I read this review and realized it’s not just most UX designers who think touch screens in cars are a bad idea (they are skewered pretty bad here): https://arstechnica.com/cars/2017/09/range-rover-velar-review

My question is, has anyone from a car company publicly discussed the usability and safety aspects of touch screens in cars? Maybe in a press release, interview, blog post, anything?

Since so many designers and lay people have spoken out against this, maybe I’ve just missed the part where car makers are making their case? Even if I would disagree, maybe there is another side to the story?

I’m speculating it wouldn’t be hard to do A/B testing and prove touch only controls used while driving are less safe. Therefore, it seems like this would have compelled a response at least on why it’s being done so much.

  • 5
    I've got no proof - but I bet it's incredibly cheaper to produce touch screen versus installing actual knobs and physical buttons. In my experience (and consider this a User Testing of 1 person), it always takes more time to find and use touch screen buttons while driving. As a driver, it is easier to feel around for unique button shapes while keeping BOTH eyes on the road. – jhurley Sep 19 '17 at 19:18
  • 3
    Ugh, I hate the touchscreen in my car! I just want a wheel, a stick shift, and missile launchers... – Austin French Sep 19 '17 at 22:18
  • I want to be in the control group for A/B tests of any new ideas in car UX, please! 8 decades of experience is good enough for me. – user67695 Sep 19 '17 at 23:18
  • 1
    Car industry really runs some of the most extensive tests in any industry, so I have to guess tests have shown this is a good approach. This being said, I hate my car's touch screen controls. Not because they're bad, but I think they require too much attention on my side (compared to physical knobs and controls) – Devin Sep 20 '17 at 5:23
  • 3
    Touchscreens are used for cost and for better sales. Despite being the inferior solution, they are perceived as being more "modern" by customers. – Peter Sep 20 '17 at 9:04


A gigantic and potentially distracting touchscreen as the primary interface doesn't seem like it belongs in a car with the primary selling point of protecting the people inside it. Yet, after just a couple hours of familiarizing myself with Volvo's new and much-improved interface for its "Sensus" system, I felt totally safe using it while cruising down the desert freeway at 75 mph.

Implementing capacitive touchscreens in the car:

So a new generation of larger capacitive touchscreens will make cars both safer and more desirable. The new, larger form factors will also, however, give automotive manufacturers a new set of challenges in ensuring that their capacitive touchscreen implementations are both reliable and robust. Synaptics’ experience in helping customers implement capacitive touchscreens in tablet-size form factors suggests that there will be broadly four issues that manufacturers will need to take into account.

Do touchscreens make driving more dangerous?

"Generally speaking the electronic elements in a car cockpit enhance driver comfort and safety," says Alexander Klotz of German automotive component and systems maker Continental.

Electronic aids and the possibilities they offer help to create a logical structure for the driver who can often let apps handle all the entertaining, navigation and climate control duties.

"If you wanted switches and knobs for all these functions you would need hundreds of them," says the expert. That is why modern cars often have a central touchscreen which combines and carries out all the necessary functions.

Car makers insist that the technology is safe, saying that the screens and apps are designed to minimise the time needed to operate them. They argue that much of the functionality can be operated by voice or by controls on the steering wheel so drivers can use the technology without taking their eyes off the road.

  • 1
    I bought a cheap plastic mp3 player for my car because it has something tat my phone doesn't: buttons that I can find without looking. When touchscreens really are touch screens - that the buttons can be felt - there will be something to discuss. Why isn't someone working on that? – user67695 Sep 20 '17 at 11:37
  • 2
    The OP was asking for some references... I think, these links make sense to his question. – DPS Sep 20 '17 at 15:40
  • 2
    DPS's answer is perfectly on point. Unfortunately the the responses don't seem that convincing. For example, in many cars voice control quality or design make it go unused, and some vehicles do not offer alternatives for important functions that are typical adjusted while driving. I guess my takeaway from their answers is that they believe with a great implementation, complementing the system as a whole, touchscreens are good and safe. I don't rule out that's possible, I just don't see it much. – whitneyland Sep 24 '17 at 19:10

You might enjoy this article from the New York Auto Show, showing all of the touch screen consoles on display that week.

What's clear from the variety of approaches to touch screen consoles is that there is no coherent, consistent approach. There are no established industry-wide UX best-practices. This leads me to think only very limited UX testing is being done, and likely only in terms of design validation rather than early-process feedback or focus groups.

The driving force behind the lack of established best-practices is twofold. Primarily, the auto industry is about innovation and forward-thinking. Looks often trump performance (ask anyone who's ever owned a DeLorean if you don't believe me), and usability is sacrificed for style.

Secondly, car manufacturers are looking more to the future than ever before, and beginning to base their design principles around the future of autonomous, electric vehicles. There are quite a few UX problems that need to be worked out before these vehicles become a reality, and while current sales are important, automotive interface designers are working on the problems of the future more than the problems of the present.

