How the listener integrates the components of speaking rate

The rate of speaking in words per minute is a function of three independent variables, namely, articulation rate and the number and durations of pauses. The present study varies each of these components separately in a factorial design in order to determine how the listener combines them into a global impression of speech rate. A model was obtained by direct scaling and linear regression that accounts reasonably well for the estimates of apparent rate obtained in this and several other studies. It provides for a trading relation between articulation rate and pause rate in which the former is much more influential in determining the listener's perception of speech rate.