Thinking / 02 Interaction × Behaviour
Physical buttons
feel easier.
But why?
Is the hardware really the reason? A reflection on digital controls, muscle memory, and the quiet value of knowing what happens next.
01 / The familiar argument
There is a familiar argument around modern car interiors: physical controls are better, touchscreens are worse.
I understand why that argument is attractive. Driving is different from using a phone or laptop. Attention matters, and anything that makes a driver look away from the road deserves scrutiny.
But I keep wondering whether we are sometimes blaming the wrong thing.
Maybe physical controls do not feel better simply because they are physical. Maybe they feel better because they are predictable.
02 / A week with a Peugeot 308
The screen hadn’t changed.
My relationship with it had.
I started thinking about this after driving a Peugeot 308 as a courtesy car.
It used a main touchscreen alongside a smaller digital control area for frequently used functions. What surprised me was how quickly that second screen stopped feeling like a screen.
The controls were large. Their positions were stable. After a couple of days, I knew where the important functions were. By the end of the week, I could reach for them with very little visual checking.
That experience did not prove that touchscreens are as safe as physical buttons. It did something more useful for me: it challenged the assumption that a digital control must always require more attention.
I had built a mental map.
03 / The value of constraints
Physical controls come with
predictability almost for free.
A physical control usually stays where it is.
The temperature knob does not disappear because another application has opened. A button does not move because the interface has changed state. Its shape, position and resistance can all become part of memory.
That makes repeated interaction easier to learn. After enough use, the user is no longer searching for the control. They are recalling where it is.
This is where I think physical interfaces have an important advantage: their constraints create stability.
A digital screen has the opposite quality. Its strength is flexibility. The same surface can become navigation, media, climate control, settings or something that did not exist when the vehicle was launched.
But that flexibility also creates a design risk.
If important controls move, disappear, change size, change hierarchy or become buried behind different states, the user cannot build the same reliable mental map.
The problem is not necessarily that the surface is digital. The problem is that the interaction has become unpredictable.
A stable anchor.
Room to change.
Keep essential actions in a familiar place. Let contextual content adapt around them.
Illustrative principle, not a reproduction of the Peugeot interface or a validated control layout.
04 / A familiar compromise
We already learn
digital spaces.
The devices around us are almost entirely digital, yet many interactions still become automatic.
A phone keyboard changes its content constantly, but its basic spatial model remains familiar. Common navigation areas tend to stay where users expect them. On a laptop, I do not need to think consciously about every gesture on the trackpad before I perform it.
These interfaces are flexible, but they still give us enough consistency to learn them.
So I do not think the useful question for vehicles is simply: physical or digital?
A better question might be:
Which interactions need enough stability that the user should be able to build muscle memory around them?
That changes the design conversation.
05 / A division of labour
Not every interaction
should be equally flexible.
Some controls are used frequently and often while the vehicle is moving. Those interactions benefit from stable position, large targets, clear feedback and minimal hierarchy.
Other functions are occasional, contextual or complex. A screen may be much better at handling those because it can reveal information only when it is relevant.
That suggests the answer does not have to be a return to dashboards filled with physical buttons. It could be a more deliberate division of labour.
Physical controls can provide tactile certainty where that certainty matters. Digital interfaces can provide flexibility where flexibility adds value. And digital controls themselves can become more predictable through fixed zones, persistent functions, generous targets, haptic feedback and consistent behaviour.
Voice, steering-wheel controls and other modalities can add another layer, but they should solve a real interaction problem rather than simply adding another way to do the same thing.
The goal is not to make everything physical. It is to make important interactions learnable, stable and trustworthy.
06 / Maybe the debate is framed too narrowly
I still understand why people say they want physical buttons back.
What they may actually be asking for is something slightly deeper: an interface that does not force them to search, interpret and re-learn basic actions every time they use it.
Physical controls achieve that naturally because their form and location are constrained. Digital interfaces have to earn it through design.
That is the part of the debate I find most interesting.
The future question may not be whether a control is made from plastic, glass or pixels. It may be whether the user can predict where it is, what it will do, and what will happen next.
