Combatting ‘Highway Hypnosis’

Picture a small, familiar moment. You’re an hour into a highway drive. Music’s playing. A while back you turned off the navigation voice, because every time it spoke it ducked your music to tell you something you already knew. The road is straight and empty and your mind has drifted somewhere pleasant.

Then you notice the sign for your exit, and you notice it too late. You needed to be a lane or two over already, and now you have a bad choice to make. Either you let the exit go, or you cut across two or three lanes of highway traffic in a few seconds to still catch it, which is a dangerous move for you and for everyone driving around you.

That moment is the whole problem, and the reason it happens is worth taking apart. It’s not carelessness. It’s two systems failing at the same time.

Two channels, both down

A driver has two ways of being told an exit is coming: the car tells them through audio, or they notice it themselves through attention. On a long highway drive, both are often unavailable at once.

The audio channel gets switched off on purpose. Drivers mute turn by turn voice because it interrupts music and podcasts, and it’s one of the most common complaints about in car navigation.

The attention channel wears down on its own. Long, monotonous highway driving produces a well documented state called highway hypnosis↗, where a driver stays awake and technically in control but starts running on autopilot, navigating from memory and prediction instead of what’s actually in front of them. Reaction to changes in the environment, like an approaching exit, slows down in a measurable way.

So here’s the failure, stated plainly. On exactly the drives where a driver’s attention is most dulled, they’ve often turned off the only system still trying to warn them. Audio off, attention down. That’s the gap I set out to design for.

It’s also why this feels like a Tesla problem specifically. Teslas have real long distance highway confidence now, and that confidence is traced almost entirely to charging infrastructure. Tesla's users are doing exactly the kind of long, monotonous highway stretch that triggers this in the first place.

The maps answered where: attention thins out on exactly the stretches where exit decisions pile up. But geography isn’t the failure. The interface is. The real question is what happens in the cabin in the last mile before an exit, and how the car tells you it’s coming. That’s where Tesla’s design runs into trouble.

How everyone else cues an exit

Modern navigation does three things at once as an exit approaches. A visual turn card that stays on screen, lane guidance showing which lane to be in, and audio that escalates, with a heads-up at two miles, again at one, once more right before. Google Maps and Waze do all three, and through CarPlay or Android Auto they sit in the driver’s line of sight. Even a basic built-in system puts the next turn in the instrument cluster, inches under where you’re already looking.

Tesla opts out of all of it.

This is a tradeoff, not a flaw

It’s worth being fair here, because Tesla’s choice is deliberate and mostly a good one. The single screen is core to the brand. A clean, minimal cabin, no cluttered cluster, one surface that updates over the air. It’s simpler to build, easier to iterate, and unmistakably Tesla. For almost all driving, it works.

But every design philosophy has an edge case where its strength turns into a cost, and highway exits are Tesla’s. The same centralization that makes the cabin elegant means there’s no cluster to fall back on, so the one moment you need a fast, glanceable cue in your sightline is the one moment a single screen serves worst. The next turn ends up in the top corner of the center display, the farthest point from the road, because that’s just where a unified screen puts it.

This isn’t a data problem or a competence problem. Tesla has the map data and the engineering to do anything. It’s that a philosophy tuned for simplicity was never tuned for this specific glance.