Americans aren’t moving enough. Despite years of public health efforts, 80% of adults fall short of physical activity guidelines and over 70% are reported to be overweight.¹ Increasing physical activity is clearly a national health priority. At the same time, climate change has intensified public awareness of transportation’s environmental impact— the largest source of greenhouse gas emissions in the United States in 2022.² As cities grow and invest in more walkable infrastructure, eco-friendly transportation is becoming both more feasible and more socially valued. Encouraging walking—a practical, low-cost activity—offers a rare win-win: reducing climate impacts while improving physical health. 

Frerich focuses on digital nudges because online platforms quietly structure so much of modern life and decision-making. Americans spend a third of their waking hours on apps, and with a multi-billion dollar app market, there is a strong incentive amongst tech giants to prioritize speed and convenience as opposed to long-term user wellbeing.^{3, 4} When Google Maps defaults to driving for directions, it unintentionally reinforces car dependence, even when a route is easily walkable.

Traditional interventions to promote walking are expensive, slow, and often tested on subjects already interested in increasing their physical activity. Frerich argues that implementing a subtle digital nudge on a map platform can expand the reach of pro-health interventions. With over 100 million annual Google Maps users, tiny tweaks to choice architecture could influence transportation behavior at scale. 

Frerich conducted a randomized online experiment with 414 U.S.-based adults recruited through Amazon’s Mechanical Turk (MTurk) platform. Participants were shown one of four Google Maps–style screenshots depicting a walkable 0.8-mile route (14-minute walk, 6-minute drive): a walk default, a drive default, a walk default coupled with a social norm graphic stating that “Most choose to walk”, and a drive default with the same social norm message. The social norm graphic also displayed the estimated amount of calories burned from the walk. 

Participants were then asked to assume that all modes of transportation offered on the screenshot are available and were instructed to select their preferred choice from walking, driving (or ridesharing), cycling, or taking public transport. 

So, did defaulting to walking directions increase the preference for it as a mode of transportation? The results establish that defaults make a difference–especially when combined with social norms. Participants assigned to the walk default and social norm message condition were 1.27 times more likely to choose walking over driving compared to those given the drive default or drive default + social norm message condition. The combination of defaults and messaging was powerful with 71.2% of participants in the walk default + social norm condition reported a preference to walk, as opposed to only 55.3% in the drive default group. 

The experiment also found that people who reported walking less than or equal to 2 times per week were significantly more influenced. For them, exposure to the walk default + social norm condition led to a 17.8% increase in choosing to walk. Conversely, participants who already walk frequently remained unaffected by the defaults, likely because they consistently chose walking regardless of the map design. 

These findings spotlight a tool in public health and even, pro-environment policies: digital choice architecture. When billions of micro-decisions, like how to travel half a mile, are influenced by app defaults, those small choices accumulate into massive collective outcomes. Even modest behavior shifts, scaled across millions of users, could lead to benefits such as increasing national physical activity levels, reducing short-trip emissions, and lowering traffic and noise pollution. 

These promising findings aren’t without their criticisms. Specifically, the study measured hypothetical choices without any real-world tracking. We often make decisions without acting upon it in the end, so whether these results translate into real-world findings opens an avenue for further research. On top of that, the routes used in the Google Maps screenshots during the experiment may be unfamiliar to participants. This unfamiliarity may inflate default effects which again emphasizes the need for deeper study rooted in reality. Despite these points, the broader implication is clear: if digital platforms have the power to shape our behavior in any way, we should be intentional about how–and toward what ends. 

We think of walking as a personal choice, but this study reminds us that it’s also a product of design. Digital defaults steer us in subtle, persistent ways—toward the car, the elevator, the fastest route. But change the default, add a cue, and suddenly more people take the physically active route. Digital defaults aren’t neutral, and when we design them with care, they can gently guide us toward choices that support both our health and our future.

This article summarizes:

Frerich, J. (2025). Walk this way: harnessing digital nudges to promote walking for transportation. Behavioural Public Policy, 1–15. doi:10.1017/bpp.2025.10

1. Fryar, C. D., Carroll, M. D., and Afful, J. (2020) ‘Prevalence of Overweight, Obesity, and Severe Obesity among Adults Aged 20 and Over: United States, 1960–1962 through 2017–2018’, National Center for Health Statistics, Division of Health and Nutrition Exam Surveys. https://www.cdc.gov/nchs/data/hestat/obesity-adult-17-18/overweight-obesity-adults-H.pdf
2. U.S. Environmental Protection Agency. (n.d.). Sources of greenhouse gas emissions. https://www.epa.gov/ghgemissions/sources-greenhouse-gas-emissions 
3. Wakefield, J. (2022), People devote third of waking time to mobile apps. BBC. Available at: https://www.bbc.com/news/technology-59952557 
4. Harris, R. (2023), The mobile app market is still growing to hit 571 billion by 2030. App Developer Magazine. https://appdevelopermagazine.com/The-mobile-app-market-is-still-growing-to-hit-571-billion-by-2030/