Trail Making Test: What It Measures and How It Works

Why the Trail Making Test Matters

If a neurologist, memory clinic, or neuropsychologist has mentioned a timed "trails" task, this is almost certainly it. The Trail Making Test has been in use for decades and remains one of the most widely administered neuropsychological measures in the world, in part because it is quick, needs only pencil and paper, and gives an examiner information on a wide range of cognitive skills (Bowie & Harvey, 2006).

It matters to families because the test does something clever with very little: by comparing how a person performs on a simple sequencing task versus a more demanding one, it gives a clinician a window into the parts of thinking that often shift early. Understanding what it does — and does not — measure helps you make sense of a result and ask better questions. If you are new to the broader topic, our overview of cognitive testing explains where brief tasks like this fit alongside other tools.

What Does "Trail Test" Mean?

"Trail test," "Trails," and "TMT" are all shorthand for the Trail Making Test. The name describes the task: when a person connects a series of scattered points in the correct order without lifting the pencil, the result is one continuous line — a trail — winding across the page.

Clinicians usually refer to the two parts as "Trails A" and "Trails B." When you see those labels in a visit summary or a neuropsychology report, they refer to the two timed parts of this one test, not to two separate tests.

What Does the Trail Making Test Measure?

Rather than testing one ability in depth, the Trail Making Test briefly samples a cluster of related skills. A recent review describes the split clearly: Part A evaluates visuoperceptual tracking and processing speed, while Part B adds divided attention, set-shifting, and cognitive flexibility (Tsiakiri et al., 2024).

In everyday language, that means:

• Visual scanning — the eyes' ability to search a page and find the next target quickly.
• Processing speed — how fast a person can take in information and act on it.
• Attention — staying focused and not losing the thread mid-task.
• Working memory — keeping track of where you are in two sequences at once.
• Set-shifting and mental flexibility — switching back and forth between two rules, which is the extra demand Part B introduces.

Researchers have tested these assumptions directly. One study of healthy older adults found that Part A relies mainly on visuoperceptual abilities, Part B relies primarily on working memory and secondarily on task-switching, and the difference between the two times minimizes visuoperceptual and working memory demands, leaving a relatively pure indicator of executive control (Sánchez-Cubillo et al., 2009).

These skills fall under what clinicians call executive function — the brain's "management" system for planning, switching, and staying on track. Because Part B leans on executive function so heavily, the relationship between the two parts is often more informative than either part alone.

How the Test Works (Part A and Part B)

The format is simple to picture. In Part A, a person draws a line connecting a series of numbered circles in ascending order, as quickly and accurately as they can. In Part B, the circles contain both numbers and letters, and the person alternates between the two sequences — number, letter, number, letter — again working against the clock (Bowie & Harvey, 2006).

That alternating rule in Part B is the whole point: it forces the brain to hold two sequences in mind and flip between them, which is far more demanding than counting up in a single line. A clinician times each part and notes any errors, pointing them out so the person can get back on track and the task keeps moving.

The test is professionally administered using standardized forms, instructions, and timing, and it is scored by trained clinicians rather than practiced at home — rehearsing a screening task can distort exactly what it is designed to reveal. It is one of many types of cognitive tests a clinician may draw on.

Trail Making A and B: What's the Difference?

Both parts use the same basic motion — drawing a line from point to point — but the key change between them is the mental rule. That design is what makes the pair useful:

• Part A is the baseline. It captures how quickly someone can scan, find, and connect targets when the rule is simple.
• Part B adds switching between two sequences. Its layout and visual search are not identical to Part A's, so the extra time is not a pure switching cost, but much of it reflects the demand of holding two sequences and alternating between them.

Part B nearly always takes longer than Part A, for everyone. What a clinician watches is whether the gap is larger than expected for the person's age and education, and whether errors cluster in Part B. A person can have an ordinary Part A time and a notably slow Part B time, which points more toward executive function than toward general slowing.

How Scores Are Interpreted

The Trail Making Test is scored primarily by time — how many seconds it takes to complete each part — rather than by a tally of right answers. Longer completion times, especially a large gap between Part A and Part B, can point toward slowed processing or reduced mental flexibility.

Crucially, there is no universal "pass" number. Performance changes naturally with age and education: a normative study of 911 community-dwelling adults aged 18 to 89 found that performance declined with older age and fewer years of schooling, which is why clinicians compare a result against norms stratified by age and education rather than a single cutoff (Tombaugh, 2004). A time that is perfectly typical for one person could be unusual for another. For a broader explanation of why this matters for any test, see our guide to age-adjusted cognitive scores, and for reading results in context, how to understand cognitive test results.

Beyond Raw Time: The B/A Ratio and B−A Difference

Many clinicians also calculate "derived" scores that compare the two parts directly. The two most common are the difference score (Part B time minus Part A time) and the ratio score (Part B time divided by Part A time). The idea is that dividing or subtracting removes much of the influence of general speed and motor ability, isolating the switching demand (Tsiakiri et al., 2024).

