A Quantitative Model for Retrospective Subjective Duration

Abstract

In retrospective, as opposed to prospective, temporal tasks, the subject is unaware that a duration judgment will be required. In previous studies, the durations to be judged retrospectively were filled with some cognitive task, like reading. To reduce cognitive effects and keep as close to the internal clock as possible, in the present study the durations (ten, ranging from 1.3 to 20 sec) were filled with noise and the method of reproduction was used. Assuming the psychophysical power law, the single retrospective reproductions could be well predicted from (a) the Parallel-Clock Model (originally developed for prospective reproduction experiments, H. Eisler, 1975), together with (b) individual parameter values obtained from prospective data and thus recovered in the retrospective, when (c) the standard durations were transformed by a common additive term. This term was interpreted as representing the influence on where in time the subjects positioned the start of the duration, depending on familiarization with the situation, and boredom.