DISSOCIATION BETWEEN DELAYED ALTERNATION AND MEMORY AFTER PEDIATRIC HEAD-INJURY - RELATIONSHIP TO MRI FINDINGS

This study investigated the usefulness of a delayed alternation task i n characterizing the cognitive sequelae of closed head injury in child ren and adolescents. Verbal learning and memory (California Verbal Lea rning Test) were also studied for comparison. Sixty-two closed head in jury patients (mean age, 9.6 years), who were studied after an average postinjury interval of 20 months, were divided according to both thei r lowest postresuscitation Glasgow Coma Scale score (3 to 8 versus 9 t o 15) and age range (5 to 7 years versus 8 to 16 years) at the time of testing. Magnetic resonance imaging was performed to evaluate the rel ationship of focal brain lesions to cognitive and memory performance. Fifty-six neurologically normal children (mean age, 9.9 years) were te sted on the same measures. The results disclosed no relationship betwe en delayed alternation performance and severity of injury. In contrast , verbal memory was impaired in the severely-injured patients, relativ e to both controls and less severely-injured patients. Frontal lobe (b ut not extrafrontal) lesion size incremented the Glasgow Coma Scale sc ore in predicting verbal memory, but there was no relationship between focal brain lesions and delayed alternation performance. In contrast to the tendency for more efficient delayed alternation performance in the 5- to 7-year-old subjects than in the 8- to 16-year-old subjects, verbal memory significantly improved with age in the closed head injur y and control groups. Notwithstanding our essentially negative finding s for delayed alternation, it is possible that this task may be useful for assessing frontal lobe injury in younger children or infants. Our results indicate limitations in extrapolating from nonhuman primate s tudies of delayed alternation to frontal lobe function in children and adolescents.