In the footsteps of a neuropsychological test classic – The Trail Making Test

There are few tests that are as widely used as the Trail Making Test (TMT). The test is one of the most often used neuropsychological tests in the world. But what makes the TMT so popular? What does it actually measure? How did it come about and what has become of it?

Once upon a time there was a test …

A precursor to the TMT was developed by John E. Partington in 1938 and named “Partington’s Pathways Test”. While it was initially intended as a test to measure motor speed, after the first studies it was found that the results correlated moderately to highly with intelligence tests. Accordingly, Partington concluded that the test captured general mental abilities. The TMT received its present name when it was incorporated in a slightly modified form into the U.S. Army’s “Army Individual Test Battery” in 1944. In the course of this, it was used as a screening for general intellectual level, but also to examine soldiers with brain injuries. The utility of the TMT for screening individuals with brain injuries was also recognized by Ralph M. Reitan in the 1950s and was therefore added to the Halstead-Reitan Neuropsychological Test Battery (HRNTB). The form contained therein, and the procedure used by Reitan to administer and score the TMT is the most widely used version (hereafter referred to as the original version).

One test, many cognitive functions

A major reason for the popularity of the TMT is its simple and fast administration. The TMT consists of two parts. In part A, the subject sees circles with numbers from 1 to 25 that appear to be randomly distributed. The task is to connect the circles with the numbers as quickly as possible in ascending order (1-2-3-…). Part B then shows circles with numbers or letters and the test takers are to connect them alternately in ascending or alphabetical order (1-A-2-B-…). Despite the seemingly simple tasks, a multitude of cognitive processes are involved in successful and rapid completion. Accordingly, the test is associated with various cognitive functions: visual (divided) attention, processing speed, motor speed, visuoperceptual functions, cognitive flexibility, inhibition, working memory, general mental ability, or fluid intelligence. Studies on its construct validity indicate that total time for completion in Part A is considered a measure of visual attention and processing speed, whereas total time for completion in Part B is even more strongly related to executive functions and general, fluid intelligence (Larrabee & Curtiss, 1994; Salthouse, 2011).

One test, many variants

Although the original version of Reitan is the most widely used, several variants of the test have been developed over time.

Cross-cultural or culture-specific variants: For the Asian region, variants without letters in Part B were developed. These were then replaced by characters for zodiac signs, for example (Wang et al., 2018). Another way to get by without letters is to use only numbers but embed them in different geometric shapes (e.g., squares and circles; Lu & Bigler, 2000; Zhao et al., 2013) or background colors. In the Color Trails Test (CTT; D’Elia et al., 1996), in Part A, all circles with odd numbers have a pink background, while all even numbers have a yellow background. Part B contains duplicates of each number from 1 to 15 embedded in pink and yellow circles. The subject is asked to quickly connect the circles in ascending order, but alternating between pink and yellow (pink 1-yellow 2-pink 3-…). The CTT was developed with the goal of providing a fair assessment of visuomotor processing speed across cultures and from infancy. However, there are mixed results on the equivalence of the CTT and the original version of the TMT (Dugbartey et al., 2000; Lee & Chan, 2010).

Extended variants: Some tests aim to capture in a more sophisticated way the different cognitive subdimensions relevant for completing the TMT tasks. The TMT variant in the Delis-Kaplan Executive Function System (D-KEFS; Delis, Kaplan & Kramer, 2001) contains Parts A and B of the original TMT, but has additional tasks on visual search, processing speed (connecting letters; A-B-C-…), and motor speed (connecting empty circles using a dashed line). In addition, the test is administered in A3 rather than A4 format, which means that there are higher requirements for visual exploration. The Comprehensive Trail Making Test (CTMT; Reynolds, 2019) is also an extended form of the original TMT. In it, three additional number linking tasks are included. Two tasks with distractors (empty circles or circles filled with irrelevant line drawings) and one task that requires switching between numbers in Arabic and spelled out form (1-Two-3-Four- …).

