Abstract
A fully automatized procedure permitting computerized evaluation of ERA from the vertex is described. It is based on rank correlation. The correlation becomes a more reliable procedure when performed on the derivative of the measurements and using a biological template. The procedure has been tested on normal adults and 1- to 4- year-old children. 10% errors are found in 159 tests, evenly balanced between false positive and false negative errors. This compares very favourably to a subjective judgement of response, where specially in children there is a clear tendency to judge false positive responses. The method and the results are discussed