神经导航.pdf

JOURNAL OF NEUROTRAUMA Volume 18, Number 11, 2001 Mary Ann Liebert, Inc. Invited Paper Neurobehavioral Assessment of Outcome following Traumatic Brain Injury in Rats: An Evaluation of Selected Measures ROBERT J. HAMM ABSTRACT Neurobehavioral assessment of outcome has played an integral part in traumatic brain injury (TBI) research. Given the fundamental role of neurobehavioral measurement, it is critical that the tasks used are of the highest psychometric quality. The purpose of this paper is to evaluate several, com- monly used neurobehavioral measures along the dimensions of reliability, sensitivity, and validity. Using both the midline and lateral fluid-percussion injury models, nine neurobehavioral measures were evaluated that assessed three different neurobehavioral constructs. Reflex suppression was measured by the duration of the suppression of the pinna, corneal, and righting reflexes. Vestibu- lomotor function was assessed with the beam-balance, beam-walking, and rotorod tasks. Cognitive function was evaluated by three measures of Morris water maze performance (goal latency, path length, cumulative distance). The evaluation of the reliability of the nine neurobehavioral measures found that all had acceptably high reliability coefficients (0.79 or higher). The analysis of each mea- sure’s sensitivity to injury found that all measures were capable of detecting injury-induced im- pairments. However, there were some substantial differences in the sensitivity of the measures of vestibulomotor and maze performance: the rotorod was the most sensitive vestibulomotor measure and goal latency and path length were equally sensitive measures of maze performance. In the as- sessment of validity, the results of a factor analysis supported the convergent and discriminat