Practice item of the Shape Trail Test (Zhao et al., 2013)

One test, many applications

The TMT has been shown to be useful in many applications and for the examination of diverse groups of individuals. Evidence on its criterion validity exists for diverse neurological disorders such as traumatic brain injury, stroke, dementia in Alzheimer’s disease and other etiologies, Huntington’s disease, Parkinson’s disease, or chronic toxic encephalopathy. Individuals with psychiatric disorders such as schizophrenia, depression, obsessive-compulsive disorder, post-traumatic stress disorder, or alcoholism also show abnormalities in TMT. It is interesting to note that cognitive impairments can be detected and severity differentiated with TMT even in preliminary stages or in mild forms of illness. For example, individuals with the huntingtin gene already show lower performance in TMT than healthy individuals in the prodromal phase, i.e., up to 15 years before the onset of motor symptoms (O’Rourke et al., 2011). Individuals with mild strokes or transitory ischemic attacks (TIA) also show deficits in both parts of the TMT, which are still detectable after four years (Nicolas et al., 2021). Individuals with schizophrenia who are treated as outpatients also have deficits, but these are smaller than those of individuals treated as inpatients (Laere et al., 2018).

Performance on the TMT also has predictive validity. For example, the extent of independence in instrumental activities of daily living such as cooking, shopping, or medication management can be predicted based on the Part B score (Cahn-Weiner et al., 2002). In individuals with neurological conditions, the incidence of falls during hospitalization (Mateen et al., 2018) has also been shown to be related to TMT performance. In addition, another frequently studied external criterion that is highly relevant to everyday life is fitness to drive. In a variety of studies with different groups of individuals, TMT has been found to be useful for predicting practical driving behavior (e.g., Grace et al., 2005; Devos et al., 2011; Asimakopulos et al., 2012). Accordingly, it is also often used in screenings for driving ability, such as in the DRIVESC2 test set.

Practice items Part A and B of the orginal version of the Trail Making Test (TMT)

Practice item of the Walking Trail-Making Test
(Persad et al., 2008)

Variants without hand motor skills

In addition, variants have been developed that do not require hand motor skills. In the Walking Trail-Making Test (Alexander et al., 2005; Persad et al., 2008; WTMT), mats marked with circled numbers and/or letters are placed on the floor. The subject must then step on the corresponding numbers/letters. Correlations with the original version of the TMT are low for Part A, but moderate to high for Part B. A less elaborate version, which also does not require hand motor skills, is simply administered orally (Oral Trail Making Test, OTMT; Ricker & Axelrod, 1994). The OTMT was developed specifically for individuals with severe motor impairments, visual impairments, or illiteracy. Here, the person simply counts aloud from 1 to 25 (Part A) or alternates between numbers and letters until the number 13 is reached (Part B). According to the authors, the main purpose is to be able to estimate performance in written TMT. At least for Part B, this has been consistently demonstrated in several studies (Axelrod & Lamberty, 2006).

Digital variants

In addition to these analog variants, there are now also digital TMTs. One of these is the Langensteinbacher version (TMT-L; Rodewald et al., 2020) in the Vienna Test System (VTS). The TMT-L is not simply a digital copy of Parts A and B of the original version of the TMT, but has been adapted to address the shortcomings of the original version. For example, the path of Part B has been shown to be longer than Part A and the individual points are also further apart. Differences in total time for completion between Parts A and B of the original version are therefore not only due to the additional demand on cognitive flexibility, but also simply to more visual exploration (Gaudino et al., 1995, Woodruff et al., 1995). In the digital TMT-L version, therefore, the layout of the two parts was adjusted so that the paths as well as the number of distractors are the same. In addition, almost the same number of items run in the left and right directions, respectively, and adjacent items lie in the area of foveal vision, thus avoiding long saccades. The advantage of the digital version is a completely standardized and automatic time measurement, which was not always consistent in the past when using the original version (exact start of time measurement as well as handling of errors). The TMT-L is available in three parallel versions and can be used with computer mouse as well as on touch screen. Despite these differences, research shows that the digital TMT-L and the original paper & pencil version produce similar results.

