We Think We’re Better Drivers Than Autonomous Vehicles

Think you’re a better driver than most? You’re not alone, and this belief may be one reason self-driving cars haven’t gained widespread acceptance.

A new Harvard study found that about 77 percent of participants rated themselves as superior to automated vehicles, while 60 percent thought other people were worse drivers. Drivers’ egos may be hindering the widespread acceptance of automation.

“We find people are OK with some level of automation but prefer higher levels of automation for others than themselves,” the researchers explain. “This is because they think that they are better at driving than increasingly automated systems. We believe this creates a barrier to adoption.”

Safer driving

The drive towards vehicle automation isn’t solely a technological or financial endeavor for automakers; it’s a pressing public safety concern. Globally, traffic accidents are the primary cause of death for individuals aged 2 through 59, with approximately 98 percent of accidents attributed to human error.

Moreover, accidents incurred a cost of $340 billion, equivalent to 1.6 percent of the US Gross Domestic Product in 2023, as reported by the US National Highway Traffic Safety Administration. However, the adoption of automated cars faces significant psychological resistance from drivers.

“Whether people utilize automated vehicles is not just a matter of how they view the technology in a vacuum, but it’s how they view it in relation to themselves,” the researchers say.

Perceived abilities

To understand how US drivers perceive their own driving abilities compared to automated vehicles, researchers conducted two sets of experiments and surveys.

In one experiment involving 579 participants representative of national demographics, half of the group rated, on a scale from zero to five, the level of driving automation they preferred for themselves, while the other half shared the level they preferred for others.

Participants favored level two for themselves on average. Level two vehicles offer multiple automated features like adaptive cruise control and lane centering, allowing human intervention at any point. For others, participants preferred level three, where the car also monitors and reacts to the environment, with human intervention still possible.

In the second experiment with 803 participants, preferences for riding in fully automated vehicles (level 5) were assessed on a scale from 0 to 100. Once again, drivers preferred such vehicles for others (52 out of 100) more than for themselves (40 out of 100).

“In both cases, people wanted less automation for themselves than they wanted for others,” the researchers explain. “But if everyone acts on this preference, then this will prolong the adoption of automated vehicles.”

Moving forward

To overcome this psychological barrier, automakers and policymakers can take several approaches:

  1. Highlight benefits beyond safety. While safety is a significant advantage of automated vehicles, emphasizing this aspect may not sway drivers who already believe they are safer drivers. Exploring other benefits, such as convenience or productivity gains, could be more persuasive.
  2. Offer incentives, such as insurance discounts or tax breaks. Financial incentives may encourage drivers to reconsider automation features and overcome their discomfort with driving what they perceive as an “inferior” vehicle.
  3. Educate consumers effectively. Research has shown that educational interventions, such as videos addressing biases, can reduce biases. Marketers should strive to create short, memorable, and motivational educational materials. For example, Metro Trains in Melbourne, Victoria, successfully launched a viral ad campaign called “Dumb ways to die,” focusing on railway safety.
  4. Be specific. People tend to overestimate their driving abilities when thinking in abstract terms, like “driving well.” Providing concrete examples or experiences can help individuals better understand the potential benefits of automated vehicles and reconsider their perceptions.
Facebooktwitterredditpinterestlinkedinmail