Trail Making Test (TMT)

Evidence Reviewed as of before: 22-04-2012
Author(s): Katie Marvin, MSc. PT
Editor(s): Nicol Korner-Bitensky, PhD OT; Annabel McDermott, OT

Purpose

The Trail Making Test (TMT) is a widely used test to assess executive function in patients with stroke. Successful performance of the TMT requires a variety of mental abilities including letter and number recognition mental flexibility, visual scanning, and motor function.

In-Depth Review

Purpose of the measure

The Trail Making Test (TMT) is a widely used test to assess executive abilities in patients with stroke. Successful performance of the TMT requires a variety of mental abilities including letter and number recognition mental flexibility, visual scanning, and motor function.

Performance is evaluated using two different visual conceptual and visuomotor tracking conditions: Part A involves connecting numbers 1-25 in ascending order; and Part B involves connecting numbers and letters in an alternating and ascending fashion.

Available versions

The TMT was originally included as a component of the Army Individual Test Battery and is also a part of the Halstead-Reitan Neuropsychological Test Battery (HNTB).

Features of the measure

Description of tasks:

The TMT is comprised of 2 tasks – Part A and B:

  • Part A: Consists of 25 circles numbered from 1 to 25 randomly distributed over a page of letter size paper. The participant is required to connect the circles with a pencil as quickly as possible in numerical sequence beginning with the number 1.
  • Part B: Consists of 25 circles numbered 1 to 13 and lettered A to L, randomly distributed over a page of paper. The participant is required to connect the circles with a pencil as quickly as possible, but alternating between numbers and letters and taking both series in ascending sequence (i.e. 1, A, 2, B, 3, C…).

What to consider before beginning:

  • The TMT requires relatively intact motor abilities (i.e. ability to hold and maneuver a pen or pencil, ability to move the upper extremity. The Oral TMT may be a more appropriate version to use if the examiner considers that the participant’s motor ability may impact his/her performance.
  • Cultural and linguistic variables may impact performance and affect scores.

Scoring and Score Interpretation:

Performance is evaluated using two different visual conceptual and visuomotor tracking conditions: Part A involves connecting numbers 1-25 in ascending order; and Part B involves connecting numbers and letters in an alternating and ascending fashion.

Time taken to complete each task and number of errors made during each task are recorded and compared with normative data. Time to complete the task is recorded in seconds, whereby the greater the number of seconds, the greater the impairment.

In some reported methods of administration, the examiner pointed out and explained mistakes during the administration.

A maximum time of 5 minutes is typically allowed for Part B. Participants who are unable to complete Part B within 5 minutes are given a score of 300 or 301 seconds. Performance on Part B has not been found to yield any more information on stroke severity than performance on Part A (Tamez et al., 2011).

Ranges and Cut-Off Scores
Normal Brain-damage
TMT Part A 1-39 seconds 40 or more seconds
TMT Part B 1-91 92 or more seconds

Adapted from Reitan (1958) as cited in Matarazzo, Wiens, Matarazzo & Goldstein (1974).

Time:

Approximately 5 to 10 minutes

Training requirements:

No training requirements have been reported.

Equipment:

  • A copy of the measure
  • Pencil or pen
  • Stopwatch

Alternative versions of the Trail Making Test

  • Color Trails (D’Elia et al., 1996)
  • Comprehensive Trail Making Test (Reynolds, 2002)
  • Delis-Kaplan Executive Function Scale (D-KEFS) – includes subtests modeled after the TMT
  • Oral TMT – an alternative for patients with motor deficits or visual impairments (Ricker & Axelrod, 1994).
  • Repeat testing – alternate forms have been developed for repeat testing purposes (Franzen et al., 1996; Lewis & Rennick, 1979)
  • Symbol Trail Making Test – developed as an alternative to the Arabic version of the TMT, for populations with no familiarity with the Arabic numerical system (Barncord & Wanlass, 2001)

Client suitability

Can be used with:

Patients with stroke and brain damage.

Should not be used with:

Patients with motor deficiencies. If motor ability may impact performance, consider using the Oral TMT.

In what languages is the measure available?