The history of the TMT shows that even for such established tests, further development is still possible, be it on a psychometric or technological level. The convincing features of the test remain: very short and simple administration, but still high sensitivity for subtle cognitive impairments in various disorders. The TMT-L is also a popular neuropsychological test in the Vienna Test System, which has also been integrated into several test sets for measuring cognitive functions in dementia (CFD), ADHD (CFADHD), schizophrenia (CFSD) or simple baseline testing (COGBAT).

References:

Alexander, N. B., Ashton-Miller, J. A., Giordani, B., Guire, K., & Schultz, A. B. (2005). Age differences in timed accurate stepping with increasing cognitive and visual demand: A Walking Trail Making Test. Journal of Gerontology60(12), 1558–1562. https://doi.org/10.1093/gerona/60.12.1558

Arbuthnott, K., & Frank, J. (2000). Trail Making Test, Part B as a measure of executive control: Validation using a set-switching paradigm. Journal of Clinical and Experimental Neuropsychology22(4), 518–528. https://doi.org/10.1076/1380-3395(200008)22:4;1-0;FT518

Asimakopulos, J., Boychuck, Z., Sondergaard, D., Poulin, V., Ménard, I., & Korner-Bitensky, N. (2012). Assessing executive function in relation to fitness to drive: A review of tools and their ability to predict safe driving. Australian Occupational Therapy Journal59(6), 402–427. https://doi.org/10.1111/j.1440-1630.2011.00963.x

Axelrod, B. N., & Lamberty, G. J. (2006). The Oral Trail Making Test. In A. M. Poreh (Ed.), The quantified process approach to neuropsychological assessment (pp. 45–51). Psychology Press.

Cahn-Weiner, D. A., Boyle, P. A., & Malloy, P. A. (2002). Tests of executive function predict instrumental activities of daily living in community-dwelling older individuals. Applied Neuropsychology 9(3), 187–191. https://doi.org/10.1207/S15324826AN0903_8

D’Elia, L., Satz, P., Uchiyama, C. L., & White, T. (1996). Color Trails Test. PAR.

Delis, D. C., Kaplan, E., & Kramer, J. H. (2001). Delis-Kaplan Executive Function System. Pearson.

Devos, H., Akinwuntan, A. E., Nieuwboer, A., Truijen, S., Tant, M., & de Weerdt, W. (2011). Screening for fitness to drive after stroke: A systematic review and meta-analysis. Neurology76(8), 747–756. https://doi.org/10.1212/WNL.0b013e31820d6300

Dugbartey, A. T., Townes, B. D., & Mahurin, R. K. (2000). Equivalence of the Color Trails Test and Trail Making Test in nonnative English-speakers. Archives of Clinical Neuropsychology15(5), 425–431. https://doi.org/10.1016/S0887-6177(99)00034-7

Gaudino, E. A., Geisler, M. W., & Squires, N. K. (1995). Construct validity in the trail making test: What makes part B harder? Journal of Clinical and Experimental Neuropsychology17(4), 529–535. https://doi.org/10.1080/01688639508405143

Grace, J., Amick, M. M., D’Abreu, A., Festa, E. K., Heindel, W. C., & Ott, B. R. (2005). Neuropsychological deficits associated with driving performance in Parkinson’s and Alzheimer’s disease. Journal of the International Neuropsychological Society11(06). https://doi.org/10.1017/S1355617705050848

Kortte, K. B., Horner, M. D., & Windham, W. K. (2002). The Trail Making Test, Part B: Cognitive Flexibility or Ability to Maintain Set? Applied Neuropsychology9(2), 106–109. https://doi.org/10.1207/S15324826AN0902_5

Laere, E., Tee, S. F., & Tang, P. Y. (2018). Assessment of cognition in schizophrenia using trail making test: A meta-analysis. Psychiatry Investigation15(10), 945–955. https://doi.org/10.30773/pi.2018.07.22

Larrabee, G. J., & Curtiss, G. (1995). Construct validity of various verbal and visual memory tests. Journal of Clinical and Experimental Neuropsychology17(4), 536–547. https://doi.org/10.1080/01688639508405144

