An experiment in time perception: separating rate of presentation from type of estimation.

An experiment was conducted to distinguish between studies which show main effects versus interaction effects regarding estimations of time as a function of the rate (fast or slow) of digit presentation and the type (immediate or delayed) of time estimation. Undergraduate volunteers (36 men and 103 women) gave retrospective estimates of time following a short-term memory filler task consisting of memorizing a series of random digits (1-5) which varied in length (25 or 50 digits). Rate was manipulated by altering the interstimulus interval between digits (100 msec. or 1,300 msec.) or by varying the length of the list. A multivariate analysis of variance showed significant main effects for type of time estimation (delayed estimates were longer than immediate), length of list (estimates for 50 digits were longer than for 25), and interstimulus interval (estimates for 1,300 msec. were longer than for 100 msec.) yet, contrary to Pedri and Hesketh in 1993, there were no significant interactions among the variables. Each of the independent variables was associated with effects on time estimation and short-term memory which were the same at all levels of the other variables. Task variables were discussed as a potential source of variance among past studies.