Arabic, Chinese and Hebrew

Summary

What does the tool measure? Executive function in patients with stroke.
What types of clients can the tool be used for? The TMT can be used with, but is not limited to, patients with stroke.
Is this a screening or assessment tool? Assessment tool
Time to administer The TMT takes approximately 5 to 10 minutes to administer.
Versions
  • Color Trails
  • Comprehensive Trail Making Test
  • Delis-Kaplan Executive Function Scale (D-KEFS)
  • Oral TMT
  • Repeat testing – alternate forms have been developed for repeat testing purposes
  • Symbol Trail Making Test
Other Languages Arabic, Chinese and Hebrew
Measurement Properties
Reliability

Test-retest:

Two studies examined the test-retest reliability of the TMT among patients with stroke and found adequate to excellent test-retest reliability.

Validity

Content:

One study examined the content validity of the TMT and found it to be a complex test that involves aspects of abstraction, visual scanning and attention.

Criterion:

Predictive:

Several studies have examined the predictive validity of the TMT and have found Part B to be predictive of fitness to drive following stroke.

Construct:

Convergent:

One study examined the convergent validity of the TMT and found poor to adequate correlations with the Category Test, Wisconsin Card Sort Test, Paced Auditory Serial Addition Task and the Visual Search and Attention Test.

Known groups:

Three studies have examined the known groups validity of the TMT and found that the TMT was able to differentiate between patients with and without brain damage however, it was not sensitive to differentiating between front and non-frontal brain damage.

Floor/Ceiling Effects One study found Part A of the TMT to have significant ceiling effects.
Does the tool detect change in patients? The responsiveness of the TMT has not formally been studied, however the TMT has been used to detect changes in a clinical trial with patients with stroke.
Acceptability The TMT is simple and easy to administer.
Feasibility The TMT is relatively inexpensive and highly portable. The TMT is public domain and can be reproduced without permission. It can be administered by individuals with minimal training in cognitive assessment.
How to obtain the tool? The Trail Making Test (TMT) can be purchased from: http://www.reitanlabs.com

Psychometric Properties

Overview

A literature search was conducted to identify all relevant publications on the psychometric properties of the Trail Making Test (TMT) involving patients with stroke.

Floor/Ceiling Effects

In a study by Mazer, Korner-Bitensky and Sofer (1989) that investigated the ability of perceptual testing to predict on-road driving outcomes in patients with stroke, part A of the TMT was found to have significant ceiling effects. For this reason, Part A was excluded from study results as it was deemed too easy for participants when evaluating the ability of the TMT to predict on-road driving test outcomes. No ceiling effects for part B were found.

Reliability

Internal consistency:

No studies were identified examining the internal consistency of the TMT in patients with stroke.

Test-retest:

Matarazzo, Wiens, Matarazzo and Goldstein (1974) examined the test-retest reliability of the TMT and other components of the Halstead Impairment Index with 29 healthy males and 16 60-year old patients with diffuse cerebrovascular disease. Adequate test-retest reliability was found for both Part A and Part B of the TMT in the healthy control group (r=0.46 and 0.44 respectively), as calculated using Pearson correlation coefficients. Excellent and adequate test-retest reliability was found for Part A and Part B of the TMT respectively (r=0.78 and 0.67), among participants with diffuse cerebrovascular disease.

Goldstein and Watson (1989) investigated the test-retest reliability of the TMT as a part of the Halstead- Reitan Battery in a sample of 150 neuropsychiatric patients, including patients with stroke. Test-retest correlations were calculated using Pearson Correlation Coefficients for the entire sample and for the sub-group of patients with stroke. Excellent test-retest reliability for both Part A and Part B were found (0.94 and 0.86 respectively) in the sub-group of patients with stroke; and adequate reliability for the entire participant sample (0.69 and 0.66 respectively).

Intra-rater:

No studies were identified examining the intra-rater reliability of the TMT in patients with stroke.

Inter-rater:

No studies were identified examining the inter-rater reliability of the TMT in patients with stroke.

Validity

Content:

O’Donnell, MacGregor, Dabrowski, Oestreicher & Romero (1994) examined the face validity of the TMT in a sample of 117 community-dwelling patients, including patients with stroke. The results suggest that the TMT is a complex test that involves aspects of abstraction, visual scanning and attention.