  • Autonomous vehicles will eliminate all of the dangers of touchscreens, so that is pretty much a null hypothesis. Also, people will probably not buy self-driving cars, because they won't have to, so I hope the car companies have thought of a future without customers. – user67695 Sep 19 '17 at 23:21
  • 1
    Autonomous vehicles will come with their own set of dangers, though they will undoubtedly be fewer than before. To think that any system will ever completely eliminate risk is folly. There's an argument to be made that the everyday consumer won't buy self-driving cars, but rather companies like Uber will be the primary customers. That said, the same could have been argued about non-autonomous vehicles, yet the world isn't dominated by taxis. – denveruxer Sep 20 '17 at 5:25
  • 1
    If autonomous cars do not reduce danger, then they are not an improvement, therefore they should not be called 'technology'. People won't use them unless they are much safer. Taxis have one big part that costs a lot of money: the driver. Eliminate that, and the cost drops way down. That is why we developed machines, computers, robots... To eliminate cost and make things possible and affordable that were not before. I can't imagine why someone would buy an autonomous car, only rich people and poor people, and enthusiasts, will still drive, just like with horses and sailboats. – user67695 Sep 20 '17 at 11:25
  • I never said that autonomous cars wouldn't reduce danger, just that they have their own sets of risks. Any form of transportation will always have some element of danger. People will buy AVs for the same reasons that they buy cars today-- they want a private, personalized experience that is as much a part of their personality as the clothes they wear. That aspect of the automotive experience has not changed, and is unlikely to do so in the future. – denveruxer Sep 25 '17 at 16:42
  • I guess I own my ancient car rather than take a taxi everywhere because it is cheaper and I don't have to request or wait. If there was no wait, and it was cheaper, the car would go to the crusher, my car is not a pet or hobby, or lifestyle statement. Most people will probably make the same decision. As for danger, well, we will just have to agree to disagree that people will expect newer technology to be safer. It has always been true so far. – user67695 Sep 25 '17 at 16:48

In a review of the safety of in-car displays it is concluded:

The empirical results illustrate that the visual design of in-car displays can have a significant impact on the potential for visual distraction. Task length and the spatial separation between interaction elements, especially those encoded sequentially, arise as two of the critical factors for the probability of in-car glances to exceed the safe glance limits in this context. The findings suggest that visual designers should try to minimize task duration as well as the duration of all visual encoding steps required for the in-car task. This means, for example, that the number of available menu items should be limited and that the distance between interaction elements encoded one after another in a task sequence should be minimized to a level where clutter is still avoided. Given a prolonged search task, an extended estimate of a safe in-car glance duration, inaccuracy in driver’s time perception ability, and a longer individual encoding step near the end of a glance, milliseconds can truly matter in this context (Gray and Boehm-Davis, 2000). The idea of minimizing visual encoding steps relates to the idea of Janssen et al. (2012) of providing shortest possible “action sets” and thus, natural breakpoints to encourage task switching and reduce distraction by secondary tasks. In short, shorter visual encoding steps should give more room for breakpoints.

Reference from - Kujala, T., & Salvucci, D. (2015). Modeling visual sampling on in-car displays: The challenge of predicting safety-critical lapses of control. International Journal of Human-Computer Studies, 79, 66-78. doi:10.1016/j.ijhcs.2015.02.009

  • In programming, we have an improvement called Object Oriented programming - making pieces of code behave in a familiar way, as an analogy to 'things'. So, how about making the controls of the dashboard behave in an Object Oriented way: this knob does this, that lever does that. No conflict, no encoding, no changes. Oops! We had that before I was born. Darn. – user67695 Sep 21 '17 at 13:30

I think we need to consider a few things when it comes to interface designing in vehicles.

  • The absolute minimum necessities - Anyone looking to buy a car today expects certain things to be available as a part of their dashboard. Navigation, climate control and multimedia other than the bare minimum. Some of these are very difficult to manufacture and use in the mechanical knobs & buttons approach (Imagine having to fit and use a whole keyboard or worse; using a dial like in the BMW SUV 5 series!)

  • Not all of these need to be used while driving - I understand that being in a car, one would expect to be driving while using these but not all of them need constant fiddling. Navigation can be set before starting the journey, climate controls are fairly simple to use and multimedia has other quick-keys (coming to that in the next point)

  • On-wheel controls - Nowadays, most vehicles have physical quick-access controls on the steering wheel for music, driving modes, climate control and even some complicated actions.

I am not saying that it's safe or that it couldn't be improved but with changing requirements and addition of features, compromises need to be made. Some manufacturers put more effort than the others but in the end, we have to accept the fact that physical controls would slowly fade away. And with each iteration of development, the problems would be fixed

  • I think we have to realize that a car is for travelling, it is not a living room or office. The driver has a job to do, he is not being entertained. When we realize that, then the appropriate things will be in the car for comfort and safety, and will be laid out conveniently. Oh! Already done! "Nothing new to see here, move along." – user67695 Sep 20 '17 at 11:33
  • @nocomprende should you be required to pull over and stop the car to adjust the climate control to be cooler or warmer? Some cars require a touch screen to be used to adjust climate controls. I've had the same car for 5 years and can't operate the touch screen without looking at it. Even if it's to a small degree, you don't think these kinds of things are a safety issue? – whitneyland Sep 20 '17 at 16:08
  • @Lee as I said, we have had knobs and levers for 100 years of cars. They work. No new designing needed. All the knobs and levers that were there when I was a child are still there, doing the same things in the same way for the same reasons. Computers didn't change what was needed by human bodies for comfort and safety, they only turned our brains to mush, apparently. – user67695 Sep 21 '17 at 13:26
  • @nocomprende you're right no new designing is needed, that's the whole point here! That these touch screens are being used when they appear they may be unsafe, and better solutions exist. I was hoping to find the reasoning behind this unstoppable trend. – whitneyland Sep 24 '17 at 19:04
  • @Lee Greed. – user67695 Sep 24 '17 at 19:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.