The purpose of this study was to explore how testing procedure influen
ces performance after traumatic brain injury, and how this influence v
aries as a function of the subject's information-processing abilities.
Twelve brain-injured subjects completed three versions of the Listeni
ng/Grammar subtest of the Test of Adolescent Language. In the first co
ndition, sentences were presented at 2-s intervals, in the second at 4
-s intervals, and in the third at variable intervals controlled by the
subject. Scores in each condition were correlated with measures of wo
rking memory storage and processing ability. Performance in the 2-s co
ndition was significantly influenced by working memory processing abil
ity, whereas performance in the 4-s condition was significantly influe
nced by working memory storage ability. Both aspects of working memory
contributed to performance in the variable condition, although only p
rocessing was significant. As a group, subjects with better working me
mory processing ability preferred the variable interval condition, whi
ch allowed greater flexibility, while subjects with poor processing ab
ility found it more difficult. The results are discussed in terms of c
linical assessment and intervention.