There is research support for this. In one study, people's "switch cost" on a separate laboratory task-switching test correlated with Part B performance, especially with the B/A ratio, which the authors concluded provides an index of executive function (Arbuthnott & Frank, 2000).

Derived scores are not immune to age, though. In a normative study of 285 healthy adults aged 18 to 90, the B/A ratio averaged about 2.8, and both derived scores were related to age — the difference score more strongly. The authors cautioned that previously suggested ratio cutoffs could misclassify healthy older adults as impaired (Drane et al., 2002). That is one more reason these numbers are interpreted by a clinician against appropriate norms, not read off a chart.

Do Errors Matter on the Trail Making Test?

Yes. Time is the headline score, but the kinds and number of errors carry their own information. Errors in Part B — such as losing the sequence or failing to switch — are thought to shed light on attention and cognitive flexibility (Tsiakiri et al., 2024).

A study of 526 older adults, each classified by a multidisciplinary consensus team, illustrates this. People with mild cognitive impairment and people with Alzheimer's disease took longer on both parts than cognitively normal peers and made more Part B errors: roughly 30% of the normal group made at least one Part B error, compared with about 59% of the MCI group and 72% of the Alzheimer's group. The authors concluded that Part B time and errors are independently meaningful scores in dementia evaluations (Ashendorf et al., 2008).

Notice what that finding also shows: nearly a third of cognitively normal older adults made an error, too. A single slip is common and is not, on its own, a sign of a problem.

The Trail Making Test and Mild Cognitive Impairment

Because it samples processing speed and executive function, the Trail Making Test is often part of the workup when someone is being evaluated for mild cognitive impairment. In the study above, both Part A and Part B times followed a clear gradient from normal cognition to MCI to Alzheimer's disease at the group level (Ashendorf et al., 2008).

Group differences do not translate neatly into individual answers, however. Some people with MCI perform within the expected range, and some healthy people perform slowly on a given day. That is why clinicians use the test as one piece of a battery — often alongside memory and language tasks such as the verbal fluency test — rather than as a stand-alone verdict.

Trails B and Driving Evaluations

Part B also shows up in a setting families may not expect: questions about whether an older adult can keep driving safely. Fitness-to-drive guidelines recommend using Trails B as one input, and a systematic review of 47 studies examined which cutoffs best relate to driving safety. The authors found some support for a commonly cited time-and-error threshold but also major methodological limitations across the research, and debate about the optimal cutoff continues (Roy & Molnar, 2013).

In practice, a Trails B result is a flag for a clinician to weigh, not a driving decision by itself. When concerns about driving come up, clinicians typically consider the whole picture and may refer for a specialized driving evaluation.

Can the Test Be Given Without Pencil and Paper?

Yes. For people who cannot manage the written version — for example, because of vision problems, orthopedic injuries, or conditions that affect movement — clinicians can use an oral version, in which the person says the alternating sequence aloud instead of drawing it. The oral form removes the visual and motor components while still assessing sequencing and set-shifting. Its Part B correlates strongly with the written Part B, but the two versions are not interchangeable score-for-score, so each needs its own norms (Mrazik et al., 2010).

The same logic explains why a clinician will ask about eyesight, hand or arm problems, and similar factors before interpreting a written result: slowing caused by difficulty seeing or moving is not the same thing as slowing in thinking.

When It's Used

Clinicians reach for the Trail Making Test in a range of situations — as part of a broader cognitive workup, when screening for changes in attention or executive function, or when tracking recovery after an event such as a stroke or head injury. A recent review of stroke research, for example, describes it as a widely used measure of processing speed and mental flexibility after stroke, including as a marker of recovery over time (Tsiakiri et al., 2024).

It rarely travels alone. More often it is one component of a larger battery, and when a deeper picture is needed it appears within a full neuropsychological evaluation that combines many tasks. Another familiar pencil-and-paper task a clinician may pair it with is the clock drawing test, which samples a different but overlapping set of skills.

What Happens Next

A slower-than-expected time on the Trail Making Test is a prompt, not a verdict. On its own, it cannot diagnose dementia, Alzheimer's disease, or any other condition. As the Alzheimer's Association explains, there is no single test that determines whether someone is living with Alzheimer's or another dementia — clinicians combine cognitive assessments with medical history, physical and neurological exams, and sometimes brain imaging or lab work (Alzheimer's Association).

When a result raises questions, a clinician may look for treatable contributors — such as thyroid problems, vitamin deficiencies, poor sleep, stress, or medication side effects — before drawing any conclusions. They may also repeat testing later to see whether performance is stable or changing. The aim at every step is clarity and next steps, not alarm.

Taking the Next Step

To see how this task compares with other tools a clinician might use, start with our guide to the different types of cognitive tests.

If you'd like a structured, repeatable way to check in on memory and thinking from home between clinician visits, explore how Orena's clinically validated at-home test works.