Normative data and screening power of a shortened version of the Token Test

Abstract

A 36-item-version of the Token Test is described and normative data obtained from its administration to 215 normal subjects are given. Years of schooling (but not age) were found to significantly affect the performance. The scores were corrected for this factor and the lower limit of the 90% tolerance interval around the mean of the adjusted scores was determined: it was found to correspond to 29 and left below it exactly 5% of the normal sample. The test was given to 200 aphasic patients. Fourteen (7%) were found to have an adjusted score of 29 or more, namely would have been classified as non-aphasic. This is a percentage remarkably smaller than that (40%) obtained with a 10 sentence comprehension test, which supports previous studies pointing to the sensitivity of the Token Test to the presence of oral language disorders. On the basis of the aphasic patients' performance, cutting scores allowing evaluation of the severity of the comprehension deficit are provided. The 36-item-version of the test appears to be an useful and convenient device to diagnose aphasic impairment of language comprehension.