Reaction time for driving, and why a tap test is not a driver test
· 6 min read · Tyamka team
Reaction time for driving in research runs from about 0.7 to 1.5 seconds, and road design assumes 2.5. Why a tap on a screen is a different task.
Reaction time for driving, as braking research measures it, runs from about 0.7 to 1.5 seconds. A well-known review by Marc Green puts it at 0.70 to 0.75 seconds when a driver knows a signal is coming, about 1.25 seconds for an unexpected but common event such as brake lights ahead, and about 1.5 seconds for a real surprise. American road design assumes 2.5 seconds, a margin meant to cover about 90% of drivers in simple to moderately complex situations. An online tap test gives much shorter numbers. Human Benchmark reports a median of about 273 ms, and our reaction time test measures the same kind of simple reaction on its first levels. The difference comes from the task, not from drivers being slow.
Reaction time for driving in research and road design
| Source | What it describes | Time |
|---|---|---|
| Green, Transportation Human Factors, 2000 | Driver fully aware when and where the signal will come | 0.70 to 0.75 s |
| Green, 2000 | Unexpected but common signal, such as the brake lights of the car ahead | about 1.25 s |
| Green, 2000 | Surprise, such as an object suddenly moving into the driver's path | about 1.5 s |
| AASHTO road design policy, US | Perception and reaction time used to calculate stopping sight distance | 2.5 s |
| UK Highway Code, table of typical stopping distances | Thinking distance of 9 m at 30 mph | about 0.67 s by our calculation |
Green's paper reviews many earlier studies of how long drivers take to get their foot onto the brake. His main point is that expectation matters most. Whether a driver expects the event changes the time by about a factor of two. Age, sex, mental load and urgency change it somewhat too.
AASHTO, the American Association of State Highway and Transportation Officials, uses 2.5 seconds for every design speed. The value is meant to cover about 90% of drivers in simple to moderately complex situations, so it is deliberately longer than a typical driver's time. Engineers use it to decide how far ahead a driver must be able to see, not to describe how fast any one person reacts.
The UK Highway Code does not state a reaction time. Its table of typical stopping distances, which we checked as reproduced by the RAC, gives 9 metres of thinking distance at 30 mph. At that speed a car covers about 13.4 metres per second, so 9 metres comes to about two thirds of a second. That division is ours.
The numbers differ because they answer different questions. One set describes drivers in experiments, one is a safety margin for designing roads, and one is a teaching figure. None of them is a norm to check yourself against with a phone.
Why braking takes longer than a tap
In a simple tap test much of the work is done for you before the signal. You know what the signal looks like and what to do when it comes, and your finger is ready. On the later levels of our game you also have to see where the circle is, and hold back when a square appears, but the task is still small and clearly defined.
On the road it is different.
- You do not know what the signal is. It might be brake lights, a pedestrian, a cyclist or a car pulling out. First you have to notice that anything has happened.
- You have to decide. Brake, steer, sound the horn or do nothing. A response chosen from several takes longer than a single known one, which is Hick's law from William Hick's work in 1952.
- The movement is bigger. Your foot has to move from the accelerator to the brake pedal, a much longer movement than pressing a finger.
- You may be looking elsewhere. A mirror, the speedometer, a sign, a passenger. The event may start while your eyes are away from it.
A simple tap test takes most of this away. That is why it can measure a simple reaction, and why it says little about what happens in a car.
What a few tenths of a second mean on the road
This table shows how far a car at 50 km/h travels before the driver's foot reaches the brake, for different reaction times. At that speed the car moves about 13.9 metres per second. It is plain arithmetic and not a stopping distance, because the distance covered while braking comes on top.
| Reaction time | Distance at 50 km/h before braking starts |
|---|---|
| 0.27 s, the Human Benchmark median for a click | about 3.8 m |
| 0.70 to 0.75 s, an expected signal (Green) | about 9.7 to 10.4 m |
| 1.25 s, brake lights ahead (Green) | about 17.4 m |
| 1.5 s, a surprise (Green) | about 20.8 m |
| 2.5 s, the design value (AASHTO) | about 34.7 m |
At 50 km/h, the gap between a click on a screen and a surprise on the road is about 17 metres, by subtracting the rows. It reflects the difference between the two tasks, not the difference between one person and another.
Why a tap test is not a check for drivers
People search for a reaction test for drivers, and some sites turn a single tap into a verdict about driving. We do not, for these reasons.
- A simple reaction test measures one narrow thing, the time to respond to a known signal when you are ready for it. Driving also depends on noticing hazards, predicting what others will do and choosing a response.
- The device adds its own delay. In one lab study the web page added 58 to 133 ms on average, depending on the device and browser. The article on reaction time on a phone and a computer explains this.
- The differences between people in a tap test are small next to the effect of expectation that Green describes.
- A game is not a medical test and not a driving assessment. If you have doubts about your fitness to drive, talk to a doctor.
So a fast time in our game does not mean you drive safely, and a slow time does not mean you drive dangerously. We draw no conclusion either way.
What a test can show
A test like ours can show your reaction time on the device you are using, and how much it varies from try to try. In our game the circle comes after a random pause of one to four seconds, and a tap before it or less than 0.12 s after it is a false start that costs one of three lives. From level 4 the circle appears in one of two places and later four, and from level 10 a square sometimes appears instead, which you must not tap.
The article on the reaction time test average gives the numbers from research and from a large online test, and explains why your own time changes between attempts.