Criterion:

Concurrent:

No studies were identified examining the concurrent validity of the TMT.

Predictive:

Mazer, Korner-Bitensky and Sofer (1998) examined the ability of the TMT and other measures of perceptual function to predict on-road driving test outcomes in 84 patients with subacute stroke. For Part B of the TMT, a cut-off score of < 3 errors demonstrated high positive predictive value (85%) and low negative predictive value (48%) for successful completion of driving evaluation. The Motor Free Visual Perception Test (MFVP) and the TMT Part B, when combined, demonstrated the highest predictive value for on-road driving test outcome. Participants who scored poorly on both the MFVP and TMT Part B had 22 times the likelihood of failing the on-road evaluation.

Devos, Akinwuntan & Nieuwboer (2011) conducted a systematic review to identify the best determinants of fitness to drive following stroke. The TMT Part B was evaluated in 2 studies (Mazer et al., 1998 and Mazer et al., 2003) and found to be one of the best predictors of passing on-road driving evaluation tests (effect size = 0.81, p<0.0001). In addition, when using a cutoff score of 90 seconds, the TMT Part B had a sensitivity of 80% and a specificity of 62% for detecting unsafe on-road performance. In a subsequent systematic review by Marshall et al. (2007), the TMT was, again, found to be one of the most useful predictors of fitness for driving post-stroke.

Construct:

Convergent/Discriminant:

O’Donnell et al. (1994) examined the convergent validity of the TMT and four other neuropsychological tests: Category Test (CAT), Wisconsin Card Sort Test (WCST), Paced Auditory Serial Addition Task (PASAT), and Visual Search and Attention Test (VSAT). The study involved 117 community-dwelling adults, including patients with stroke. Poor to adequate correlations were found between the TMT and the other measures (CAT r=0.38; WCST r=0.31; PASAT r=0.44; VAST r=0.30), using Pearson product-moment correlations.

Known groups:

Reitan (1955) examined the ability of the TMT to differentiate between patients with and without organic brain damage, including patients with stroke. Highly significant differences in mean and sum scores were found between the two groups (p<0.001) on both parts of the TMT, suggesting that the TMT is able to different between patients with and without brain damage.

Corrigan and Hinkeldey (1987) examined the relationship between Part A and Part B of the TMT. Data was collected from the charts of 497 patients receiving treatment at a rehabilitation centre. Patients with traumatic brain injury and stroke comprised a large majority of the sample. A difference (B-A) and a ratio (B/A) score were calculated. The difference score was highly correlated with intelligence and severity of impairment and only moderately correlated with age, education and memory functioning. The B/A ratio appeared to show greatest sensitivity to differences in cerebral lateralization of damage.

Tamez et al. (2011) examined the effects of frontal versus non-frontal stroke and severity of stroke on TMT performance in 689 patients with stroke. The TMT, Digit Span and National Institute of Health Stroke Scale (NIHSS) were administered within 72 hours of hospital admission. Stroke severity was classified according to the NIHSS, and frontal or non-frontal lesions by CT or MRI scans. Performance on both Part A and Part B of the TMT were significantly correlated with stroke severity using the NIHSS. Patients with frontal and non-frontal lesions were found to score equally on Part A and Part B (p>0.05). Results of this study suggest that the TMT is sensitive to brain damage, however, there is little evidence to support the widely held assumption that Trails B is more sensitive to frontal lesions than Part A.

Sensitivity/ Specificity:

No studies were identified examining the specificity of the TMT in patients with stroke.

Responsiveness

Barker-Collo, Feigin, Lawes, Senior and Parag (2000) assessed the course of recovery of attention span in 43 patients with acute stroke over a 6-month period. The TMT and other measures of attention were administered at baseline (within 4 weeks following stroke onset), 6 weeks, and 6 months after stroke. Although the responsiveness of the TMT was not formally assessed in this study, the scale was sensitive enough to detect an improvement in attention at 6 weeks and 6 months following stroke.

