Abstract
An attempt was made to explain the temporal accuracy with which a series of responses could be matched to a series of signals, by reference to the continuously changing temporal relationships between these events. Although approximately half of the responses were made too soon and half too late, this distribution was not fortuitous. The temporal position of a response in relation to the signal which initiated it was influenced both by the position of the previous response, and by the position of the signal for the next response. Not only did these events determine whether a response would be made early or late, but to a large extent they influenced the size of this error. Thus, while the character of the temporal structure of events around a signal determined the relative accuracy of the response, the mean signal speed in relation to the mean response rate determined the frequency with which particular temporal structures occurred. The underlying factors that actually determined the character of the response were shown to be masked when the effects on response of the signal speed variable were presented as mean accuracy values.
Keywords

This publication has 1 reference indexed in Scilit: