Trail Making Test online: Parts A and B
Click 25 circles in order, then 25 more while alternating numbers and letters. Your times show scanning speed and the cost of switching.
After Part B you see your result and can keep it with a free account.
This task needs a mouse or trackpad and a larger screen. Open this page on a computer to take it.
Start the Stroop testStroop works on a phone: name the ink color and ignore the word.
How it works
2 timed parts on the same kind of board, each with a short practice first. About 2 minutes in all.
- 1 · Part A
Numbers
Click the circles from 1 to 25 in order, as fast as you can.
- 2 · Part B
Numbers and letters
Alternate between the two sequences: 1, A, 2, B, 3, C and on to 13. That is 25 circles again.
- 3 · Before each part
Practice, then the clock
An 8-circle practice and a 3-second countdown come first. The clock starts with your first click on a circle and stops at the last one.
The board fits your screen, and the next circle in the sequence is outlined, as in the practice. A wrong click flashes red and is counted, then you carry on from where you were.
Trail Making began in the Army Individual Test Battery of 1944 and later became part of the Halstead-Reitan battery. This version keeps its two parts and its 25-circle trails.
What it measures
Both parts ask you to find the next circle, move to it and click. Part B adds a switch on every move.
Part A
- Visual scanning: finding the next circle among the others on the board.
- Processing speed: how quickly you move through a simple, familiar sequence.
- Pointer control: reaching and clicking each circle accurately.
Part B
- Everything Part A asks for, on a new layout.
- Switching: moving from counting to the alphabet and back on every click.
- Keeping your place in two sequences at once, which draws on working memory.
Subtracting Part A from Part B removes most of the shared speed and movement, so the difference is read as the cost of switching between sequences. Researchers also call this set-shifting or cognitive flexibility.
A test, not a diagnosis.
Your times describe one attempt at one task, and they cannot diagnose or rule out any condition.
What you get
Your result appears the moment you finish Part B.
On the result screen
- Your time for Part A and for Part B, in seconds.
- The B minus A difference, which shows what switching cost you, and the B to A ratio.
- Your errors in each part.
- Size-adjusted times, so results from small and large screens compare, with the raw times alongside.
- A CSV file of every click, if you want the raw data.
When you keep it
- Saved to your results with a free account, on any device.
- Shown beside your other tests as you add them, so you can compare areas of thinking.
- A suggested next test right after, so your results keep building.
How this version is scored
The runner logs every click and its time. The result screen is built from that log.
- Completion time
- From your first click on a circle to your click on the last one, for each part. Practice runs are not scored, and each part stops after 5 minutes.
- Errors
- A click on any circle other than the next one. It flashes red and adds 1 to the count for that part.
- B minus A
- Part B time minus Part A time, in seconds. The bigger the gap, the more the switching slowed you down.
- B to A ratio
- Part B time divided by Part A time. It puts the switching cost in proportion to your own speed.
- Size-adjusted time
- On a smaller screen the circles sit closer together, so the runner rescales each time as size-adjusted time = raw time × reference path ÷ ideal path, where the ideal path is the straight-line route through your layout and the reference path is the average route over 20 layouts on a standard 800 × 600 board.
Why errors do not stop the clock. In the paper test an examiner points out each error straight away and you return to the last correct circle while the clock runs, so the time spent correcting is part of the score. This version follows the same rule without an examiner: a wrong click costs you time, and it is also counted on its own.
Raw and adjusted times. The result screen leads with size-adjusted times. Your raw times appear in its full data panel and in the CSV, and both are saved with your result.
Reference values
Published averages for the paper-and-pencil test, from 911 community-dwelling adults aged 18 to 89 (Tombaugh, 2004).
| Age | Part A | Part B |
|---|---|---|
| 18 to 24 | 23 | 49 |
| 25 to 34 | 24 | 51 |
| 35 to 44 | 29 | 58 |
| 45 to 54 | 32 | 64 |
| 55 to 59 | 35 | 79 |
| 60 to 64 | 33 | 75 |
| 65 to 69 | 39 | 91 |
| 70 to 74 | 42 | 110 |
| 75 to 79 | 51 | 131 |
| 80 to 84 | 58 | 153 |
From age 55, Tombaugh reports averages separately by education. The rows from 55 on are for adults with up to 12 years of schooling; adults with more schooling were faster, especially on Part B.
Not directly comparable. These are paper times: a pencil line drawn while an examiner watches and corrects errors. Here you click instead of drawing, the next circle is outlined, the board follows your screen and nobody is watching. Each of these changes the times.
The result screen shows your own times in seconds. It does not turn them into a percentile against this table.
Before you start
A steady setup makes your times easier to read.
For a cleaner attempt
- Use a computer with a mouse or trackpad you know well.
- Make the browser window large, so the board has room.
- Sit somewhere quiet and stay on the tab until both parts are done.
- Use the practice to settle on a pace you can keep.
Interpret cautiously when
- You are tired, unwell, rushed or distracted.
- Your vision, hand movement or pointer setup makes clicking harder.
- You did the task recently; practice alone makes it faster.
- The alphabet is less familiar to you, for example in a second language.
Questions
What does Part B of the Trail Making Test measure?
Part B asks for everything Part A does and adds a switch on every move, from a number to a letter and back. The extra time it takes over Part A is read as the cost of switching, often called set-shifting or cognitive flexibility. Keeping your place in two sequences also draws on working memory.
What is a normal time on the Trail Making Test?
On the paper version, adults aged 18 to 24 averaged about 23 seconds on Part A and 49 seconds on Part B in Tombaugh's 2004 sample, and times rise with age. A browser version with a mouse is not directly comparable, so use those averages as context, not as a cut-off. Your result screen shows your own times rather than a percentile.
Can I do the Trail Making Test on my phone?
No. The task needs precise pointing and room for 25 circles, so the tests hub asks for a computer with a mouse or trackpad. On a phone you can start with the Stroop test and come back to this one on a computer.
How often can I repeat the Trail Making Test?
As often as you like: you can retake it from the tests hub. Times usually improve with repetition because the task gets familiar, so a faster second run is not always a real change. Compare attempts made in similar conditions, with some time between them.
Sources
- Tombaugh (2004): Trail Making Test A and B, normative data stratified by age and education
- Reitan (1958): Validity of the Trail Making Test as an indicator of organic brain damage
- Bowie and Harvey (2006): Administration and interpretation of the Trail Making Test, Nature Protocols
- Strauss, Sherman and Spreen (2006): A Compendium of Neuropsychological Tests, 3rd edition
- Giovagnoli et al. (1996): Trail Making Test, normative values from 287 Italian adults
- Germine et al. (2012): Is the Web as good as the lab? Web and lab performance compared