References

  • Barker-Collo, S., Feigin, V., Lawes, C., Senior, H., & Parag, V. (2010). Natural history of attention deficits and their influence on functional recovery from acute stages to 6 months after stroke.Neuroepidemiology, 35(4), 255-262.
  • Barncord, S.W. & Wanlass, R.L. (2001). The Symbol Trail Making Test: Test development and utility as a measure of cognitive impairment. Applied Neuropsychology, 8, 99-103
  • Corrigan, J. D. & Hinkeldey, N. S. (1987). Relationships between Parts A and B of the Trail Making Test. Journal of Clinical Psychology, 43(4), 402-409.
  • D’Elia, L.F., Satz, P., Uchiyama, C.I. & White, T. (1996). Color Trails Test. Odessa, Fla.:PAR.
  • Devos, H., Akinwuntan, A. E., Nieuwboer, A., Truijen, S., Tant, M., & De Weerdt, W. Screening for fitness to drive after stroke: a systematic review and meta-analysis.Neurology, 76(8), 747-756.
  • Elkin-Frankston, S., Lebowitz, B.K., Kapust, L.R., Hollis, A.M., & O’Connor, M.G. (2007). The use of the Colour Trails Test in the assessment of driver competence: Preliminary reports of a culture-fair instrument. Archives of Clinical Neuropsychology, 22, 631-635.
  • Goldstein, G. & Watson, J.R. (1989). Test-retest reliability of the Halstead-Reitan Battery and the WAIS in a Neuropsychiatric Population. The Clinical Neuropsychologist, 3(3), 265-273.
  • O’Donnell, J.P., Macgregor, L.A., Dabrowski, J.J., Oestreicher, J.M., & Romero, J.J. (1994). Construct validity of neuropsychological tests of conceptual and attentional abilities. Journal of Clinical Psychology, 50(4), 596-560.
  • Mark, V. W., Woods, A. J., Mennemeier, M., Abbas, S., & Taub, E. Cognitive assessment for CI therapy in the outpatient clinic.Neurorehabilitation, 21(2), 139-146.
  • Marshall, S.C., Molnar, F., Man-Son-Hing, M., Blair, R., Brosseau, L., Finestone, H.M., Lamothe, C, Korner-Bitensky, N., & Wilson, K. (2007). Predictors of driving ability following stroke: A systematic review. Topics in Stroke Rehabilitation, 14(1):98-114.
  • Matarazzo, J.D., Wiens, A.N., Matarazzo, R.G., & Goldstein, S.G. (1974). Psychometric and clinical test-retest reliability of the Halstead Impairment Index in a sample of healthy, young, normal men. The Journal of Nervous and Mental Disease, 188(1), 37-49.
  • Mazer, B.L., Korner-Bitensky, N.A., & Sofer, S. (1998). Predicting ability to drive after stroke. Archives of Physical Medicine and Rehabilitation, 79, 743-750.
  • Mazer, B.L., Sofer, S., Korner-Bitensky, N., Gelinas, I., Hanley, J. & Wood-Dauphinee, S. (2003). Effectiveness of a visual attention retraining program on the driving performance of clients with stroke. Archives of Physical Medicine and Rehabilitation, 84, 541-550.
  • Reitan, R.M. (1955). The relation of the Trail Making Test to organic brain damage. Journal of Consulting Psychology, 19(5), 393-394.
  • Reynolds, C. (2002). Comprehensive Trail Making Test. Austin, Tex,: Pro-Ed.
  • Ricker, J.H. & Axelrod, B.N. (1994). Analysis of an oral paradigm for the Trail Making Test. Assessment, 1, 47-51.
  • Strauss, E., Sherman, E.M.S., & Spreen, O. (2006).A Compendium of neuropsychological tests: Administration, norms, and commentary.(3rd. ed.).NY. Oxford University Press.
  • Tamez, E., Myersona, J., Morrisb, L., Whitea, D. A., Baum C., & Connor, L. T. (2011). Assessing executive abilities following acute stroke with the trail making test and digit span.Behavioural Neurology, 24(3), 177-185.

See The Measure

How to obtain the Trail Making Test (TMT)?

The Trail Making Test (TMT) can be purchased from:

Reitan Neuropsychology Laboratory
P.O. Box 66080
Tucson, AZ
85728

http://www.reitanlabs.com

Table of contents
Survey