Lee, T. M. C., & Chan, C. C. H. (2000). Comparison of the Trail Making and Color Trails Tests in a Chinese context: A preliminary report. Perceptual and Motor Skills90(1), 187–190. https://doi.org/10.2466/pms.2000.90.1.187

Lu, L., & Bigler, E. D. (2000). Performance on original and a Chinese version of Trail Making Test Part B: A normative bilingual sample. Applied Neuropsychology7(4), 243–246. https://doi.org/10.1207/S15324826AN0704_6

Mateen, B. A., Bussas, M., Doogan, C., Waller, D., Saverino, A., Király, F. J., & Playford, E. D. (2018). The Trail Making test: a study of its ability to predict falls in the acute neurological in-patient population. Clinical Rehabilitation32(10), 1396–1405. https://doi.org/10.1177/0269215518771127

Nicolas, K., Goodin, P., Visser, M. M., Michie, P. T., Bivard, A., Levi, C., Parsons, M. W., & Karayanidis, F. (2021). Altered functional connectivity and cognition persists 4 years after a transient ischemic attack or minor stroke. Frontiers in Neurology12https://doi.org/10.3389/fneur.2021.612177

Oosterman, J. M., Vogels, R. L. C., van Harten, B., Gouw, A. A., Poggesi, A., Scheltens, P., Kessels, R. P. C., & Scherder, E. J. A. (2010). Assessing mental flexibility: Neuroanatomical and neuropsychological correlates of the trail making test in elderly people. Clinical Neuropsychologist24(2), 203–219. https://doi.org/10.1080/13854040903482848

O’Rourke, J. J. F., Beglinger, L. J., Smith, M. M., Mills, J., Moser, D. J., Rowe, K. C., Langbehn, D. R., Duff, K., Stout, J. C., Harrington, D. L., Carlozzi, N., & Paulsen, J. S. (2011). The trail making test in prodromal huntington disease: Contributions of disease progression to test performance. Journal of Clinical and Experimental Neuropsychology33(5), 567–579. https://doi.org/10.1080/13803395.2010.541228

Persad, C. C., Jones, J. L., Ashton-Miller, J. A., Alexander, N. B., & Giordani, B. (2008). Executive function and gait in older adults with cognitive impairment. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences63(12), 1350–1355. https://doi.org/10.1093/gerona/63.12.1350

Salthouse, T. A. (2011). What cognitive abilities are involved in trail-making performance? Intelligence39(4), 222–232. https://doi.org/10.1016/j.intell.2011.03.001

Reynolds, C. R. (2019). Comprehensive Trail Making Test-Second Edition (CTMT-2). PAR.

Ricker, J. H., & Axelrod, B. N. (1994). Analysis of an oral paradigm for the Trail Making Test. Assessment1(1), 47–51. https://doi.org/10.1177/1073191194001001007

Rodewald, K., Weisbrod, M., & Aschenbrenner, S. (2020). Trail Making Test Langensteinbacher Version. SCHUHFRIED.

Wang, R. Y., Zhou, J. H., Huang, Y. C., & Yang, Y. R. (2018). Reliability of the Chinese Version of the Trail Making Test and Stroop Color and Word Test among Older Adults. International Journal of Gerontology12(4), 336–339. https://doi.org/10.1016/j.ijge.2018.06.003

Woodruff, G. R., Mendoza, J. E., Dickson, A. L., Blanchard, E., & Christenberry, L. B. (1995). The effects of configural differences on the trail making test. Archives of Clinical Neuropsychology10(4), 408–408. https://doi.org/10.1093/arclin/10.4.408

Zhao, Q., Guo, Q., Li, F., Zhou, Y., Wang, B., & Hong, Z. (2013). The Shape Trail Test: Application of a new variant of the Trail Making Test. PLoS ONE8(2), e57333. https://doi.org/10.1371/journal.pone.0057333

Newsletter

Newsletter

Stay up to date on the latest tests, practical tips and tricks on subjects related to digital assessment or interesting offers for continuing professional development.