OBJECTIVE

The Trail Making Test (TMT) is a psychomotor task that is used routinely in clinical evaluations and consists of two parts: TMT-A and TMT-B. The purpose of the present study was to examine the relations of two derived-TMT indices with age, education, and gender and to establish normative data.

BACKGROUND

Prior research has suggested that these derived indices provide purer measures of the executive skills required to complete TMT-B. It has also been suggested that these scores can be effectively used to detect cognitive impairment, and that they are relatively free from the impact of age.

METHOD

Difference scores (B minus sign A) and ratio scores (B:A) were computed from the TMT performance of 285 adults who have no history of substance abuse or psychiatric or neurologic disorder and who are not currently being treated with any type of psychotropic medication.

RESULTS

Subjects ranged in age from 18 to 90 years and in education level from 6 to 20 years. For the entire sample, (B minus sign A) scores ranged from 6 to 248 seconds (M, 62.15; SD, 49.44), and (B:A) scores ranged from 1.23 to 9.54 (M, 2.81; SD, 1.13). Significant correlations were noted with age and (B minus sign A) and (B:A) scores, accounting for 28% and 6% of the variance, respectively. Subjects were assigned to one of eight age groups, which were formed by decade. Univariate analyses of variance were significant for (B minus sign A) (p < 0.0001) and (B:A) (p < 0.0004) scores. Post hoc analyses indicated that the two oldest age groups had significantly larger (B minus sign A) scores than the younger age groups. However, post hoc analyses with (B:A) scores indicated that only the 70-79-year-old age group significantly differed from the three youngest age groups. No other age groups significantly differed from another on this measure.

CONCLUSIONS

Normative data are presented for clinical use, and recommendations are made for future research. In addition, the current results suggest that normal aging has a significant impact on cognitive flexibility and indicate that previous suggestions for impairment cut-off scores based on the (B:A) ratio may result in erroneous classification of healthy older adults.