The Impact of Language on the Equivalence of Trail Making Tests: Findings from Three Pediatric Cohorts with Different Language Dominance

Abstract

The effect of different language backgrounds on performance and the functional equivalence of trail making tests were examined. The children's version of the Trail Making Test A and B (TMT) and the Children's Color Trails Test 1 and 2 (CCTT) were employed with right-handed (n = 79) participants. Children were classified into three groups according to language proficiency: Chinese dominant (CDL), Chinese-English bilinguals (CEB), and English dominant (EDL). In general, the CDL group exhibited the best performance on Children's Color Trails 1 and 2 and Trail Making Test A and B. In examining the functional equivalence of TMT and CCTT, both tests were influenced by language background and intelligence but not gender. The results suggest that language backgrounds do exert a modest effect on trail making tests and that diverging executive demands on CCTT 2 versus TMT B may impact their functional equivalence with different groups of participants. Therefore, caution should be exercised when comparing or replacing CCTT with TMT in children from different and diverging language backgrounds.

Children's Color Trails Test 1 & 2: Test–Retest Reliability and Factorial Validity

“…Unlike existing, traditional trail-making tests that have been shown to be affected by learning disabilities (Rourke & Finlayson, 1975), language, and to a lesser extent, culture (Leon-Carreon, 1989;Mok, Tsang, Lee, & Llorente, 2008;Rosin & Levett, 1989a, 1989b as a result of their dependence on letters of the alphabet for a categorical shift, the CCTT replaces the use of letters with colors as categorical shift, as noted above (Llorente et al, 2003). Such alteration, in addition to other changes including the addition of non-verbal instructions, has been shown to reduce the weaknesses found in traditional trail-making tests in some instances (Lee, Cheung, Chan, & Chan, 2000;Llorente et al, 2003;Maj et al, 1993Maj et al, , 1994.…”

“…Such alteration, in addition to other changes including the addition of non-verbal instructions, has been shown to reduce the weaknesses found in traditional trail-making tests in some instances (Lee, Cheung, Chan, & Chan, 2000;Llorente et al, 2003;Maj et al, 1993Maj et al, , 1994. However, other complex factors have emerged on CCTT performance not found in other trail-making tests such as a possible Stroop effect (Mok et al, 2008).…”

“…Although the CCTT is purported to measure several neuropsychological domains, including attention and cognitive flexibility, little is known about its factorial validity when employed with clinical samples. Therefore, despite the fact that surveys conducted over the last three decades in several continents indicate that trail-making tests continue to be one of the most widely used measures of neuropsychological assessment (Bassa & Schlebusch, 1984;Butler, Retzlaff, & Vanderploeg, 1991;Rabin, Barr, & Burton, 2005;Tsoi & Sundberg, 1989), and the increased worldwide attention and usage of the CCTT (Baron, 2004;Mok et al, 2008;Strauss, Sherman, & Spreen, 2006), few studies have investigated its psychometric properties with clinical, pediatric populations (cf. Strauss et al, 2006).…”

Trail Making Test: Normative data for the Latin American Spanish-speaking pediatric population

“…Some studies have found that boys perform better than girls in both versions (Lee et al, 2002), while other studies have found that overall girls perform better in the TMT-B compared to boys (Barr, 2003). However, other studies have found no difference between the sexes (Anderson, Lajoie, & Bell, 1997;Beltrán Dulcey & Solís-Uribe, 2012;León-Carrión, 1989;Mok et al, 2008). In the present study, sex was associated with total time in the TMT-A for Chile, Cuba, Mexico, and Peru so that boys needed less time to complete the test than girls.…”

“…Regarding gender, there are studies that have shown significant differences (Lee et al, 2002), while others have not found any influence (Beltrán Dulcey & Solís-Uribe, 2012;León-Carrión, 1989;Mok, Tsang, Lee, & Llorente, 2008). Finally, although the influence of parental education on children's performance in TMT has not been studied, Van der Elst, Hurks, Wassenberg, Meijs, and Jolles (2011) have shown that high parental schooling correlates with better child performance on verbal fluency tests, so it would be advisable to include this variable in future research concerning the neuropsychological performance of children.…”