The billion-person experiment

We are test subjects in the digital age and are still learning the consequences

Photo by Cottonbro studio, pexels.com

Twenty years ago, almost no one imagined we would spend 11 hours a day with digital media, interrupt conversations to glance at our phones, or struggle to stay focused long enough to finish a book.

Digital media has brought innumerable benefits, and we can’t go back to a world without it. We can connect to parents and children and coworkers at a distance. We can organize work more efficiently, conduct everyday tasks online, get the latest news.

And yet there’s another side. A person who I once interviewed, a writer, told me that she can’t read more than a few pages in a book. She also said that she couldn’t control checking her phone or scrolling through social media. She wasn’t unusual. Over two decades of studying digital behavior, I’ve heard versions of her story again and again.

These weren’t design goals. They were unintended consequences that emerged years after these technologies were deployed. We have become both the beneficiaries and unintended victims of our digital revolution.

If no one intended these outcomes, why did they happen?

One reason is that technology is rarely evaluated for how it might reshape people’s attention, cognition, emotional well-being, or relationships over months and years. User testing typically asks whether people can use an interface, not how a technology changes them after living with it over time. Once a product is released, companies rarely look back.

Many effects of technology use occur weeks, months or years after a technology is introduced.

Take the case of Facebook. It began as a campus social network. Soon advertising became central, and then algorithmic ranking transformed what people saw. Misinformation, polarization, and other societal harms emerged gradually, not immediately. In 2003, few people could have anticipated how profoundly Facebook would reshape individual and public life.

Now consider a design process that takes a long-term view. A company developing a new social media platform involves not only adult users, but also teenagers, parents, teachers, psychologists, pediatricians, sociologists, ethicists, and futurists. Rather than relying solely on short laboratory studies, the technology would be evaluated in real-world settings over months or even years to understand how it affects people’s thinking, well-being, learning, and relationships.

In other words, imagine a design process that asks not only whether a technology works, but also what risks it may create for individuals and society over time.

Just how pharmaceutical trials take time to understand the efficacy and dangers of new drugs, the same is true of new technological systems. Software differs from pharmaceuticals in one critical aspect: it continually changes through updates, new algorithms, and new features. Harms may not be evident right away but may slowly emerge as people’s practices change as the technology evolves. There are so many factors at play that can affect users: individual differences, changes in society, malicious actors, and new design decisions, like the choice of Facebook to introduce algorithms in 2009. That makes ongoing monitoring even more important.

This lesson suggests a different approach for technology design. I propose four basic principles.

Ask what technology does to people, not just what people do with it. User testing should go beyond identifying problems with hardware, software, or interface design. Both laboratory and field studies should examine how technologies affect cognition, emotional well-being, and relationships. Because people respond differently, studies should include diverse populations to understand who benefits, who is harmed, and under what conditions.

Study technology over time. While initial user studies can uncover immediate problems, like cognitive overload, it’s important to examine consequences over a longer term. Will people adapt or will harms persist? Even after deployment, technology companies should be responsible for continuously monitoring the technology use for long-term effects and modifying them whenever significant risks emerge.

Put human experts on design teams. Technologies that shape people’s lives should be designed not only with users but also with psychologists, sociologists, health experts, ethicists, and policy specialists. For products aimed at children and adolescents, parents, educators, and developmental psychologists should play a central role throughout the design process.

Create independent oversight. Existing institutions evaluate issues such as privacy, competition, and technical safety, but few systematically assess the long-term effects of technology on cognition, mental health, or relationships. Independent oversight would almost certainly slow some innovation, but we already accept similar tradeoffs when products can affect public health, like with FDA review of pharmaceuticals and medical devices. An oversight body could determine when long-term evaluation is required and require changes if significant harms emerge, even years after deployment.

We already ask whether new technologies work and are usable. The more important question is whether, over time, they can help the people who use them flourish.

Users should not have to become unwitting participants in a decades-long experiment whose consequences are only understood after billions of people have adopted the technology. We require evidence that medicines or medical devices are safe before prescribing them widely. As digital technologies increasingly shape our attention, relationships, and health, we should expect the same standard of care for the systems that increasingly shape our lives.

************************************************************

You can read more about how digital media reshapes our attention in my book Attention Span.