Accuracy, Reliability, and Refractoriness in a Coincidence-Anticipation Task

Abstract The purpose of this study was to obtain further evidence concerning human performance in a particular type of anticipation task commonly termed the transit reaction. It was found that: (a) Performance can be exceptionally accurate; (b) the absolute error for performance with knowledge of results was significantly smaller than the error for performance without knowledge of results; (c) a small systematic constant error was found to persist for performance with and without knowledge of results; (d) both absolute and constant errors increased with the introduction of catch trials; (e) the correlation coefficients (.86—.54) suggested that the measures were at least modestly reliable; and (f) the estimated refractory period was 149 msec.