Do we need to memorize telephone numbers? Probably not. Offloading our memory of trivialities onto computers can allow us to focus on things that matter.

When I watch a movie or television show, I often experience the “it’s that guy who was in that thing” effect. Instead of memorizing the name of every actor, I simply turn to the Internet Movie Database and within seconds I am reminded of who this actor is and where I saw him before. I don’t think that’s such a bad thing.

There may be genuine downsides to relying on technology to remember things for us, and more (and better) studies, as always, are welcomed, but the evidence simply isn’t in to justify media headlines that boldly assert that our smartphones are wreaking havoc on our memory.

Take-home message:
- The term “digital amnesia” was coined not by scientists but by a cybersecurity firm that sells solutions to help protect the information we store digitally
- Experiments described in a seminal paper on the way in which we turn to technology to remember things for us have failed to show the same results in the hands of other scientists, although the experiments were not identical
- The effect that our reliance on technology has on our memory is still not clear, as the scientific research on this question is still in its infancy


@CrackedScience

Back to top