Response time measurements in the iOS app-based Token Test, the Brief Token Test in the elderly

A pen and paper Token Test utilizing binary scoring of accuracy-inaccuracy in responses evaluates linguistic functions including verbal comprehension. However, the test might not be sensitive enough to detect the aging effect on performance in normal older adults unless the test utilizes a multidimensional scoring system. We hypothesized that the measurement of response time (RT) instead of accuracy would be a feasible way to measure deterioration of verbal comprehension as a function of age in this linguistically vulnerable population.

Data were acquired from 174 community-dwelling normal older adults (147 females and 27 males, mean age 75.9 ± 6.13 years, range 65-94 years). In order to automatically quantify accuracies and RT, we developed an iOS app-based Token Test, comprising of four subtests (24 items, 6 items in each subtest), and administered it to each participant.

RT of the second, third and fourth subtests (P<0.001, P<0.01, P<0.05, respectively) and the average RT across all four subtests (P<0.01) had significant partial correlation coefficients with age. The results of the multiple linear regression analyses showed that the models for RT and average RT analyses of the three subtests (2, 3 and 4) included the age variable.

RT measurement of the iOS app-based Token Test could be highly useful and convenient to better detect abnormalities in linguistic efficiency. RT measurement in the Token Test could be used as a screening tool for linguistic slowing in the elderly population.